Compare commits
7 Commits
522a8f8d46
..
clean
| Author | SHA1 | Date | |
|---|---|---|---|
 jalf
|
a5dc093a15 | ||
 Johnny Fernandes
|
1ed2b7a7a0 | ||
|
Johnny Fernandes
|
afd26f47d2 | ||
|
DiogoCosta18
|
3bff7eefb0 | ||
|
DiogoCosta18
|
f46320f81e | ||
|
DiogoCosta18
|
2062a91985 | ||
|
DiogoCosta18
|
9ae334410d |
@@ -67,3 +67,6 @@ generator/outputs/samples/*
|
|||||||
.venv/
|
.venv/
|
||||||
.ipynb_checkpoints/
|
.ipynb_checkpoints/
|
||||||
__pycache__/
|
__pycache__/
|
||||||
|
|
||||||
|
#Presentation
|
||||||
|
presentation_inputs.zip
|
||||||
@@ -1,125 +1,264 @@
|
|||||||
# DRL_PROJ — DeepFake Detection
|
# Deep learning face project
|
||||||
|
|
||||||
Deep learning project for binary deepfake detection on the DeepFakeFace dataset.
|
This repository contains a two-part deep learning project on the
|
||||||
|
DeepFakeFace (DFF) dataset:
|
||||||
|
|
||||||
## Project structure
|
1. **Classifier:** detect whether a face image is real or fake.
|
||||||
|
2. **Generator:** train generative models that produce new fake face images.
|
||||||
|
|
||||||
```
|
The project is written as an experimental report. The notebooks are the main
|
||||||
|
deliverable: they show the pipeline, the intermediate failures, the ablations,
|
||||||
|
the decisions, and the final models. Read them in order.
|
||||||
|
|
||||||
|
## Project story
|
||||||
|
|
||||||
|
The work follows the same principle in both parts: start with a simple
|
||||||
|
baseline, inspect what fails, change one important factor at a time, and keep
|
||||||
|
the evidence tied to saved logs and saved artifacts.
|
||||||
|
|
||||||
|
For the **classifier**, the story moves from dataset understanding to
|
||||||
|
preprocessing, baseline models, controlled ablations, Grad-CAM inspection,
|
||||||
|
stronger model families, and data scaling. The final practical classifier is a
|
||||||
|
ResNet50-style pipeline using face crops, 224×224 inputs, ImageNet/default
|
||||||
|
normalization, and no stochastic augmentation at validation/test time.
|
||||||
|
|
||||||
|
For the **generator**, the story starts with raw baseline failures, then locks
|
||||||
|
the data pipeline before comparing three parallel model-family branches:
|
||||||
|
GAN, VAE, and DDPM. The final comparison keeps quality versus speed central:
|
||||||
|
DDPM gives the best saved FID and visual quality, GAN is the best
|
||||||
|
quality-speed compromise, and VAE is the fastest but smoothest option.
|
||||||
|
|
||||||
|
## How to read the project
|
||||||
|
|
||||||
|
Start with the classifier notebooks, then read the generator notebooks. The
|
||||||
|
generator has one linear setup stage followed by three parallel branches:
|
||||||
|
GAN, VAE, and DDPM. Those branches are numbered in reading order, but they are
|
||||||
|
conceptually parallel experiments after the pipeline is selected.
|
||||||
|
|
||||||
|
### Classifier notebooks
|
||||||
|
|
||||||
|
Read these first:
|
||||||
|
|
||||||
|
1. `classifier/notebooks/01_eda.ipynb`
|
||||||
|
Dataset composition, real/fake source mapping, image statistics, and
|
||||||
|
shortcut risks.
|
||||||
|
2. `classifier/notebooks/02_preprocessing.ipynb`
|
||||||
|
Deterministic preprocessing, train-only augmentation, face crops, and
|
||||||
|
normalization.
|
||||||
|
3. `classifier/notebooks/03_phase1_analysis.ipynb`
|
||||||
|
SimpleCNN and ResNet18 controlled baselines.
|
||||||
|
4. `classifier/notebooks/04_phase2_analysis.ipynb`
|
||||||
|
Resolution, normalization, source holdouts, facecrop, and augmentation
|
||||||
|
ablations.
|
||||||
|
5. `classifier/notebooks/05_gradcam_analysis.ipynb`
|
||||||
|
Qualitative localization analysis across the classifier pipeline.
|
||||||
|
6. `classifier/notebooks/06_phase3_model_family_analysis.ipynb`
|
||||||
|
Stronger pretrained model families and the ResNet50 practical choice.
|
||||||
|
7. `classifier/notebooks/07_phase4_data_scaling_analysis.ipynb`
|
||||||
|
Data scaling for strong backbones and the final classifier decision.
|
||||||
|
|
||||||
|
### Generator notebooks
|
||||||
|
|
||||||
|
Read these after the classifier:
|
||||||
|
|
||||||
|
1. `generator/notebooks/01_baseline_sanity_check.ipynb`
|
||||||
|
Raw baseline failures and why the data pipeline must be fixed first.
|
||||||
|
2. `generator/notebooks/02_pipeline_selection.ipynb`
|
||||||
|
Controlled pipeline ablations: resolution, alignment, augmentation, and
|
||||||
|
raw/aligned mixing.
|
||||||
|
3. `generator/notebooks/03_gan_stability_progression.ipynb`
|
||||||
|
GAN branch: DCGAN → WGAN-GP → spectral normalization + GroupNorm +
|
||||||
|
self-attention → 128×128 check.
|
||||||
|
4. `generator/notebooks/04_vae_loss_progression.ipynb`
|
||||||
|
VAE branch: MSE + KL → perceptual loss → PatchGAN adversarial loss.
|
||||||
|
5. `generator/notebooks/05_ddpm_recipe_progression.ipynb`
|
||||||
|
DDPM branch: linear schedule → cosine schedule → v-prediction → wider
|
||||||
|
backbone.
|
||||||
|
6. `generator/notebooks/06_final_family_comparison.ipynb`
|
||||||
|
Final comparison of the selected GAN, VAE, and DDPM recipes under saved
|
||||||
|
Phase 5 conditions.
|
||||||
|
7. `generator/notebooks/07_final_sample_showcase.ipynb`
|
||||||
|
Curated final sample examples from saved outputs. This is qualitative
|
||||||
|
showcase material, not a replacement for FID.
|
||||||
|
|
||||||
|
## What the notebooks do
|
||||||
|
|
||||||
|
The notebooks are analysis/report chapters. They load existing configs, logs,
|
||||||
|
figures, saved sample grids, checkpoints, and prediction summaries. They are
|
||||||
|
not intended to launch new training runs.
|
||||||
|
|
||||||
|
When a notebook shows a plot or image grid, the surrounding markdown explains:
|
||||||
|
|
||||||
|
- what the artifact shows;
|
||||||
|
- why it is needed;
|
||||||
|
- how it supports the phase decision;
|
||||||
|
- what limitation remains.
|
||||||
|
|
||||||
|
This is important because the project is evaluated not only by final
|
||||||
|
performance, but by the documented evolution of the solution.
|
||||||
|
|
||||||
|
## Repository layout
|
||||||
|
|
||||||
|
```text
|
||||||
DRL_PROJ/
|
DRL_PROJ/
|
||||||
classifier/ ← discriminative model (real vs. fake classifier)
|
classifier/
|
||||||
src/ ← model definitions, training, evaluation, preprocessing
|
configs/ experiment configs by phase
|
||||||
configs/ ← experiment configs organised by phase
|
notebooks/ classifier report notebooks
|
||||||
phase1/ ← baseline models (SimpleCNN, ResNet18)
|
outputs/ saved logs, figures, Grad-CAM panels, checkpoints
|
||||||
phase2/ ← architecture sweep (ResNet variants, face-crop)
|
src/ classifier data, models, training, evaluation
|
||||||
phase3/ ← EfficientNet, ViT, frequency-aware training
|
tests/ unit and smoke tests
|
||||||
phase4/ ← ensemble strategies
|
tools/ facecrop, Grad-CAM, inference, reevaluation helpers
|
||||||
tools/ ← analyse.py, ensemble.py, inference.py, facecrop.py
|
|
||||||
notebooks/ ← EDA, preprocessing, evaluation, GradCAM
|
generator/
|
||||||
outputs/ ← models, logs, figures (gitignored except .pt/.json)
|
configs/ generator configs by phase/family
|
||||||
run.py ← main training entry point
|
notebooks/ generator report notebooks and notebook builder
|
||||||
generator/ ← generative model (GAN / VAE / diffusion) — in progress
|
outputs/ saved logs, sample grids, final showcase artifacts
|
||||||
pipeline/ ← Vast.ai ephemeral GPU orchestration
|
src/ generator data, models, training, metrics
|
||||||
data/ ← dataset root (gitignored)
|
tests/ unit and smoke tests
|
||||||
cropped/ ← MTCNN pre-cropped faces (gitignored)
|
tools/ sampling and utility scripts
|
||||||
classifier/ ← bbox crops for the classifier
|
|
||||||
generator/ ← landmark-aligned crops for the generator
|
data/ original DFF dataset root, not committed
|
||||||
|
cropped/ preprocessed face crops, not committed
|
||||||
|
docs/ project statement and supporting documents
|
||||||
|
pipeline/ optional remote/GPU orchestration helpers
|
||||||
```
|
```
|
||||||
|
|
||||||
|
## Rebuilding the generator notebooks
|
||||||
|
|
||||||
|
The generator notebooks are generated from a single source file:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
cd generator/notebooks
|
||||||
|
python _build.py
|
||||||
|
```
|
||||||
|
|
||||||
|
That builder writes the numbered generator notebooks listed above. It uses
|
||||||
|
existing saved logs and artifacts; it does not train models.
|
||||||
|
|
||||||
## Setup
|
## Setup
|
||||||
|
|
||||||
Create a local environment when you want to run the code directly on a machine you control:
|
Create a conda environment and install the project requirements:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 -m venv .venv
|
conda create -n drl python=3.12
|
||||||
source .venv/bin/activate
|
conda activate drl
|
||||||
python -m pip install --upgrade pip setuptools wheel
|
python -m pip install --upgrade pip setuptools wheel
|
||||||
python -m pip install -r requirements.txt
|
python -m pip install -r requirements.txt
|
||||||
```
|
```
|
||||||
|
|
||||||
## Local Training
|
Use **Python 3.12**; some dependencies (for example `facenet-pytorch`) are
|
||||||
|
unreliable on 3.13+.
|
||||||
|
|
||||||
|
The raw dataset should be placed under `data/`. Preprocessed crops are stored
|
||||||
|
under `cropped/`. These folders are intentionally not committed. To download
|
||||||
|
and extract the dataset:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 classifier/run.py classifier/configs/phase2/p2_resnet18_facecrop.json
|
python classifier/tools/fetch_ds.py
|
||||||
python3 classifier/run.py classifier/configs/phase3/p3_efficientnet_b0.json
|
python classifier/tools/fetch_ds.py --data-dir /path/to/DFF
|
||||||
```
|
```
|
||||||
|
|
||||||
## Ephemeral Vast.ai Pipeline
|
Expected layout under the data root: `wiki/<identity>/*.jpg`,
|
||||||
|
`inpainting/...`, `text2img/...`, `insight/...`.
|
||||||
|
|
||||||
The deployment/orchestration path now lives under [`pipeline/`](/run/host/mnt/shared/UP/DRL/DRL_PROJ/pipeline/README.md).
|
## Classifier — training
|
||||||
|
|
||||||
One-time setup:
|
From the repository root:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
cat > pipeline/.env <<'EOF'
|
# CPU (slow but valid)
|
||||||
VAST_API_KEY=<your-api-key>
|
python classifier/run.py classifier/configs/phase4/p4_convnext_tiny_100pct.json
|
||||||
VAST_SSH_PRIVATE_KEY=/home/your-user/.ssh/id_ed25519
|
|
||||||
EOF
|
# GPU when CUDA is available
|
||||||
|
python classifier/run.py classifier/configs/phase4/p4_convnext_tiny_100pct.json --use-gpu
|
||||||
```
|
```
|
||||||
|
|
||||||
End-to-end ephemeral run:
|
Training uses 5-fold stratified group cross-validation. Per-fold checkpoints
|
||||||
|
are saved as `classifier/outputs/models/{run_name}_fold{k}_best.pt` (and
|
||||||
|
`_final.pt`). Override data or output locations with `--data-dir` and
|
||||||
|
`--output-root`.
|
||||||
|
|
||||||
|
**Primary delivery model** (best Phase 4 detector): config
|
||||||
|
`classifier/configs/phase4/p4_convnext_tiny_100pct.json` with per-fold
|
||||||
|
weights `classifier/outputs/models/p4_convnext_tiny_100pct_fold*_best.pt`.
|
||||||
|
|
||||||
|
## Classifier — inference
|
||||||
|
|
||||||
|
Classify a single image as real or fake:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 -m pipeline run classifier/configs/phase2/p2_resnet18_facecrop.json --upload-data
|
python classifier/tools/inference.py image.jpg classifier/configs/phase4/p4_convnext_tiny_100pct.json
|
||||||
```
|
```
|
||||||
|
|
||||||
Interactive offer selection:
|
This loads the config and the matching checkpoint, runs the image through the
|
||||||
|
model, and prints a result like:
|
||||||
|
|
||||||
|
```
|
||||||
|
Image : image.jpg
|
||||||
|
Model : p4_convnext_tiny_100pct (convnext_tiny)
|
||||||
|
Device: cuda
|
||||||
|
Result: FAKE (confidence: 74.7%)
|
||||||
|
P(fake): 0.7466 P(real): 0.2534
|
||||||
|
```
|
||||||
|
|
||||||
|
If you omit `--checkpoint`, the tool automatically looks for a saved
|
||||||
|
checkpoint under `classifier/outputs/models/` — first the single-run
|
||||||
|
`{run_name}_best.pt`, then CV fold files `{run_name}_fold{k}_best.pt`, then
|
||||||
|
`{run_name}_fold{k}_final.pt`. To use a specific fold:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 -m pipeline offers --select-offer
|
python classifier/tools/inference.py image.jpg classifier/configs/phase4/p4_convnext_tiny_100pct.json \
|
||||||
|
--checkpoint classifier/outputs/models/p4_convnext_tiny_100pct_fold0_best.pt
|
||||||
```
|
```
|
||||||
|
|
||||||
You can override the ranking mode per run:
|
## Generator — training
|
||||||
|
|
||||||
|
From the repository root:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 -m pipeline offers --sort price
|
python generator/run.py generator/configs/phase0/p0_vae.json
|
||||||
python3 -m pipeline offers --sort performance
|
python generator/run.py generator/configs/phase0/p0_ddpm.json
|
||||||
python3 -m pipeline offers --sort performance --price 0.14
|
|
||||||
```
|
```
|
||||||
|
|
||||||
You can also filter by region:
|
Generator training expects real-face images (default source is `wiki`); use
|
||||||
|
`--data-dir` to point at your dataset tree. Checkpoints are saved under
|
||||||
|
`generator/outputs/models/{run_name}_final_ema.pt` (EMA shadow) and
|
||||||
|
`{run_name}_best_ema.pt` (lowest-FID snapshot).
|
||||||
|
|
||||||
|
## Generator — inference (sampling)
|
||||||
|
|
||||||
|
Generate 4×4 sample grids from Phase 5 EMA checkpoints:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
python3 -m pipeline offers --select-offer --region europe
|
python generator/tools/sampling.py --models p5_gan p5_vae p5_ddpm --samples 10
|
||||||
python3 -m pipeline offers --select-offer --region Portugal
|
|
||||||
python3 -m pipeline offers --select-offer --region US
|
|
||||||
python3 -m pipeline offers --select-offer --region europe --price 0.14
|
|
||||||
```
|
```
|
||||||
|
|
||||||
To inspect which region strings are currently available from the search results:
|
Options:
|
||||||
|
|
||||||
```bash
|
- `--models` — which models to sample from (`p5_gan`, `p5_vae`, `p5_ddpm`;
|
||||||
python3 -m pipeline offers --list-regions
|
defaults to all three).
|
||||||
```
|
- `--samples` — number of grids per model (default 10).
|
||||||
|
- `--output-dir` — where to write the PNGs (default
|
||||||
|
`generator/outputs/samples/final_comparison/`).
|
||||||
|
- `--truncation` — optional latent truncation for the GAN (lower = less
|
||||||
|
diversity but sharper).
|
||||||
|
- `--device` — `cuda` or `cpu` (default: auto-detect).
|
||||||
|
|
||||||
That command:
|
Each grid is a 4×4 PNG of 16 images sampled from the model's EMA weights.
|
||||||
- ensures your SSH public key is registered with Vast.ai
|
GAN samples are drawn from random latent vectors, VAE samples decode from the
|
||||||
- searches offers using the filters in `pipeline/defaults/vast.json`
|
learned prior, and DDPM samples use 50-step DDIM.
|
||||||
- creates an instance
|
|
||||||
- waits for SSH readiness
|
|
||||||
- syncs the repo
|
|
||||||
- uploads `data/` when `--upload-data` is set
|
|
||||||
- runs `python3 classifier/run.py ...`
|
|
||||||
- downloads `classifier/outputs/`
|
|
||||||
- for generator runs, rsyncs `generator/outputs/` back every 25 epochs and again at completion
|
|
||||||
- destroys the instance automatically unless `--keep-on-failure` is set
|
|
||||||
|
|
||||||
Useful commands:
|
## Final takeaway
|
||||||
|
|
||||||
```bash
|
The project is best understood as a sequence of controlled decisions:
|
||||||
python3 -m pipeline up
|
|
||||||
python3 -m pipeline status <instance_id>
|
|
||||||
python3 -m pipeline down <instance_id>
|
|
||||||
```
|
|
||||||
|
|
||||||
To override the default Vast search/runtime settings, copy `pipeline/defaults/vast.json`, edit it, and pass:
|
1. cleanly define the data and preprocessing;
|
||||||
|
2. establish simple baselines;
|
||||||
|
3. improve one factor at a time;
|
||||||
|
4. compare model families using saved evidence;
|
||||||
|
5. report both performance and limitations.
|
||||||
|
|
||||||
```bash
|
The classifier becomes reliable through source-aware preprocessing, stronger
|
||||||
python3 -m pipeline run classifier/configs/phase3/p3_efficientnet_b0.json --pipeline-config /path/to/vast.override.json
|
pretrained backbones, and scaling. The generator improves by first locking the
|
||||||
```
|
face-aligned pipeline and then selecting the best recipe inside each model
|
||||||
|
family before the final GAN/VAE/DDPM comparison.
|
||||||
The default policy in `pipeline/defaults/vast.json` now targets:
|
|
||||||
- `1x` GPU
|
|
||||||
- `RTX 3090` or `RTX 3090 Ti`
|
|
||||||
- `<= $0.20/hour`
|
|
||||||
- sorted by `dlperf` descending
|
|
||||||
- uses `vastai/pytorch:latest` as the default image
|
|
||||||
|
|||||||
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"extends": "../shared.json",
|
||||||
|
"run_name": "smoke",
|
||||||
|
"backbone": "simple_cnn",
|
||||||
|
"cnn_preset": "micro",
|
||||||
|
"dropout": 0.0,
|
||||||
|
"epochs": 1,
|
||||||
|
"cv_folds": 2,
|
||||||
|
"image_size": 64,
|
||||||
|
"batch_size": 8,
|
||||||
|
"num_workers": 0,
|
||||||
|
"early_stopping_patience": 0,
|
||||||
|
"subsample": 1.0,
|
||||||
|
"augment": false,
|
||||||
|
"lr": 0.001,
|
||||||
|
"T_max": 1,
|
||||||
|
"data_dir": "data"
|
||||||
|
}
|
||||||
BIN
Binary file not shown.
+228
-240
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
+2
-2
@@ -19,7 +19,7 @@
|
|||||||
"- Do augmentation and source-holdout runs reveal instability in attention?\n",
|
"- Do augmentation and source-holdout runs reveal instability in attention?\n",
|
||||||
"- Are errors visually plausible, or do they suggest shortcut behavior?\n",
|
"- Are errors visually plausible, or do they suggest shortcut behavior?\n",
|
||||||
"\n",
|
"\n",
|
||||||
"Roadmap link: after this qualitative check, `06_phase3_model_family_analysis.ipynb` compares stronger pretrained backbones and `07_phase4_data_scaling_analysis.ipynb` records the planned data-scaling analysis.\n"
|
"Roadmap link: after this qualitative check, `06_model_families.ipynb` compares stronger pretrained backbones and `07_data_scaling.ipynb` records the planned data-scaling analysis.\n"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1701,7 +1701,7 @@
|
|||||||
"\n",
|
"\n",
|
||||||
"The key limitation from Phase 2 still stands: high in-distribution AUC does not guarantee source-agnostic generalization. The Grad-CAM panels help make that limitation visible, but the source-holdout pairwise AUC values are the primary quantitative evidence.\n",
|
"The key limitation from Phase 2 still stands: high in-distribution AUC does not guarantee source-agnostic generalization. The Grad-CAM panels help make that limitation visible, but the source-holdout pairwise AUC values are the primary quantitative evidence.\n",
|
||||||
"\n",
|
"\n",
|
||||||
"Next: `06_phase3_model_family_analysis.ipynb` asks whether stronger pretrained model families improve on the selected Phase 2 pipeline.\n"
|
"Next: `06_model_families.ipynb` asks whether stronger pretrained model families improve on the selected Phase 2 pipeline.\n"
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
+2
-2
@@ -12,7 +12,7 @@
|
|||||||
"\n",
|
"\n",
|
||||||
"This notebook is evidence-only. It reads existing configs and logs, uses the saved log schema as the source of truth, and does not train, reevaluate, or invent missing results.\n",
|
"This notebook is evidence-only. It reads existing configs and logs, uses the saved log schema as the source of truth, and does not train, reevaluate, or invent missing results.\n",
|
||||||
"\n",
|
"\n",
|
||||||
"Roadmap link: `01_eda` -> `02_preprocessing` -> `03_phase1_analysis` -> `04_phase2_analysis` -> `05_gradcam_analysis` -> this model-family comparison -> `07_phase4_data_scaling_analysis`.\n"
|
"Roadmap link: `01_eda` -> `02_preprocessing` -> `03_baselines` -> `04_ablation_questions` -> `05_gradcam` -> this model-family comparison -> `07_data_scaling`.\n"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1311,7 +1311,7 @@
|
|||||||
"\n",
|
"\n",
|
||||||
"The report decision should therefore be: Phase 2 found the right input pipeline, and Phase 3 selects ResNet50 as the best classifier backbone for this task. ConvNeXt-Tiny is the AUC winner, but ResNet50 is the better detector when the full metric set is considered. The remaining caution is unchanged: source-wise and pairwise behavior must stay in the analysis because high global AUC alone does not prove source-agnostic generalization.\n",
|
"The report decision should therefore be: Phase 2 found the right input pipeline, and Phase 3 selects ResNet50 as the best classifier backbone for this task. ConvNeXt-Tiny is the AUC winner, but ResNet50 is the better detector when the full metric set is considered. The remaining caution is unchanged: source-wise and pairwise behavior must stay in the analysis because high global AUC alone does not prove source-agnostic generalization.\n",
|
||||||
"\n",
|
"\n",
|
||||||
"Next: `07_phase4_data_scaling_analysis.ipynb` turns from architecture choice to data scaling. It asks whether the selected top model families, especially ResNet50, improve further when trained with more of the available data.\n"
|
"Next: `07_data_scaling.ipynb` turns from architecture choice to data scaling. It asks whether the selected top model families, especially ResNet50, improve further when trained with more of the available data.\n"
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
+4
-4
@@ -1878,9 +1878,9 @@
|
|||||||
"id": "3a55e559",
|
"id": "3a55e559",
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"source": [
|
"source": [
|
||||||
"The 100% confusion matrices show that all three models are very close overall once they are trained with the full dataset. The deciding factor is the real-image error pattern: if the goal is to avoid labeling genuine images as fake, ConvNeXt-Tiny is the best practical choice because it has the lowest false positive rate and the best balanced accuracy. ResNet50 still has the strongest fake recall and macro F1, so it remains a strong alternative when missed fakes are the bigger concern. EfficientNet-B0 sits between them: strong and efficient, but not the top choice on either error profile.\n",
|
"The 100% confusion matrices show that all three models are very close overall once they are trained with the full dataset. The deciding factor is the pattern of false negatives (missed fakes): if the goal is to avoid missing fake images, ResNet50 is the best practical choice because it has the highest fake recall and the strongest macro F1. ConvNeXt-Tiny still has the lowest false positive rate and the best balanced accuracy, so it remains the safer choice when avoiding false alarms on real images is the priority. EfficientNet-B0 sits between them: strong and efficient, but not the top choice on either error profile.\n",
|
||||||
"\n",
|
"\n",
|
||||||
"For this project, ConvNeXt-Tiny is the best practical classifier once the full-dataset confusion matrices are considered. It matches the top group in overall performance while making the fewest mistakes on real images, which is the most important operational priority here. ResNet50 remains the strongest detector for catching fakes, but ConvNeXt-Tiny is the safer final choice when false positives on real images matter most."
|
"For this project, ResNet50 is the best practical classifier when the main operational priority is to minimize missed fakes (false negatives). It matches the top group in overall performance while giving the strongest fake detection metrics. ConvNeXt-Tiny remains the preferred alternative when false positives on real images are the biggest concern."
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1894,7 +1894,7 @@
|
|||||||
"|---|---|---|\n",
|
"|---|---|---|\n",
|
||||||
"| Data scale | Use 100% for the final Phase 4 setting | Every backbone improves from 20% to 50% to 100% across AUC, accuracy, and F1. |\n",
|
"| Data scale | Use 100% for the final Phase 4 setting | Every backbone improves from 20% to 50% to 100% across AUC, accuracy, and F1. |\n",
|
||||||
"| Best AUC model | ConvNeXt-Tiny at 100% | Highest mean AUC: `0.9954`, with very tight fold stability. |\n",
|
"| Best AUC model | ConvNeXt-Tiny at 100% | Highest mean AUC: `0.9954`, with very tight fold stability. |\n",
|
||||||
"| Best practical detector | ConvNeXt-Tiny at 100% | Lowest false positive rate on real images and best balanced accuracy in the full-dataset confusion matrices. |\n",
|
"| Best practical detector (minimize missed fakes) | ResNet50 at 100% | Highest fake recall and strongest macro F1 at 100%, making it the preferred choice when false negatives are the main concern. |\n",
|
||||||
"| Efficient model alternative | EfficientNet-B0 at 100% | Very close AUC `0.9949` with much smaller checkpoint size, but slightly weaker operating metrics. |\n",
|
"| Efficient model alternative | EfficientNet-B0 at 100% | Very close AUC `0.9949` with much smaller checkpoint size, but slightly weaker operating metrics. |\n",
|
||||||
"| Remaining limitation | Source behavior still matters | `insight` remains the weakest fake source, even though scaling raises it to around `0.990` pairwise AUC. |"
|
"| Remaining limitation | Source behavior still matters | `insight` remains the weakest fake source, even though scaling raises it to around `0.990` pairwise AUC. |"
|
||||||
]
|
]
|
||||||
@@ -1942,7 +1942,7 @@
|
|||||||
"\n",
|
"\n",
|
||||||
"Phase 4 shows that data scale is a real improvement lever. At 20%, the top backbones were already strong, and with 50% and 100% data they converge to very similar overall performance. ConvNeXt-Tiny remains the best AUC model at `0.9954`, while ResNet50 remains strongest on fake-catching thresholded metrics (accuracy/F1 and fake recall).\n",
|
"Phase 4 shows that data scale is a real improvement lever. At 20%, the top backbones were already strong, and with 50% and 100% data they converge to very similar overall performance. ConvNeXt-Tiny remains the best AUC model at `0.9954`, while ResNet50 remains strongest on fake-catching thresholded metrics (accuracy/F1 and fake recall).\n",
|
||||||
"\n",
|
"\n",
|
||||||
"For the final practical decision, the confusion matrices are decisive: because this project prioritizes avoiding false alarms on real images, `p4_convnext_tiny_100pct` is the recommended classifier. It achieves top-tier overall performance while producing the lowest false positive rate and best balanced accuracy at 100% data. ResNet50 stays the preferred alternative when the priority is maximizing fake detection, and source-wise diagnostics remain important because `insight` is still the weakest fake source."
|
"For the final practical decision, the confusion matrices are decisive. Because this project prioritizes avoiding missed fakes (false negatives), `p4_resnet50_100pct` is the recommended classifier. It achieves top-tier fake detection metrics while staying competitive on overall performance. ConvNeXt-Tiny remains the preferred alternative when minimizing false positives on real images is the operational priority, and source-wise diagnostics remain important because `insight` is still the weakest fake source."
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
@@ -1,16 +0,0 @@
|
|||||||
run,n_candidates,n_images,heldout_source,candidate_auc,candidate_acc,panel,expanded_panel,fine_panel
|
|
||||||
p1_simplecnn_baseline,192,10,,0.8378182870370371,0.7395833333333334,classifier\outputs\gradcam\p1_simplecnn_baseline\panel.png,,
|
|
||||||
p1_resnet18_baseline,192,10,,0.9769965277777778,0.9270833333333334,classifier\outputs\gradcam\p1_resnet18_baseline\panel.png,,
|
|
||||||
p2a_t1_original,192,10,,0.9984085648148149,0.9895833333333334,classifier\outputs\gradcam\p2a_t1_original\panel.png,,
|
|
||||||
p2a_t2_real_norm,192,10,,0.9939236111111112,0.9791666666666666,classifier\outputs\gradcam\p2a_t2_real_norm\panel.png,,
|
|
||||||
p2a_t3_holdout_text2img,240,10,text2img,0.9264322916666667,0.775,classifier\outputs\gradcam\p2a_t3_holdout_text2img\panel.png,,
|
|
||||||
p2a_t3_holdout_inpainting,240,10,inpainting,0.9819878472222222,0.9333333333333333,classifier\outputs\gradcam\p2a_t3_holdout_inpainting\panel.png,,
|
|
||||||
p2a_t3_holdout_insight,240,10,insight,0.9549696180555556,0.7625,classifier\outputs\gradcam\p2a_t3_holdout_insight\panel.png,,
|
|
||||||
p2b_simplecnn_224,192,10,,0.8207465277777778,0.7447916666666666,classifier\outputs\gradcam\p2b_simplecnn_224\panel.png,,
|
|
||||||
p2b_resnet18_224,192,10,,0.9984085648148149,0.9895833333333334,classifier\outputs\gradcam\p2b_resnet18_224\panel.png,,
|
|
||||||
p2c_simplecnn_facecrop,192,10,,0.8058449074074073,0.7552083333333334,classifier\outputs\gradcam\p2c_simplecnn_facecrop\panel.png,,
|
|
||||||
p2c_resnet18_facecrop,192,16,,0.9911747685185185,0.890625,classifier\outputs\gradcam\p2c_resnet18_facecrop\panel.png,classifier\outputs\gradcam\p2c_resnet18_facecrop\panel_expanded.png,classifier\outputs\gradcam\p2c_resnet18_facecrop\panel_fine_layer3.png
|
|
||||||
p2d_simplecnn_aug,192,10,,0.7565104166666666,0.65625,classifier\outputs\gradcam\p2d_simplecnn_aug\panel.png,,
|
|
||||||
p2d_resnet18_aug,192,10,,0.9733796296296297,0.9270833333333334,classifier\outputs\gradcam\p2d_resnet18_aug\panel.png,,
|
|
||||||
p2e_simplecnn_facecrop_aug,192,10,,0.7358217592592592,0.6510416666666666,classifier\outputs\gradcam\p2e_simplecnn_facecrop_aug\panel.png,,
|
|
||||||
p2e_resnet18_facecrop_aug,192,10,,0.9910300925925927,0.9322916666666666,classifier\outputs\gradcam\p2e_resnet18_facecrop_aug\panel.png,,
|
|
||||||
|
@@ -1,4 +0,0 @@
|
|||||||
panel,path
|
|
||||||
architecture capacity,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\evolution\05_evolution_architecture_capacity.png
|
|
||||||
resnet pipeline,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\evolution\05_evolution_resnet_pipeline.png
|
|
||||||
simplecnn pipeline,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\evolution\05_evolution_simplecnn_pipeline.png
|
|
||||||
|
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p1_resnet18_baseline",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9270833333333334,
|
|
||||||
"auc_roc": 0.9769965277777778,
|
|
||||||
"f1": 0.950354609929078,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
44,
|
|
||||||
4
|
|
||||||
],
|
|
||||||
[
|
|
||||||
10,
|
|
||||||
134
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 7.202164852060378e-06,
|
|
||||||
"logit": -11.841121673583984,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\83\\2667583_1985-05-02_2009.jpg",
|
|
||||||
"basename": "2667583_1985-05-02_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 8.81600226421142e-06,
|
|
||||||
"logit": -11.638933181762695,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\08\\3523408_1963-02-01_2015.jpg",
|
|
||||||
"basename": "3523408_1963-02-01_2015.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.020292282104492,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\17\\32590217_1942-09-19_2014.jpg",
|
|
||||||
"basename": "32590217_1942-09-19_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999988079071045,
|
|
||||||
"logit": 13.679031372070312,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 18.888574600219727,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\97\\3832797_1984-01-19_2010.jpg",
|
|
||||||
"basename": "3832797_1984-01-19_2010.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9967204928398132,
|
|
||||||
"logit": 5.7167792320251465,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\49\\25392649_1991-05-29_2013.jpg",
|
|
||||||
"basename": "25392649_1991-05-29_2013.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8840320110321045,
|
|
||||||
"logit": 2.031179428100586,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\56\\14853156_1981-05-28_2009.jpg",
|
|
||||||
"basename": "14853156_1981-05-28_2009.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0010116391349583864,
|
|
||||||
"logit": -6.895171165466309,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\10\\726410_1912-05-31_1997.jpg",
|
|
||||||
"basename": "726410_1912-05-31_1997.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0036017918027937412,
|
|
||||||
"logit": -5.622715473175049,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\81\\3028481_1951-07-05_1977.jpg",
|
|
||||||
"basename": "3028481_1951-07-05_1977.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4881739914417267,
|
|
||||||
"logit": -0.04731296747922897,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\01_confident_true_wiki_wiki_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\02_confident_true_wiki_wiki_2667583_1985-05-02_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\03_confident_true_inpainting_inpainting_3523408_1963-02-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\04_confident_true_insight_insight_32590217_1942-09-19_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\05_confident_true_text2img_text2img_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\06_strong_false_positive_wiki_3832797_1984-01-19_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\07_strong_false_positive_wiki_25392649_1991-05-29_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\08_strong_false_negative_insight_14853156_1981-05-28_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\09_strong_false_negative_insight_726410_1912-05-31_1997.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\10_borderline_wiki_3028481_1951-07-05_1977.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_resnet18_baseline\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p1_simplecnn_baseline",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.7395833333333334,
|
|
||||||
"auc_roc": 0.8378182870370371,
|
|
||||||
"f1": 0.8031496062992126,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
40,
|
|
||||||
8
|
|
||||||
],
|
|
||||||
[
|
|
||||||
42,
|
|
||||||
102
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\67\\779067_1968-10-08_2008.jpg",
|
|
||||||
"basename": "779067_1968-10-08_2008.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.06221456080675125,
|
|
||||||
"logit": -2.7129321098327637,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.10906809568405151,
|
|
||||||
"logit": -2.1002955436706543,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9499694108963013,
|
|
||||||
"logit": 2.9437952041625977,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9545132517814636,
|
|
||||||
"logit": 3.0437798500061035,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\89\\9507989_1976-07-19_2007.jpg",
|
|
||||||
"basename": "9507989_1976-07-19_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9725574851036072,
|
|
||||||
"logit": 3.567837715148926,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8278583884239197,
|
|
||||||
"logit": 1.5705249309539795,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8256536722183228,
|
|
||||||
"logit": 1.555131196975708,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\25\\1932725_1949-01-25_1979.jpg",
|
|
||||||
"basename": "1932725_1949-01-25_1979.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.1434662640094757,
|
|
||||||
"logit": -1.786793828010559,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\41\\37585341_1994-09-01_2015.jpg",
|
|
||||||
"basename": "37585341_1994-09-01_2015.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.1799710988998413,
|
|
||||||
"logit": -1.5165432691574097,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\74\\3290874_1984-06-18_2004.jpg",
|
|
||||||
"basename": "3290874_1984-06-18_2004.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4988362491130829,
|
|
||||||
"logit": -0.004655048251152039,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\01_confident_true_wiki_wiki_779067_1968-10-08_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\02_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\03_confident_true_inpainting_inpainting_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\04_confident_true_insight_insight_34379104_1950-12-07_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\05_confident_true_text2img_text2img_9507989_1976-07-19_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\06_strong_false_positive_wiki_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\07_strong_false_positive_wiki_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\08_strong_false_negative_text2img_1932725_1949-01-25_1979.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\09_strong_false_negative_inpainting_37585341_1994-09-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\10_borderline_inpainting_3290874_1984-06-18_2004.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p1_simplecnn_baseline\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2a_t1_original",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9895833333333334,
|
|
||||||
"auc_roc": 0.9984085648148149,
|
|
||||||
"f1": 0.993006993006993,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
48,
|
|
||||||
0
|
|
||||||
],
|
|
||||||
[
|
|
||||||
2,
|
|
||||||
142
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\74\\3290874_1984-06-18_2004.jpg",
|
|
||||||
"basename": "3290874_1984-06-18_2004.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 3.208382226560502e-09,
|
|
||||||
"logit": -19.557498931884766,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 6.348270176204096e-08,
|
|
||||||
"logit": -16.572498321533203,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.80807876586914,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\81\\3028481_1951-07-05_1977.jpg",
|
|
||||||
"basename": "3028481_1951-07-05_1977.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 19.85223960876465,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\10\\726410_1912-05-31_1997.jpg",
|
|
||||||
"basename": "726410_1912-05-31_1997.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.96767807006836,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.00015543345944024622,
|
|
||||||
"logit": -8.769137382507324,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\10\\726410_1912-05-31_1997.jpg",
|
|
||||||
"basename": "726410_1912-05-31_1997.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0615716315805912,
|
|
||||||
"logit": -2.7240052223205566,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\57\\31543457_1910-02-06_1970.jpg",
|
|
||||||
"basename": "31543457_1910-02-06_1970.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5928639769554138,
|
|
||||||
"logit": 0.3758176565170288,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7217857837677002,
|
|
||||||
"logit": 0.9533369541168213,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.897236168384552,
|
|
||||||
"logit": 2.1668851375579834,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\01_confident_true_wiki_wiki_3290874_1984-06-18_2004.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\02_confident_true_wiki_wiki_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\03_confident_true_inpainting_inpainting_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\04_confident_true_insight_insight_3028481_1951-07-05_1977.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\05_confident_true_text2img_text2img_726410_1912-05-31_1997.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\06_strong_false_negative_insight_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\07_strong_false_negative_inpainting_726410_1912-05-31_1997.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\08_borderline_inpainting_31543457_1910-02-06_1970.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\09_borderline_insight_3720944_1920-12-28_1963.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\10_expanded_context_insight_34379104_1950-12-07_2013.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t1_original\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2a_t2_real_norm",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9791666666666666,
|
|
||||||
"auc_roc": 0.9939236111111112,
|
|
||||||
"f1": 0.9861111111111112,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
46,
|
|
||||||
2
|
|
||||||
],
|
|
||||||
[
|
|
||||||
2,
|
|
||||||
142
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\21\\24644621_1964-12-15_2010.jpg",
|
|
||||||
"basename": "24644621_1964-12-15_2010.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 2.2610427549807355e-05,
|
|
||||||
"logit": -10.697076797485352,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\46\\6385546_1980-05-23_2013.jpg",
|
|
||||||
"basename": "6385546_1980-05-23_2013.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 8.002227696124464e-05,
|
|
||||||
"logit": -9.433125495910645,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999998807907104,
|
|
||||||
"logit": 16.21304702758789,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\83\\2667583_1985-05-02_2009.jpg",
|
|
||||||
"basename": "2667583_1985-05-02_2009.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999998807907104,
|
|
||||||
"logit": 15.857626914978027,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 18.31481170654297,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\02\\10386202_1915-02-10_1949.jpg",
|
|
||||||
"basename": "10386202_1915-02-10_1949.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8437501788139343,
|
|
||||||
"logit": 1.686400294303894,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\81\\3028481_1951-07-05_1977.jpg",
|
|
||||||
"basename": "3028481_1951-07-05_1977.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5785955786705017,
|
|
||||||
"logit": 0.31701087951660156,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0017513168277218938,
|
|
||||||
"logit": -6.345634460449219,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\21\\24644621_1964-12-15_2010.jpg",
|
|
||||||
"basename": "24644621_1964-12-15_2010.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.48223310708999634,
|
|
||||||
"logit": -0.0710974782705307,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4771732687950134,
|
|
||||||
"logit": -0.0913703665137291,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\01_confident_true_wiki_wiki_24644621_1964-12-15_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\02_confident_true_wiki_wiki_6385546_1980-05-23_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\03_confident_true_inpainting_inpainting_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\04_confident_true_insight_insight_2667583_1985-05-02_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\05_confident_true_text2img_text2img_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\06_strong_false_positive_wiki_10386202_1915-02-10_1949.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\07_strong_false_positive_wiki_3028481_1951-07-05_1977.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\08_strong_false_negative_insight_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\09_strong_false_negative_insight_24644621_1964-12-15_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\10_borderline_wiki_3720944_1920-12-28_1963.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t2_real_norm\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2a_t3_holdout_inpainting",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 240,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9333333333333333,
|
|
||||||
"auc_roc": 0.9819878472222222,
|
|
||||||
"f1": 0.9567567567567568,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
47,
|
|
||||||
1
|
|
||||||
],
|
|
||||||
[
|
|
||||||
15,
|
|
||||||
177
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": "inpainting",
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 8.817832686247584e-09,
|
|
||||||
"logit": -18.546489715576172,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\70\\22732870_1995-03-08_2014.jpg",
|
|
||||||
"basename": "22732870_1995-03-08_2014.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 8.841730902986455e-09,
|
|
||||||
"logit": -18.54378318786621,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\32\\39602032_1988-03-10_2013.jpg",
|
|
||||||
"basename": "39602032_1988-03-10_2013.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 16.797229766845703,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\19\\14136419_1997-11-01_2013.jpg",
|
|
||||||
"basename": "14136419_1997-11-01_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.396644592285156,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\08\\3523408_1963-02-01_2015.jpg",
|
|
||||||
"basename": "3523408_1963-02-01_2015.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 21.35198211669922,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\81\\3028481_1951-07-05_1977.jpg",
|
|
||||||
"basename": "3028481_1951-07-05_1977.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999240636825562,
|
|
||||||
"logit": 9.485453605651855,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\92\\3958892_1987-01-22_2013.jpg",
|
|
||||||
"basename": "3958892_1987-01-22_2013.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 5.744245572714135e-05,
|
|
||||||
"logit": -9.764669418334961,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.00028320401906967163,
|
|
||||||
"logit": -8.169059753417969,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\02\\10386202_1915-02-10_1949.jpg",
|
|
||||||
"basename": "10386202_1915-02-10_1949.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.41957107186317444,
|
|
||||||
"logit": -0.32453441619873047,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\56\\22613456_1986-02-10_2012.jpg",
|
|
||||||
"basename": "22613456_1986-02-10_2012.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6162223815917969,
|
|
||||||
"logit": 0.473544716835022,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\01_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\02_confident_true_wiki_wiki_22732870_1995-03-08_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\03_confident_true_inpainting_inpainting_39602032_1988-03-10_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\04_confident_true_insight_insight_14136419_1997-11-01_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\05_confident_true_text2img_text2img_3523408_1963-02-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\06_strong_false_positive_wiki_3028481_1951-07-05_1977.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\07_strong_false_negative_inpainting_3958892_1987-01-22_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\08_strong_false_negative_insight_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\09_borderline_insight_10386202_1915-02-10_1949.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\10_borderline_inpainting_22613456_1986-02-10_2012.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_inpainting\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2a_t3_holdout_insight",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 240,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.7625,
|
|
||||||
"auc_roc": 0.9549696180555556,
|
|
||||||
"f1": 0.8267477203647416,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
47,
|
|
||||||
1
|
|
||||||
],
|
|
||||||
[
|
|
||||||
56,
|
|
||||||
136
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": "insight",
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\74\\3290874_1984-06-18_2004.jpg",
|
|
||||||
"basename": "3290874_1984-06-18_2004.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 5.1686233418224425e-12,
|
|
||||||
"logit": -25.988414764404297,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 9.782129967161879e-12,
|
|
||||||
"logit": -25.3504638671875,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\15\\16719015_1940-10-10_1964.jpg",
|
|
||||||
"basename": "16719015_1940-10-10_1964.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 20.643247604370117,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\62\\38862_1966-09-09_2005.jpg",
|
|
||||||
"basename": "38862_1966-09-09_2005.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.999998927116394,
|
|
||||||
"logit": 13.801375389099121,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\08\\3523408_1963-02-01_2015.jpg",
|
|
||||||
"basename": "3523408_1963-02-01_2015.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 18.072370529174805,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999977350234985,
|
|
||||||
"logit": 13.005270004272461,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\12\\11714712_1959-02-08_2007.jpg",
|
|
||||||
"basename": "11714712_1959-02-08_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 1.4099471590256485e-10,
|
|
||||||
"logit": -22.68229866027832,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\05\\308405_1966-11-29_2008.jpg",
|
|
||||||
"basename": "308405_1966-11-29_2008.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 8.343223001361366e-09,
|
|
||||||
"logit": -18.601816177368164,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\07\\30502207_1953-11-09_1992.jpg",
|
|
||||||
"basename": "30502207_1953-11-09_1992.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4951499104499817,
|
|
||||||
"logit": -0.019400928169488907,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\20\\22164720_1946-06-24_2007.jpg",
|
|
||||||
"basename": "22164720_1946-06-24_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5142862796783447,
|
|
||||||
"logit": 0.05716080591082573,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\01_confident_true_wiki_wiki_3290874_1984-06-18_2004.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\02_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\03_confident_true_inpainting_inpainting_16719015_1940-10-10_1964.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\04_confident_true_insight_insight_38862_1966-09-09_2005.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\05_confident_true_text2img_text2img_3523408_1963-02-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\06_strong_false_positive_wiki_3720944_1920-12-28_1963.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\07_strong_false_negative_insight_11714712_1959-02-08_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\08_strong_false_negative_insight_308405_1966-11-29_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\09_borderline_insight_30502207_1953-11-09_1992.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\10_borderline_insight_22164720_1946-06-24_2007.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_insight\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2a_t3_holdout_text2img",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 240,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.775,
|
|
||||||
"auc_roc": 0.9264322916666667,
|
|
||||||
"f1": 0.8383233532934131,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
46,
|
|
||||||
2
|
|
||||||
],
|
|
||||||
[
|
|
||||||
52,
|
|
||||||
140
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": "text2img",
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\23\\35796523_1952-12-01_1992.jpg",
|
|
||||||
"basename": "35796523_1952-12-01_1992.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 2.9518876232259572e-08,
|
|
||||||
"logit": -17.33823585510254,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\21\\24644621_1964-12-15_2010.jpg",
|
|
||||||
"basename": "24644621_1964-12-15_2010.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 5.34162118981385e-08,
|
|
||||||
"logit": -16.74515151977539,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\97\\3832797_1984-01-19_2010.jpg",
|
|
||||||
"basename": "3832797_1984-01-19_2010.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999995231628418,
|
|
||||||
"logit": 14.63173770904541,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\17\\32590217_1942-09-19_2014.jpg",
|
|
||||||
"basename": "32590217_1942-09-19_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999998807907104,
|
|
||||||
"logit": 16.324398040771484,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\50\\31704050_1988-05-11_2013.jpg",
|
|
||||||
"basename": "31704050_1988-05-11_2013.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999866485595703,
|
|
||||||
"logit": 11.222184181213379,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\35\\20728335_1988-07-07_2014.jpg",
|
|
||||||
"basename": "20728335_1988-07-07_2014.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9033180475234985,
|
|
||||||
"logit": 2.2346484661102295,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8656692504882812,
|
|
||||||
"logit": 1.8631978034973145,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\98\\43349098_1994-08-30_2014.jpg",
|
|
||||||
"basename": "43349098_1994-08-30_2014.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 7.18730461812811e-07,
|
|
||||||
"logit": -14.14577865600586,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\68\\14609668_1956-12-09_2009.jpg",
|
|
||||||
"basename": "14609668_1956-12-09_2009.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 9.012396731122863e-07,
|
|
||||||
"logit": -13.919493675231934,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\57\\6707957_1985-09-04_2015.jpg",
|
|
||||||
"basename": "6707957_1985-09-04_2015.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4953322410583496,
|
|
||||||
"logit": -0.018671687692403793,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\01_confident_true_wiki_wiki_35796523_1952-12-01_1992.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\02_confident_true_wiki_wiki_24644621_1964-12-15_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\03_confident_true_inpainting_inpainting_3832797_1984-01-19_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\04_confident_true_insight_insight_32590217_1942-09-19_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\05_confident_true_text2img_text2img_31704050_1988-05-11_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\06_strong_false_positive_wiki_20728335_1988-07-07_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\07_strong_false_positive_wiki_3720944_1920-12-28_1963.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\08_strong_false_negative_text2img_43349098_1994-08-30_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\09_strong_false_negative_text2img_14609668_1956-12-09_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\10_borderline_text2img_6707957_1985-09-04_2015.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2a_t3_holdout_text2img\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2b_resnet18_224",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9895833333333334,
|
|
||||||
"auc_roc": 0.9984085648148149,
|
|
||||||
"f1": 0.993006993006993,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
48,
|
|
||||||
0
|
|
||||||
],
|
|
||||||
[
|
|
||||||
2,
|
|
||||||
142
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\74\\3290874_1984-06-18_2004.jpg",
|
|
||||||
"basename": "3290874_1984-06-18_2004.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 3.208382226560502e-09,
|
|
||||||
"logit": -19.557498931884766,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 6.348270176204096e-08,
|
|
||||||
"logit": -16.572498321533203,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.80807876586914,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\81\\3028481_1951-07-05_1977.jpg",
|
|
||||||
"basename": "3028481_1951-07-05_1977.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 19.85223960876465,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\10\\726410_1912-05-31_1997.jpg",
|
|
||||||
"basename": "726410_1912-05-31_1997.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.96767807006836,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.00015543345944024622,
|
|
||||||
"logit": -8.769137382507324,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\10\\726410_1912-05-31_1997.jpg",
|
|
||||||
"basename": "726410_1912-05-31_1997.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0615716315805912,
|
|
||||||
"logit": -2.7240052223205566,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\57\\31543457_1910-02-06_1970.jpg",
|
|
||||||
"basename": "31543457_1910-02-06_1970.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5928639769554138,
|
|
||||||
"logit": 0.3758176565170288,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7217857837677002,
|
|
||||||
"logit": 0.9533369541168213,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.897236168384552,
|
|
||||||
"logit": 2.1668851375579834,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\01_confident_true_wiki_wiki_3290874_1984-06-18_2004.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\02_confident_true_wiki_wiki_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\03_confident_true_inpainting_inpainting_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\04_confident_true_insight_insight_3028481_1951-07-05_1977.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\05_confident_true_text2img_text2img_726410_1912-05-31_1997.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\06_strong_false_negative_insight_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\07_strong_false_negative_inpainting_726410_1912-05-31_1997.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\08_borderline_inpainting_31543457_1910-02-06_1970.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\09_borderline_insight_3720944_1920-12-28_1963.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\10_expanded_context_insight_34379104_1950-12-07_2013.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_resnet18_224\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2b_simplecnn_224",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.7447916666666666,
|
|
||||||
"auc_roc": 0.8207465277777778,
|
|
||||||
"f1": 0.8122605363984674,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
37,
|
|
||||||
11
|
|
||||||
],
|
|
||||||
[
|
|
||||||
38,
|
|
||||||
106
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.09357129782438278,
|
|
||||||
"logit": -2.2707886695861816,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\67\\779067_1968-10-08_2008.jpg",
|
|
||||||
"basename": "779067_1968-10-08_2008.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.12308616936206818,
|
|
||||||
"logit": -1.9635241031646729,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9532476663589478,
|
|
||||||
"logit": 3.0150113105773926,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9883313775062561,
|
|
||||||
"logit": 4.439114093780518,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\46\\6385546_1980-05-23_2013.jpg",
|
|
||||||
"basename": "6385546_1980-05-23_2013.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9493488669395447,
|
|
||||||
"logit": 2.9308156967163086,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.868328869342804,
|
|
||||||
"logit": 1.8862627744674683,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8662444949150085,
|
|
||||||
"logit": 1.8681539297103882,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\35\\20728335_1988-07-07_2014.jpg",
|
|
||||||
"basename": "20728335_1988-07-07_2014.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.21595068275928497,
|
|
||||||
"logit": -1.289421796798706,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\27\\2103427_1986-06-19_2012.jpg",
|
|
||||||
"basename": "2103427_1986-06-19_2012.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.21677419543266296,
|
|
||||||
"logit": -1.2845648527145386,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4993416368961334,
|
|
||||||
"logit": -0.002633415162563324,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\01_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\02_confident_true_wiki_wiki_779067_1968-10-08_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\03_confident_true_inpainting_inpainting_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\04_confident_true_insight_insight_34379104_1950-12-07_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\05_confident_true_text2img_text2img_6385546_1980-05-23_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\06_strong_false_positive_wiki_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\07_strong_false_positive_wiki_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\08_strong_false_negative_text2img_20728335_1988-07-07_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\09_strong_false_negative_insight_2103427_1986-06-19_2012.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\10_borderline_insight_9143052_1931-01-21_1991.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2b_simplecnn_224\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,204 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2c_resnet18_facecrop",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.890625,
|
|
||||||
"auc_roc": 0.9911747685185185,
|
|
||||||
"f1": 0.9219330855018587,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
47,
|
|
||||||
1
|
|
||||||
],
|
|
||||||
[
|
|
||||||
20,
|
|
||||||
124
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\74\\3290874_1984-06-18_2004.jpg",
|
|
||||||
"basename": "3290874_1984-06-18_2004.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 2.0144028667345992e-08,
|
|
||||||
"logit": -17.72035789489746,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 2.1860007137775028e-08,
|
|
||||||
"logit": -17.638607025146484,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\97\\3832797_1984-01-19_2010.jpg",
|
|
||||||
"basename": "3832797_1984-01-19_2010.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999995231628418,
|
|
||||||
"logit": 14.502941131591797,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\83\\2667583_1985-05-02_2009.jpg",
|
|
||||||
"basename": "2667583_1985-05-02_2009.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 17.030344009399414,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\79\\25469179_1988-04-23_2008.jpg",
|
|
||||||
"basename": "25469179_1988-04-23_2008.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999992847442627,
|
|
||||||
"logit": 14.198725700378418,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\57\\31543457_1910-02-06_1970.jpg",
|
|
||||||
"basename": "31543457_1910-02-06_1970.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6529371738433838,
|
|
||||||
"logit": 0.6319749355316162,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\49\\25392649_1991-05-29_2013.jpg",
|
|
||||||
"basename": "25392649_1991-05-29_2013.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.004319223575294018,
|
|
||||||
"logit": -5.4403510093688965,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\15\\16719015_1940-10-10_1964.jpg",
|
|
||||||
"basename": "16719015_1940-10-10_1964.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.007250246591866016,
|
|
||||||
"logit": -4.919443130493164,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\23\\35796523_1952-12-01_1992.jpg",
|
|
||||||
"basename": "35796523_1952-12-01_1992.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5090859532356262,
|
|
||||||
"logit": 0.03634767234325409,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\02\\10386202_1915-02-10_1949.jpg",
|
|
||||||
"basename": "10386202_1915-02-10_1949.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5188130736351013,
|
|
||||||
"logit": 0.07528790831565857,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\08\\3523408_1963-02-01_2015.jpg",
|
|
||||||
"basename": "3523408_1963-02-01_2015.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.46507686376571655,
|
|
||||||
"logit": -0.1399204581975937,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\61\\9200161_1991-02-11_2014.jpg",
|
|
||||||
"basename": "9200161_1991-02-11_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4395407438278198,
|
|
||||||
"logit": -0.24302612245082855,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\39\\451239_1954-05-11_2007.jpg",
|
|
||||||
"basename": "451239_1954-05-11_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.39794081449508667,
|
|
||||||
"logit": -0.4140525758266449,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\00\\12611800_1950-08-17_2011.jpg",
|
|
||||||
"basename": "12611800_1950-08-17_2011.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6231862902641296,
|
|
||||||
"logit": 0.5030946731567383,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\41\\37585341_1994-09-01_2015.jpg",
|
|
||||||
"basename": "37585341_1994-09-01_2015.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.34164801239967346,
|
|
||||||
"logit": -0.6559587717056274,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\25\\1932725_1949-01-25_1979.jpg",
|
|
||||||
"basename": "1932725_1949-01-25_1979.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.3312528729438782,
|
|
||||||
"logit": -0.7025240063667297,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\01_confident_true_wiki_wiki_3290874_1984-06-18_2004.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\02_confident_true_wiki_wiki_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\03_confident_true_inpainting_inpainting_3832797_1984-01-19_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\04_confident_true_insight_insight_2667583_1985-05-02_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\05_confident_true_text2img_text2img_25469179_1988-04-23_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\06_strong_false_positive_wiki_31543457_1910-02-06_1970.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\07_strong_false_negative_text2img_25392649_1991-05-29_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\08_strong_false_negative_inpainting_16719015_1940-10-10_1964.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\09_borderline_insight_35796523_1952-12-01_1992.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\10_borderline_text2img_10386202_1915-02-10_1949.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\11_expanded_context_text2img_3523408_1963-02-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\12_expanded_context_insight_9200161_1991-02-11_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\13_expanded_context_text2img_451239_1954-05-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\14_expanded_context_insight_12611800_1950-08-17_2011.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\15_expanded_context_inpainting_37585341_1994-09-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\16_expanded_context_insight_1932725_1949-01-25_1979.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\panel.png",
|
|
||||||
"expanded_panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\panel_expanded.png",
|
|
||||||
"fine_panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_resnet18_facecrop\\panel_fine_layer3.png"
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2c_simplecnn_facecrop",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.7552083333333334,
|
|
||||||
"auc_roc": 0.8058449074074073,
|
|
||||||
"f1": 0.8226415094339623,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
36,
|
|
||||||
12
|
|
||||||
],
|
|
||||||
[
|
|
||||||
35,
|
|
||||||
109
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\93\\28699293_1957-04-26_2009.jpg",
|
|
||||||
"basename": "28699293_1957-04-26_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.09115341305732727,
|
|
||||||
"logit": -2.2996323108673096,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.10735085606575012,
|
|
||||||
"logit": -2.118091106414795,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9165919423103333,
|
|
||||||
"logit": 2.3969175815582275,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\17\\32590217_1942-09-19_2014.jpg",
|
|
||||||
"basename": "32590217_1942-09-19_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7911618947982788,
|
|
||||||
"logit": 1.331943154335022,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\99\\7319399_1981-08-23_2010.jpg",
|
|
||||||
"basename": "7319399_1981-08-23_2010.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9600318074226379,
|
|
||||||
"logit": 3.1788814067840576,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\57\\31543457_1910-02-06_1970.jpg",
|
|
||||||
"basename": "31543457_1910-02-06_1970.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9373878836631775,
|
|
||||||
"logit": 2.7061376571655273,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7406934499740601,
|
|
||||||
"logit": 1.0495760440826416,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\91\\42596391_1977-03-09_2009.jpg",
|
|
||||||
"basename": "42596391_1977-03-09_2009.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.13060443103313446,
|
|
||||||
"logit": -1.895625114440918,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\93\\28699293_1957-04-26_2009.jpg",
|
|
||||||
"basename": "28699293_1957-04-26_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.16693732142448425,
|
|
||||||
"logit": -1.6074904203414917,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\23\\35796523_1952-12-01_1992.jpg",
|
|
||||||
"basename": "35796523_1952-12-01_1992.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5004183053970337,
|
|
||||||
"logit": 0.0016733184456825256,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\01_confident_true_wiki_wiki_28699293_1957-04-26_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\02_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\03_confident_true_inpainting_inpainting_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\04_confident_true_insight_insight_32590217_1942-09-19_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\05_confident_true_text2img_text2img_7319399_1981-08-23_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\06_strong_false_positive_wiki_31543457_1910-02-06_1970.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\07_strong_false_positive_wiki_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\08_strong_false_negative_insight_42596391_1977-03-09_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\09_strong_false_negative_inpainting_28699293_1957-04-26_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\10_borderline_inpainting_35796523_1952-12-01_1992.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2c_simplecnn_facecrop\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2d_resnet18_aug",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9270833333333334,
|
|
||||||
"auc_roc": 0.9733796296296297,
|
|
||||||
"f1": 0.948905109489051,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
48,
|
|
||||||
0
|
|
||||||
],
|
|
||||||
[
|
|
||||||
14,
|
|
||||||
130
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\67\\779067_1968-10-08_2008.jpg",
|
|
||||||
"basename": "779067_1968-10-08_2008.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.00046918127918615937,
|
|
||||||
"logit": -7.6640520095825195,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\08\\3523408_1963-02-01_2015.jpg",
|
|
||||||
"basename": "3523408_1963-02-01_2015.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0006820649723522365,
|
|
||||||
"logit": -7.289703369140625,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\97\\3832797_1984-01-19_2010.jpg",
|
|
||||||
"basename": "3832797_1984-01-19_2010.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999998807907104,
|
|
||||||
"logit": 15.925747871398926,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\97\\3832797_1984-01-19_2010.jpg",
|
|
||||||
"basename": "3832797_1984-01-19_2010.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999997615814209,
|
|
||||||
"logit": 15.087677001953125,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\33\\12318533_1982-06-16_2007.jpg",
|
|
||||||
"basename": "12318533_1982-06-16_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9999996423721313,
|
|
||||||
"logit": 14.999021530151367,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\51\\18249151_1970-12-16_2004.jpg",
|
|
||||||
"basename": "18249151_1970-12-16_2004.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0321250855922699,
|
|
||||||
"logit": -3.405465602874756,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.0697367787361145,
|
|
||||||
"logit": -2.5907397270202637,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\27\\2103427_1986-06-19_2012.jpg",
|
|
||||||
"basename": "2103427_1986-06-19_2012.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4919249415397644,
|
|
||||||
"logit": -0.03230302780866623,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\72\\43568472_1968-07-26_2014.jpg",
|
|
||||||
"basename": "43568472_1968-07-26_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5248959064483643,
|
|
||||||
"logit": 0.09966614097356796,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.47507616877555847,
|
|
||||||
"logit": -0.09977804869413376,
|
|
||||||
"selection": "expanded_context"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\01_confident_true_wiki_wiki_779067_1968-10-08_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\02_confident_true_wiki_wiki_3523408_1963-02-01_2015.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\03_confident_true_inpainting_inpainting_3832797_1984-01-19_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\04_confident_true_insight_insight_3832797_1984-01-19_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\05_confident_true_text2img_text2img_12318533_1982-06-16_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\06_strong_false_negative_insight_18249151_1970-12-16_2004.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\07_strong_false_negative_insight_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\08_borderline_text2img_2103427_1986-06-19_2012.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\09_borderline_insight_43568472_1968-07-26_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\10_expanded_context_wiki_34379104_1950-12-07_2013.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_resnet18_aug\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2d_simplecnn_aug",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.65625,
|
|
||||||
"auc_roc": 0.7565104166666666,
|
|
||||||
"f1": 0.725,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
39,
|
|
||||||
9
|
|
||||||
],
|
|
||||||
[
|
|
||||||
57,
|
|
||||||
87
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\67\\779067_1968-10-08_2008.jpg",
|
|
||||||
"basename": "779067_1968-10-08_2008.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.10822787880897522,
|
|
||||||
"logit": -2.10897159576416,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.16694103181362152,
|
|
||||||
"logit": -1.6074637174606323,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8760586977005005,
|
|
||||||
"logit": 1.955625057220459,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7972180843353271,
|
|
||||||
"logit": 1.36899733543396,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\text2img\\59\\40098159_1902-01-15_1988.jpg",
|
|
||||||
"basename": "40098159_1902-01-15_1988.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.8695456981658936,
|
|
||||||
"logit": 1.8969472646713257,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\06\\25476706_1975-03-27_2009.jpg",
|
|
||||||
"basename": "25476706_1975-03-27_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7830404043197632,
|
|
||||||
"logit": 1.283473253250122,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\wiki\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.764281153678894,
|
|
||||||
"logit": 1.1762961149215698,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\95\\27859495_1952-08-08_2014.jpg",
|
|
||||||
"basename": "27859495_1952-08-08_2014.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.19983962178230286,
|
|
||||||
"logit": -1.3872970342636108,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\insight\\56\\14853156_1981-05-28_2009.jpg",
|
|
||||||
"basename": "14853156_1981-05-28_2009.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.20330362021923065,
|
|
||||||
"logit": -1.3657732009887695,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\data\\inpainting\\51\\18249151_1970-12-16_2004.jpg",
|
|
||||||
"basename": "18249151_1970-12-16_2004.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5001071095466614,
|
|
||||||
"logit": 0.00042841583490371704,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\01_confident_true_wiki_wiki_779067_1968-10-08_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\02_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\03_confident_true_inpainting_inpainting_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\04_confident_true_insight_insight_34379104_1950-12-07_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\05_confident_true_text2img_text2img_40098159_1902-01-15_1988.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\06_strong_false_positive_wiki_25476706_1975-03-27_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\07_strong_false_positive_wiki_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\08_strong_false_negative_insight_27859495_1952-08-08_2014.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\09_strong_false_negative_insight_14853156_1981-05-28_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\10_borderline_inpainting_18249151_1970-12-16_2004.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2d_simplecnn_aug\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2e_resnet18_facecrop_aug",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.9322916666666666,
|
|
||||||
"auc_roc": 0.9910300925925927,
|
|
||||||
"f1": 0.9530685920577617,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
47,
|
|
||||||
1
|
|
||||||
],
|
|
||||||
[
|
|
||||||
12,
|
|
||||||
132
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\13\\25021613_1996-10-24_2012.jpg",
|
|
||||||
"basename": "25021613_1996-10-24_2012.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 4.415973648974614e-07,
|
|
||||||
"logit": -14.632866859436035,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\56\\14853156_1981-05-28_2009.jpg",
|
|
||||||
"basename": "14853156_1981-05-28_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 1.1105105386377545e-06,
|
|
||||||
"logit": -13.710689544677734,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\44\\3720944_1920-12-28_1963.jpg",
|
|
||||||
"basename": "3720944_1920-12-28_1963.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 18.160795211791992,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\79\\25469179_1988-04-23_2008.jpg",
|
|
||||||
"basename": "25469179_1988-04-23_2008.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 24.2982120513916,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\15\\16719015_1940-10-10_1964.jpg",
|
|
||||||
"basename": "16719015_1940-10-10_1964.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 1.0,
|
|
||||||
"logit": 22.898853302001953,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\57\\31543457_1910-02-06_1970.jpg",
|
|
||||||
"basename": "31543457_1910-02-06_1970.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6879352927207947,
|
|
||||||
"logit": 0.7904843091964722,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\37\\3272137_1943-04-22_2010.jpg",
|
|
||||||
"basename": "3272137_1943-04-22_2010.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.000333707983372733,
|
|
||||||
"logit": -8.004910469055176,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\46\\6385546_1980-05-23_2013.jpg",
|
|
||||||
"basename": "6385546_1980-05-23_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.010300145484507084,
|
|
||||||
"logit": -4.565243721008301,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\49\\25392649_1991-05-29_2013.jpg",
|
|
||||||
"basename": "25392649_1991-05-29_2013.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4931047856807709,
|
|
||||||
"logit": -0.027582719922065735,
|
|
||||||
"selection": "borderline"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\39\\451239_1954-05-11_2007.jpg",
|
|
||||||
"basename": "451239_1954-05-11_2007.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.5096341967582703,
|
|
||||||
"logit": 0.03854157030582428,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\01_confident_true_wiki_wiki_25021613_1996-10-24_2012.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\02_confident_true_wiki_wiki_14853156_1981-05-28_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\03_confident_true_inpainting_inpainting_3720944_1920-12-28_1963.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\04_confident_true_insight_insight_25469179_1988-04-23_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\05_confident_true_text2img_text2img_16719015_1940-10-10_1964.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\06_strong_false_positive_wiki_31543457_1910-02-06_1970.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\07_strong_false_negative_text2img_3272137_1943-04-22_2010.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\08_strong_false_negative_insight_6385546_1980-05-23_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\09_borderline_text2img_25392649_1991-05-29_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\10_borderline_text2img_451239_1954-05-11_2007.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_resnet18_facecrop_aug\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,138 +0,0 @@
|
|||||||
{
|
|
||||||
"run": "p2e_simplecnn_facecrop_aug",
|
|
||||||
"fold": 0,
|
|
||||||
"n_candidates": 192,
|
|
||||||
"candidate_metrics": {
|
|
||||||
"accuracy": 0.6510416666666666,
|
|
||||||
"auc_roc": 0.7358217592592592,
|
|
||||||
"f1": 0.7196652719665272,
|
|
||||||
"confusion_matrix": [
|
|
||||||
[
|
|
||||||
39,
|
|
||||||
9
|
|
||||||
],
|
|
||||||
[
|
|
||||||
58,
|
|
||||||
86
|
|
||||||
]
|
|
||||||
]
|
|
||||||
},
|
|
||||||
"heldout_source": null,
|
|
||||||
"selected": [
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\31\\2346431_1980-06-11_2007.jpg",
|
|
||||||
"basename": "2346431_1980-06-11_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.1969025731086731,
|
|
||||||
"logit": -1.4057669639587402,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\93\\28699293_1957-04-26_2009.jpg",
|
|
||||||
"basename": "28699293_1957-04-26_2009.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.21053381264209747,
|
|
||||||
"logit": -1.3217107057571411,
|
|
||||||
"selection": "confident_true_wiki"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\52\\9143052_1931-01-21_1991.jpg",
|
|
||||||
"basename": "9143052_1931-01-21_1991.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7768715620040894,
|
|
||||||
"logit": 1.2475274801254272,
|
|
||||||
"selection": "confident_true_inpainting"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\insight\\04\\34379104_1950-12-07_2013.jpg",
|
|
||||||
"basename": "34379104_1950-12-07_2013.jpg",
|
|
||||||
"source": "insight",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.7978224158287048,
|
|
||||||
"logit": 1.372739553451538,
|
|
||||||
"selection": "confident_true_insight"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\text2img\\46\\6385546_1980-05-23_2013.jpg",
|
|
||||||
"basename": "6385546_1980-05-23_2013.jpg",
|
|
||||||
"source": "text2img",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.9004967212677002,
|
|
||||||
"logit": 2.202756643295288,
|
|
||||||
"selection": "confident_true_text2img"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\13\\25021613_1996-10-24_2012.jpg",
|
|
||||||
"basename": "25021613_1996-10-24_2012.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6920362710952759,
|
|
||||||
"logit": 0.8096563220024109,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\wiki\\29\\19558629_1985-02-28_2007.jpg",
|
|
||||||
"basename": "19558629_1985-02-28_2007.jpg",
|
|
||||||
"source": "wiki",
|
|
||||||
"label": 0,
|
|
||||||
"pred": 1,
|
|
||||||
"prob_fake": 0.6103906035423279,
|
|
||||||
"logit": 0.44895440340042114,
|
|
||||||
"selection": "strong_false_positive"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\67\\779067_1968-10-08_2008.jpg",
|
|
||||||
"basename": "779067_1968-10-08_2008.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.22937096655368805,
|
|
||||||
"logit": -1.2118664979934692,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\93\\28699293_1957-04-26_2009.jpg",
|
|
||||||
"basename": "28699293_1957-04-26_2009.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.24612900614738464,
|
|
||||||
"logit": -1.1193655729293823,
|
|
||||||
"selection": "strong_false_negative"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\cropped\\classifier\\inpainting\\25\\1932725_1949-01-25_1979.jpg",
|
|
||||||
"basename": "1932725_1949-01-25_1979.jpg",
|
|
||||||
"source": "inpainting",
|
|
||||||
"label": 1,
|
|
||||||
"pred": 0,
|
|
||||||
"prob_fake": 0.4987809658050537,
|
|
||||||
"logit": -0.004876129329204559,
|
|
||||||
"selection": "borderline"
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"image_paths": [
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\01_confident_true_wiki_wiki_2346431_1980-06-11_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\02_confident_true_wiki_wiki_28699293_1957-04-26_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\03_confident_true_inpainting_inpainting_9143052_1931-01-21_1991.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\04_confident_true_insight_insight_34379104_1950-12-07_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\05_confident_true_text2img_text2img_6385546_1980-05-23_2013.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\06_strong_false_positive_wiki_25021613_1996-10-24_2012.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\07_strong_false_positive_wiki_19558629_1985-02-28_2007.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\08_strong_false_negative_inpainting_779067_1968-10-08_2008.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\09_strong_false_negative_inpainting_28699293_1957-04-26_2009.png",
|
|
||||||
"c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\10_borderline_inpainting_1932725_1949-01-25_1979.png"
|
|
||||||
],
|
|
||||||
"panel_path": "c:\\Users\\diogo\\Documents\\MIA_UP\\2_Semestre\\DRL\\DRL_2\\DRL_PROJ\\classifier\\outputs\\gradcam\\p2e_simplecnn_facecrop_aug\\panel.png",
|
|
||||||
"expanded_panel_path": null,
|
|
||||||
"fine_panel_path": null
|
|
||||||
}
|
|
||||||
@@ -1,37 +0,0 @@
|
|||||||
{
|
|
||||||
"phase": "phase2",
|
|
||||||
"best_existing_run": "p2c_resnet18_facecrop",
|
|
||||||
"best_existing_auc": 0.9755,
|
|
||||||
"decisions": [
|
|
||||||
{
|
|
||||||
"choice": "input size",
|
|
||||||
"decision": "224x224",
|
|
||||||
"evidence": "ResNet18 improves from AUC 0.9366 to 0.9660.",
|
|
||||||
"confidence": "high"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"choice": "face crop",
|
|
||||||
"decision": "enable",
|
|
||||||
"evidence": "Best Phase 2 run is p2c_resnet18_facecrop with AUC 0.9755.",
|
|
||||||
"confidence": "medium-high"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"choice": "augmentation",
|
|
||||||
"decision": "disable for current 20% setting",
|
|
||||||
"evidence": "p2e_resnet18_facecrop_aug reaches AUC 0.9737, below facecrop-only 0.9755; SimpleCNN drops sharply.",
|
|
||||||
"confidence": "low"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"choice": "normalization",
|
|
||||||
"decision": "ImageNet/default",
|
|
||||||
"evidence": "real_norm is only +0.0018 AUC and is less aligned with pretrained ImageNet weights.",
|
|
||||||
"confidence": "medium"
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"choice": "source generalization",
|
|
||||||
"decision": "report as limitation and diagnostic target",
|
|
||||||
"evidence": "Holding out text2img and insight drops pairwise AUC to 0.7595 and 0.8421.",
|
|
||||||
"confidence": "high"
|
|
||||||
}
|
|
||||||
]
|
|
||||||
}
|
|
||||||
@@ -1,104 +0,0 @@
|
|||||||
{
|
|
||||||
"phase": "phase4",
|
|
||||||
"best_auc_run": "p4_convnext_tiny_100pct",
|
|
||||||
"best_auc": 0.9954,
|
|
||||||
"selected_detector_run": "p4_convnext_tiny_100pct",
|
|
||||||
"selected_detector_reason": "Lowest false positive rate on real images and best balanced accuracy at 100% data.",
|
|
||||||
"notes": [
|
|
||||||
"20% rows come from the matching Phase 3 runs and act as scaling anchors.",
|
|
||||||
"50% and 100% rows come from Phase 4 logs.",
|
|
||||||
"ConvNeXt-Tiny is the AUC winner and also the practical detector when false positives on real images matter most."
|
|
||||||
],
|
|
||||||
"summary_rows": [
|
|
||||||
{
|
|
||||||
"model": "ResNet50",
|
|
||||||
"scale_pct": 20,
|
|
||||||
"run": "p3_resnet50",
|
|
||||||
"auc": 0.9857046296296297,
|
|
||||||
"accuracy": 0.9532083333333334,
|
|
||||||
"f1": 0.9687705256330055,
|
|
||||||
"balanced_accuracy": 0.9385277777777778,
|
|
||||||
"macro_f1": 0.9377251247311893
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "ResNet50",
|
|
||||||
"scale_pct": 50,
|
|
||||||
"run": "p4_resnet50_50pct",
|
|
||||||
"auc": 0.9921022333333335,
|
|
||||||
"accuracy": 0.9607166666666667,
|
|
||||||
"f1": 0.973649056957159,
|
|
||||||
"balanced_accuracy": 0.9536555555555555,
|
|
||||||
"macro_f1": 0.9482411270477236
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "ResNet50",
|
|
||||||
"scale_pct": 100,
|
|
||||||
"run": "p4_resnet50_100pct",
|
|
||||||
"auc": 0.9949671268518518,
|
|
||||||
"accuracy": 0.9692166666666667,
|
|
||||||
"f1": 0.9793757797895071,
|
|
||||||
"balanced_accuracy": 0.9638111111111111,
|
|
||||||
"macro_f1": 0.9593474615494658
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "EfficientNet-B0",
|
|
||||||
"scale_pct": 20,
|
|
||||||
"run": "p3_efficientnet_b0",
|
|
||||||
"auc": 0.9844867592592592,
|
|
||||||
"accuracy": 0.9450000000000001,
|
|
||||||
"f1": 0.9628492625462333,
|
|
||||||
"balanced_accuracy": 0.9397222222222222,
|
|
||||||
"macro_f1": 0.9284971237471479
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "EfficientNet-B0",
|
|
||||||
"scale_pct": 50,
|
|
||||||
"run": "p4_efficientnet_b0_50pct",
|
|
||||||
"auc": 0.9911186185185186,
|
|
||||||
"accuracy": 0.9576499999999999,
|
|
||||||
"f1": 0.9714437706354593,
|
|
||||||
"balanced_accuracy": 0.9541444444444445,
|
|
||||||
"macro_f1": 0.9446884831225546
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "EfficientNet-B0",
|
|
||||||
"scale_pct": 100,
|
|
||||||
"run": "p4_efficientnet_b0_100pct",
|
|
||||||
"auc": 0.9949471324074075,
|
|
||||||
"accuracy": 0.96835,
|
|
||||||
"f1": 0.9787227370502695,
|
|
||||||
"balanced_accuracy": 0.9656777777777779,
|
|
||||||
"macro_f1": 0.9584469550394183
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "ConvNeXt-Tiny",
|
|
||||||
"scale_pct": 20,
|
|
||||||
"run": "p3_convnext_tiny",
|
|
||||||
"auc": 0.9867751388888889,
|
|
||||||
"accuracy": 0.9473333333333332,
|
|
||||||
"f1": 0.9650225390458838,
|
|
||||||
"balanced_accuracy": 0.9261666666666667,
|
|
||||||
"macro_f1": 0.9292641404681354
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "ConvNeXt-Tiny",
|
|
||||||
"scale_pct": 50,
|
|
||||||
"run": "p4_convnext_tiny_50pct",
|
|
||||||
"auc": 0.9926127555555556,
|
|
||||||
"accuracy": 0.96065,
|
|
||||||
"f1": 0.9735230102651047,
|
|
||||||
"balanced_accuracy": 0.9562111111111111,
|
|
||||||
"macro_f1": 0.9484169329127863
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"model": "ConvNeXt-Tiny",
|
|
||||||
"scale_pct": 100,
|
|
||||||
"run": "p4_convnext_tiny_100pct",
|
|
||||||
"auc": 0.9953774120370371,
|
|
||||||
"accuracy": 0.9679166666666668,
|
|
||||||
"f1": 0.9783980914742021,
|
|
||||||
"balanced_accuracy": 0.9662444444444445,
|
|
||||||
"macro_f1": 0.9579703600254157
|
|
||||||
}
|
|
||||||
]
|
|
||||||
}
|
|
||||||
@@ -0,0 +1,90 @@
|
|||||||
|
"""
|
||||||
|
End-to-end smoke test: synthetic DFF layout -> short CV train -> inference.
|
||||||
|
|
||||||
|
Run from the classifier package root:
|
||||||
|
|
||||||
|
cd classifier && conda activate drl && python -m unittest tests.smoke_test -v
|
||||||
|
"""
|
||||||
|
import io
|
||||||
|
import subprocess
|
||||||
|
import sys
|
||||||
|
import unittest
|
||||||
|
from pathlib import Path
|
||||||
|
|
||||||
|
from PIL import Image
|
||||||
|
|
||||||
|
_CLASSIFIER_ROOT = Path(__file__).resolve().parents[1]
|
||||||
|
_SMOKE_CFG = _CLASSIFIER_ROOT / "configs" / "smoke" / "smoke.json"
|
||||||
|
|
||||||
|
|
||||||
|
def _write_synthetic_dff(data_root: Path, *, n_identities: int = 16) -> None:
|
||||||
|
"""Minimal DeepFakeFace-style tree: wiki + three fake sources, shared basenames per identity."""
|
||||||
|
sources = ("wiki", "inpainting", "text2img", "insight")
|
||||||
|
for i in range(n_identities):
|
||||||
|
stem = f"id{i:03d}"
|
||||||
|
for src in sources:
|
||||||
|
d = data_root / src / f"person_{i:03d}"
|
||||||
|
d.mkdir(parents=True, exist_ok=True)
|
||||||
|
buf = io.BytesIO()
|
||||||
|
Image.new(
|
||||||
|
"RGB",
|
||||||
|
(96, 96),
|
||||||
|
color=(min(20 + i * 11, 255), 80, 120 if src == "wiki" else 40),
|
||||||
|
).save(buf, format="JPEG")
|
||||||
|
# Same filename stem across all sources → one CV group per identity (matches DFF).
|
||||||
|
(d / f"{stem}.jpg").write_bytes(buf.getvalue())
|
||||||
|
|
||||||
|
|
||||||
|
class SmokeTrainInferTests(unittest.TestCase):
|
||||||
|
def test_local_smoke_train_then_inference(self):
|
||||||
|
import tempfile
|
||||||
|
|
||||||
|
with tempfile.TemporaryDirectory() as td:
|
||||||
|
tmp = Path(td)
|
||||||
|
data_dir = tmp / "data"
|
||||||
|
out_root = tmp / "outputs"
|
||||||
|
_write_synthetic_dff(data_dir)
|
||||||
|
|
||||||
|
sys.path.insert(0, str(_CLASSIFIER_ROOT))
|
||||||
|
import run as train_run
|
||||||
|
|
||||||
|
train_run.main(
|
||||||
|
str(_SMOKE_CFG),
|
||||||
|
data_dir_override=str(data_dir),
|
||||||
|
output_root=str(out_root),
|
||||||
|
use_gpu=False,
|
||||||
|
)
|
||||||
|
|
||||||
|
models_dir = out_root / "models"
|
||||||
|
ck_fold0 = models_dir / "smoke_fold0_best.pt"
|
||||||
|
if not ck_fold0.is_file():
|
||||||
|
ck_fold0 = models_dir / "smoke_fold0_final.pt"
|
||||||
|
self.assertTrue(
|
||||||
|
ck_fold0.is_file(),
|
||||||
|
f"Expected fold-0 checkpoint under {models_dir}",
|
||||||
|
)
|
||||||
|
|
||||||
|
probe = tmp / "probe.jpg"
|
||||||
|
probe.write_bytes((data_dir / "wiki" / "person_000" / "id000.jpg").read_bytes())
|
||||||
|
|
||||||
|
cmd = [
|
||||||
|
sys.executable,
|
||||||
|
str(_CLASSIFIER_ROOT / "tools" / "inference.py"),
|
||||||
|
str(probe),
|
||||||
|
str(_SMOKE_CFG),
|
||||||
|
"--checkpoint",
|
||||||
|
str(ck_fold0),
|
||||||
|
]
|
||||||
|
proc = subprocess.run(
|
||||||
|
cmd,
|
||||||
|
cwd=str(_CLASSIFIER_ROOT),
|
||||||
|
capture_output=True,
|
||||||
|
text=True,
|
||||||
|
timeout=300,
|
||||||
|
)
|
||||||
|
self.assertEqual(proc.returncode, 0, proc.stderr + proc.stdout)
|
||||||
|
self.assertIn("P(fake)", proc.stdout)
|
||||||
|
|
||||||
|
|
||||||
|
if __name__ == "__main__":
|
||||||
|
unittest.main()
|
||||||
@@ -18,8 +18,9 @@ def parse_args():
|
|||||||
|
|
||||||
|
|
||||||
def iter_config_paths(config_root: Path):
|
def iter_config_paths(config_root: Path):
|
||||||
for sub in ("phase1", "phase2"):
|
for sub in sorted(config_root.iterdir()):
|
||||||
yield from sorted((config_root / sub).glob("*.json"))
|
if sub.is_dir() and sub.name not in ("smoke",):
|
||||||
|
yield from sorted(sub.glob("*.json"))
|
||||||
|
|
||||||
|
|
||||||
def main():
|
def main():
|
||||||
|
|||||||
@@ -2,8 +2,8 @@
|
|||||||
Download the DeepFakeFace dataset from HuggingFace and extract it.
|
Download the DeepFakeFace dataset from HuggingFace and extract it.
|
||||||
|
|
||||||
Usage:
|
Usage:
|
||||||
python tools/download_data.py
|
python tools/fetch_ds.py
|
||||||
python tools/download_data.py --data-dir /mnt/data/DFF
|
python tools/fetch_ds.py --data-dir /mnt/data/DFF
|
||||||
"""
|
"""
|
||||||
import argparse
|
import argparse
|
||||||
import zipfile
|
import zipfile
|
||||||
|
|||||||
@@ -19,9 +19,27 @@ from PIL import Image
|
|||||||
|
|
||||||
from src.models import get_model, load_checkpoint
|
from src.models import get_model, load_checkpoint
|
||||||
from src.preprocessing import get_transforms
|
from src.preprocessing import get_transforms
|
||||||
|
from src.utils import load_config
|
||||||
|
|
||||||
|
|
||||||
# Defaults checkpoint to outputs/models/{run_name}_best.pt when not supplied
|
def _default_checkpoint(cfg: dict, checkpoint_path: Path | None) -> Path:
|
||||||
|
"""Resolve checkpoint: explicit path, single-fold `*_best.pt`, or CV `*_fold{k}_best.pt` / `*_final.pt`."""
|
||||||
|
run_name = cfg["run_name"]
|
||||||
|
models_dir = ROOT / "outputs" / "models"
|
||||||
|
if checkpoint_path is not None:
|
||||||
|
return Path(checkpoint_path)
|
||||||
|
candidates: list[Path] = [models_dir / f"{run_name}_best.pt"]
|
||||||
|
for k in range(32):
|
||||||
|
candidates.append(models_dir / f"{run_name}_fold{k}_best.pt")
|
||||||
|
for k in range(32):
|
||||||
|
candidates.append(models_dir / f"{run_name}_fold{k}_final.pt")
|
||||||
|
for p in candidates:
|
||||||
|
if p.is_file():
|
||||||
|
return p
|
||||||
|
return models_dir / f"{run_name}_best.pt"
|
||||||
|
|
||||||
|
|
||||||
|
# Defaults checkpoint under outputs/models/ (single-run or CV best/final).
|
||||||
def predict(image_path, config_path, checkpoint_path=None):
|
def predict(image_path, config_path, checkpoint_path=None):
|
||||||
image_path = Path(image_path)
|
image_path = Path(image_path)
|
||||||
config_path = Path(config_path)
|
config_path = Path(config_path)
|
||||||
@@ -35,10 +53,9 @@ def predict(image_path, config_path, checkpoint_path=None):
|
|||||||
sys.exit(1)
|
sys.exit(1)
|
||||||
|
|
||||||
try:
|
try:
|
||||||
with open(config_path) as f:
|
cfg = load_config(str(config_path))
|
||||||
cfg = json.load(f)
|
except (json.JSONDecodeError, OSError, ValueError) as e:
|
||||||
except json.JSONDecodeError as e:
|
print(f"Error: Failed to load config: {e}")
|
||||||
print(f"Error: Invalid JSON in config: {e}")
|
|
||||||
sys.exit(1)
|
sys.exit(1)
|
||||||
|
|
||||||
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
||||||
@@ -50,10 +67,7 @@ def predict(image_path, config_path, checkpoint_path=None):
|
|||||||
print(f"Error: Failed to build model: {e}")
|
print(f"Error: Failed to build model: {e}")
|
||||||
sys.exit(1)
|
sys.exit(1)
|
||||||
|
|
||||||
if checkpoint_path is None:
|
checkpoint_path = _default_checkpoint(cfg, Path(checkpoint_path) if checkpoint_path else None)
|
||||||
checkpoint_path = ROOT / "outputs" / "models" / f"{cfg['run_name']}_best.pt"
|
|
||||||
else:
|
|
||||||
checkpoint_path = Path(checkpoint_path)
|
|
||||||
|
|
||||||
if not checkpoint_path.exists():
|
if not checkpoint_path.exists():
|
||||||
print(f"Error: Checkpoint not found: {checkpoint_path}")
|
print(f"Error: Checkpoint not found: {checkpoint_path}")
|
||||||
|
|||||||
@@ -1,642 +0,0 @@
|
|||||||
# Deepfake Detection Classifier - Implementation Plan
|
|
||||||
|
|
||||||
## Overview
|
|
||||||
This document provides a comprehensive implementation plan for refactoring the deepfake detection classifier project. Each task includes a checkbox to track completion.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 0: Pre-Implementation Setup
|
|
||||||
|
|
||||||
### Infrastructure and Configuration
|
|
||||||
- [x] Create `classifier/configs/shared.json` with shared parameters:
|
|
||||||
- seed: 42
|
|
||||||
- val_ratio: 0.1
|
|
||||||
- test_ratio: 0.1
|
|
||||||
- batch_size: 32
|
|
||||||
- optimizer: {type: "adamw", lr: 1e-4, weight_decay: 1e-4}
|
|
||||||
- scheduler: {type: "cosine_annealing", T_max: 15}
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
- num_workers: 4
|
|
||||||
- cv_folds: 5
|
|
||||||
- data_dir: "data"
|
|
||||||
- face_crop_margin: 0.6
|
|
||||||
|
|
||||||
- [x] Implement config loading/merging so experiment configs inherit `shared.json` defaults and override only the variables under test
|
|
||||||
- [x] Resolve shared nested fields such as `optimizer.lr`, `optimizer.weight_decay`, and `scheduler.T_max` into the training arguments used by the runner
|
|
||||||
- [x] Update existing configs to reference `shared.json` or otherwise document which shared defaults they intentionally override
|
|
||||||
- [x] Define one CV protocol for all phases:
|
|
||||||
- outer fold: held-out test fold
|
|
||||||
- inner validation split: group-aware split from the remaining training folds for early stopping/model selection
|
|
||||||
- final reported metrics: aggregate held-out test-fold results across the 5 outer folds
|
|
||||||
|
|
||||||
### Data Preparation
|
|
||||||
- [x] Verify dataset structure and integrity
|
|
||||||
- [x] Check that real and fake images are properly organized by source
|
|
||||||
- [x] Verify no data leakage between train/val/test splits or CV folds (group-aware by basename)
|
|
||||||
|
|
||||||
### Cleanup
|
|
||||||
- [x] Remove `classifier/tools/ensemble.py` (not part of reorganization plan, conflicts with explainability goals)
|
|
||||||
- [x] Remove robustness evaluation from `classifier/tools/analyze.py` (lines 51-104, 82-104, 144) - not part of experimental plan
|
|
||||||
- [x] Remove any unused or obsolete config files from previous experiments (see detailed list below)
|
|
||||||
- [X] Clean up old output directories if needed (keep important results for reference)
|
|
||||||
|
|
||||||
#### Config Files to Remove (39 total)
|
|
||||||
|
|
||||||
**Root configs (6):**
|
|
||||||
- [x] `classifier/configs/resnet18_quick.json`
|
|
||||||
- [x] `classifier/configs/resnet18.json`
|
|
||||||
- [x] `classifier/configs/simple_cnn_large.json`
|
|
||||||
- [x] `classifier/configs/simple_cnn_micro.json`
|
|
||||||
- [x] `classifier/configs/simple_cnn_small.json`
|
|
||||||
- [x] `classifier/configs/simple_cnn.json`
|
|
||||||
|
|
||||||
**Phase 1 old configs (7):**
|
|
||||||
- [x] `classifier/configs/phase1/p1_cnn_base.json` (uses lr=1e-3, epochs=20 - should be 1e-4, 15)
|
|
||||||
- [x] `classifier/configs/phase1/p1_cnn_aug.json`
|
|
||||||
- [x] `classifier/configs/phase1/p1_resnet18_base.json` (duplicate of new baseline)
|
|
||||||
- [x] `classifier/configs/phase1/p1_resnet18_aug.json`
|
|
||||||
- [x] `classifier/configs/phase1/holdout/` (entire directory - 6 configs, source holdout not in new plan)
|
|
||||||
|
|
||||||
**Phase 2 old configs (7):**
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet18_224.json` (should be p2a_resnet18_224.json)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet18_facecrop.json` (should be p2b_resnet18_facecrop.json)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet18_frozen.json` (frozen backbone not in new plan)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet34_224.json` (ResNet34 should be in Phase 3)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet34.json` (ResNet34 should be in Phase 3)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet50_frozen.json` (ResNet50 should be in Phase 3)
|
|
||||||
- [x] `classifier/configs/phase2/p2_resnet50.json` (ResNet50 should be in Phase 3)
|
|
||||||
|
|
||||||
**Phase 3 old configs (4):**
|
|
||||||
- [x] `classifier/configs/phase3/p3_efficientnet_b2.json` (EfficientNet-B2 not in new plan, only B0)
|
|
||||||
- [x] `classifier/configs/phase3/p3_resnet18_facecrop_full.json` (ResNet18 full dataset should be Phase 4)
|
|
||||||
- [x] `classifier/configs/phase3/p3_resnet18_freqaug.json` (frequency augmentation not in new plan)
|
|
||||||
- [x] `classifier/configs/phase3/p3_vit_b16.json` (ViT not in new plan, replaced with ConvNeXt/MobileNet)
|
|
||||||
- Note: `p3_efficientnet_b0.json` - REMOVED (will be recreated after Phase2 with correct settings)
|
|
||||||
|
|
||||||
**Source holdout (6):**
|
|
||||||
- [x] `classifier/configs/source_holdout/` (entire directory - 6 configs, source holdout not in new plan)
|
|
||||||
|
|
||||||
**Ablation (3):**
|
|
||||||
- [x] `classifier/configs/ablation/` (entire directory - 3 configs, ablation studies not in new plan)
|
|
||||||
|
|
||||||
**Configs to KEEP (3):**
|
|
||||||
- ✅ `classifier/configs/shared.json`
|
|
||||||
- ✅ `classifier/configs/phase1/p1_simplecnn_baseline.json`
|
|
||||||
- ✅ `classifier/configs/phase1/p1_resnet18_baseline.json`
|
|
||||||
|
|
||||||
**Phase 2 alias configs removed (8):**
|
|
||||||
- [x] `classifier/configs/phase2/p2b_resnet18_128.json` (alias for p1_resnet18_baseline)
|
|
||||||
- [x] `classifier/configs/phase2/p2b_simplecnn_128.json` (alias for p1_simplecnn_baseline)
|
|
||||||
- [x] `classifier/configs/phase2/p2c_resnet18_nofacecrop.json` (alias for p2b_resnet18_224)
|
|
||||||
- [x] `classifier/configs/phase2/p2c_simplecnn_nofacecrop.json` (alias for p2b_simplecnn_224)
|
|
||||||
- [x] `classifier/configs/phase2/p2d_resnet18_noaug.json` (alias for p2b_resnet18_224)
|
|
||||||
- [x] `classifier/configs/phase2/p2d_simplecnn_noaug.json` (alias for p2b_simplecnn_224)
|
|
||||||
- [x] `classifier/configs/phase2/p2e_resnet18_facecrop_only.json` (alias for p2c_resnet18_facecrop)
|
|
||||||
- [x] `classifier/configs/phase2/p2e_simplecnn_facecrop_only.json` (alias for p2c_simplecnn_facecrop)
|
|
||||||
|
|
||||||
Note: Comparison pairs (baseline vs treatment) are defined in the analysis notebook as a mapping dict, not as separate config files.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 1: Architecture Baseline
|
|
||||||
|
|
||||||
### 1.1 Experiment Configs
|
|
||||||
- [x] Create `classifier/configs/phase1/p1_simplecnn_baseline.json`
|
|
||||||
- backbone: simple_cnn
|
|
||||||
- cnn_preset: medium
|
|
||||||
- dropout: 0.0
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4 (consistent with ResNet)
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 128
|
|
||||||
- data_dir: data
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
- subsample: 0.2
|
|
||||||
- face_crop: false
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase1/p1_resnet18_baseline.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 128
|
|
||||||
- data_dir: data
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
- subsample: 0.2
|
|
||||||
- face_crop: false
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
|
|
||||||
### 1.2 Code Updates
|
|
||||||
- [x] Implement 5-fold stratified group cross-validation by basename in training pipeline
|
|
||||||
- [x] Update `classifier/src/training/trainer.py` to support CV
|
|
||||||
- [x] Update `classifier/src/evaluation/evaluate.py` to support CV
|
|
||||||
- [x] Ensure all metrics report mean ± std and confidence intervals across folds
|
|
||||||
|
|
||||||
### 1.3 Training
|
|
||||||
- [x] Train SimpleCNN with 5-fold stratified group CV (via pipeline: `python -m pipeline run classifier/configs/phase1/p1_simplecnn_baseline.json`)
|
|
||||||
- [x] Train ResNet18 with 5-fold stratified group CV (via pipeline: `python -m pipeline run classifier/configs/phase1/p1_resnet18_baseline.json`)
|
|
||||||
- [x] Save all checkpoints and metrics (pipeline automatically fetches outputs to classifier/outputs/)
|
|
||||||
|
|
||||||
### 1.4 Analysis
|
|
||||||
- [x] Use `classifier/notebooks/03_phase1_analysis.ipynb` for Phase 1 analysis
|
|
||||||
- [x] Compare SimpleCNN vs ResNet18 performance
|
|
||||||
- [x] Overall metrics (AUC, Accuracy, F1) with mean ± std and confidence intervals
|
|
||||||
- [x] Per-source metrics (text2img, inpainting, insight)
|
|
||||||
- [x] Train/val/test performance curves
|
|
||||||
- [x] Confusion matrices
|
|
||||||
- [x] Statistical significance testing
|
|
||||||
- [x] Generate Grad-CAM visualizations (10-20 images per model)
|
|
||||||
- [x] Document conclusions: Which baseline is better and why
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 2: Preprocessing Impact
|
|
||||||
|
|
||||||
### 2.1 Shortcut Analysis (2A)
|
|
||||||
- [x] Create `classifier/configs/phase2/p2a_t1_original.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- augment: false
|
|
||||||
- normalization: imagenet
|
|
||||||
- data_dir: data
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2a_t2_real_norm.json`
|
|
||||||
- extends: p2a_t1_original.json
|
|
||||||
- normalization: real_norm
|
|
||||||
- **Normalization**: Calculate mean/std from real training images only within each fold
|
|
||||||
|
|
||||||
- [x] Geometry diagnostic was explored and then removed from the codebase (`src/evaluation/geometry.py` no longer exists):
|
|
||||||
- Current pipeline always square-crops before resize, reducing rectangle-vs-square shortcut risk.
|
|
||||||
- Shortcut analysis now relies on normalization and held-out-source evidence artifacts.
|
|
||||||
|
|
||||||
- [ ] Train the 2 shortcut configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare results:
|
|
||||||
- Standard vs matched-geometry eval for `p2a_t1_original` (letterboxing impact)
|
|
||||||
- `p2a_t1_original` vs `p2a_t2_real_norm` (color distribution shortcut)
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2a_t3_holdout_text2img.json`
|
|
||||||
- extends: p2a_t1_original.json
|
|
||||||
- train_sources: ["wiki", "inpainting", "insight"]
|
|
||||||
- eval_sources: ["wiki", "inpainting", "insight", "text2img"]
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2a_t3_holdout_inpainting.json`
|
|
||||||
- extends: p2a_t1_original.json
|
|
||||||
- train_sources: ["wiki", "text2img", "insight"]
|
|
||||||
- eval_sources: ["wiki", "text2img", "insight", "inpainting"]
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2a_t3_holdout_insight.json`
|
|
||||||
- extends: p2a_t1_original.json
|
|
||||||
- train_sources: ["wiki", "text2img", "inpainting"]
|
|
||||||
- eval_sources: ["wiki", "text2img", "inpainting", "insight"]
|
|
||||||
|
|
||||||
- [ ] Train the 3 source holdout configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare held-out source performance vs in-source performance:
|
|
||||||
- Calculate AUC for held-out source (text2img, inpainting, insight)
|
|
||||||
- Compute Δ (in-source AUC - held-out AUC)
|
|
||||||
- If Δ > 0.05-0.10, model is learning source-specific features
|
|
||||||
|
|
||||||
### 2.2 Resolution Impact (2B)
|
|
||||||
- [x] Create `classifier/configs/phase2/p2b_simplecnn_224.json`
|
|
||||||
- backbone: simple_cnn
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
- data_dir: data
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2b_resnet18_224.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
- data_dir: data
|
|
||||||
|
|
||||||
- [ ] Train both 224 configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare 128×128 vs 224×224 for each model
|
|
||||||
- 128 baseline is `p1_*_baseline` (comparison mapping in notebook)
|
|
||||||
|
|
||||||
### 2.3 Facecrop Impact (2C)
|
|
||||||
- [x] Create `classifier/configs/phase2/p2c_simplecnn_facecrop.json`
|
|
||||||
- backbone: simple_cnn
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
- data_dir: cropped/classifier
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2c_resnet18_facecrop.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- augment: false
|
|
||||||
- seed: 42
|
|
||||||
- data_dir: cropped/classifier
|
|
||||||
|
|
||||||
- [ ] Train both facecrop configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare `p2b_resnet18_224` (no facecrop) vs `p2c_resnet18_facecrop` for each model
|
|
||||||
- No-facecrop baseline is `p2b_*_224` (comparison mapping in notebook)
|
|
||||||
|
|
||||||
### 2.4 Augmentation Impact (2D)
|
|
||||||
- [x] Create `classifier/configs/phase2/p2d_simplecnn_aug.json`
|
|
||||||
- backbone: simple_cnn
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- augment: {hflip_p: 0.5, rotation_degrees: 10, brightness: 0.2, contrast: 0.2, saturation: 0.1, hue: 0.02, grayscale_p: 0.1, blur_p: 0.1, erase_p: 0.2, noise_p: 0.3, noise_std: 0.04}
|
|
||||||
- data_dir: data
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2d_resnet18_aug.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- augment: {hflip_p: 0.5, rotation_degrees: 10, brightness: 0.2, contrast: 0.2, saturation: 0.1, hue: 0.02, grayscale_p: 0.1, blur_p: 0.1, erase_p: 0.2, noise_p: 0.3, noise_std: 0.04}
|
|
||||||
- data_dir: data
|
|
||||||
|
|
||||||
- [ ] Train both augmentation configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare `p2b_resnet18_224` (no aug) vs `p2d_resnet18_aug` for each model
|
|
||||||
- No-aug baseline is `p2b_*_224` (comparison mapping in notebook)
|
|
||||||
|
|
||||||
### 2.5 Augmentation + Facecrop (2E)
|
|
||||||
- [x] Create `classifier/configs/phase2/p2e_simplecnn_facecrop_aug.json`
|
|
||||||
- backbone: simple_cnn
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- augment: {hflip_p: 0.5, rotation_degrees: 10, brightness: 0.2, contrast: 0.2, saturation: 0.1, hue: 0.02, grayscale_p: 0.1, blur_p: 0.1, erase_p: 0.2, noise_p: 0.3, noise_std: 0.04}
|
|
||||||
- data_dir: cropped/classifier
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase2/p2e_resnet18_facecrop_aug.json`
|
|
||||||
- backbone: resnet18
|
|
||||||
- image_size: 224
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- augment: {hflip_p: 0.5, rotation_degrees: 10, brightness: 0.2, contrast: 0.2, saturation: 0.1, hue: 0.02, grayscale_p: 0.1, blur_p: 0.1, erase_p: 0.2, noise_p: 0.3, noise_std: 0.04}
|
|
||||||
- data_dir: cropped/classifier
|
|
||||||
|
|
||||||
- [ ] Train both facecrop+aug configs with 5-fold stratified group CV
|
|
||||||
- [ ] Compare `p2c_resnet18_facecrop` (facecrop only) vs `p2e_resnet18_facecrop_aug` for each model
|
|
||||||
- Facecrop-only baseline is `p2c_*_facecrop` (comparison mapping in notebook)
|
|
||||||
|
|
||||||
### 2.6 Phase 2 Analysis
|
|
||||||
- [ ] Use `classifier/notebooks/04_phase2_analysis.ipynb` for Phase 2 analysis
|
|
||||||
- [ ] For each experiment (2A-2E):
|
|
||||||
- [ ] Load 5-fold stratified group CV results (mean ± std and confidence intervals)
|
|
||||||
- [ ] Generate overall metrics (AUC, Accuracy, F1)
|
|
||||||
- [ ] Generate per-source metrics (text2img, inpainting, insight)
|
|
||||||
- [ ] Calculate train/val gap
|
|
||||||
- [ ] Calculate pairwise source AUC variance (wiki-vs-source AUC variance)
|
|
||||||
- [ ] Statistical significance testing vs baseline
|
|
||||||
- [ ] Generate comparison visualizations (bar charts, heatmaps)
|
|
||||||
- [ ] For 2C (Shortcut Analysis):
|
|
||||||
- [ ] Compare original-test vs alternative geometry evidence if reintroduced in a dedicated tool/notebook
|
|
||||||
- [ ] Compare ImageNet vs real-image-only normalization (color distribution shortcuts)
|
|
||||||
- [ ] Load source holdout results (3 configs)
|
|
||||||
- [ ] Calculate held-out source AUC vs in-source AUC for each holdout experiment
|
|
||||||
- [ ] Compute Δ (in-source AUC - held-out AUC)
|
|
||||||
- [ ] If Δ > 0.05-0.10, model is learning source-specific features
|
|
||||||
- [ ] Generate source holdout comparison table
|
|
||||||
- [ ] For each model/condition:
|
|
||||||
- [ ] Generate Grad-CAM visualizations (10-20 images per condition)
|
|
||||||
- [ ] Organize by experiment, prediction type, and source
|
|
||||||
- [ ] Answer key questions:
|
|
||||||
- [ ] Which preprocessing choices are statistically significant?
|
|
||||||
- [ ] Do certain sources benefit more from specific preprocessing?
|
|
||||||
- [ ] Is there an interaction between facecrop and augmentation?
|
|
||||||
- [ ] Are shortcuts being learned (resolution, color distribution)?
|
|
||||||
- [ ] Is the model learning source-specific features (source holdout)?
|
|
||||||
- [ ] Does augmentation remove shortcuts or over-regularize?
|
|
||||||
- [ ] What features do models focus on (based on Grad-CAM)?
|
|
||||||
- [ ] Generate comprehensive metrics comparison table
|
|
||||||
- [ ] Use paired fold-wise statistical tests for model comparisons, with bootstrap confidence intervals for key metrics where useful
|
|
||||||
- [ ] Provide evidence-based conclusions for each experiment
|
|
||||||
- [ ] Provide recommendations for Phase 3 (best preprocessing settings)
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 3: Extended Architecture Exploration
|
|
||||||
|
|
||||||
### 3.1 Experiment Configs
|
|
||||||
Use the best preprocessing choices from Phase 2. The placeholders below assume 224×224, face crop enabled, and no augmentation unless Phase 2 results justify different settings.
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase3/p3_resnet34.json`
|
|
||||||
- backbone: resnet34
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 224
|
|
||||||
- augment: false (placeholder until Phase 2 results confirm)
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase3/p3_resnet50.json`
|
|
||||||
- backbone: resnet50
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 224
|
|
||||||
- augment: false (placeholder until Phase 2 results confirm)
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase3/p3_efficientnet_b0.json`
|
|
||||||
- backbone: efficientnet_b0
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 224
|
|
||||||
- augment: false (placeholder until Phase 2 results confirm)
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase3/p3_convnext_tiny.json`
|
|
||||||
- backbone: convnext_tiny
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 5e-5 (reduced for ConvNeXt stability)
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 224
|
|
||||||
- augment: false (placeholder until Phase 2 results confirm)
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
|
|
||||||
- [x] Create `classifier/configs/phase3/p3_mobilenetv3_small.json`
|
|
||||||
- backbone: mobilenet_v3_small
|
|
||||||
- pretrained: true
|
|
||||||
- epochs: 15
|
|
||||||
- batch_size: 32
|
|
||||||
- lr: 1e-4
|
|
||||||
- weight_decay: 1e-4
|
|
||||||
- image_size: 224
|
|
||||||
- augment: false (placeholder until Phase 2 results confirm)
|
|
||||||
- subsample: 0.2
|
|
||||||
- seed: 42
|
|
||||||
- early_stopping_patience: 5
|
|
||||||
|
|
||||||
- [x] Remove `p3a_mobilenet_v3_large.json` (not in plan, MobileNet V3 Large fills no distinct niche)
|
|
||||||
|
|
||||||
### 3.2 Model Implementation
|
|
||||||
- [x] Implement ConvNeXt-Tiny in `classifier/src/models/convnext.py`
|
|
||||||
- [x] Implement MobileNetV3-Small in `classifier/src/models/mobilenet.py`
|
|
||||||
- [x] Register both models in `classifier/src/models/__init__.py`
|
|
||||||
|
|
||||||
### 3.3 Training
|
|
||||||
- [ ] Train ResNet34 with 5-fold stratified group CV
|
|
||||||
- [ ] Train ResNet50 with 5-fold stratified group CV
|
|
||||||
- [ ] Train EfficientNet-B0 with 5-fold stratified group CV
|
|
||||||
- [ ] Train ConvNeXt-Tiny with 5-fold stratified group CV
|
|
||||||
- [ ] Train MobileNetV3-Small with 5-fold stratified group CV
|
|
||||||
- [ ] Save all checkpoints and metrics
|
|
||||||
|
|
||||||
### 3.4 Analysis
|
|
||||||
- [ ] Use `classifier/notebooks/05_phase3_analysis.ipynb` for Phase 3 analysis
|
|
||||||
- [ ] Load 5-fold stratified group CV results for all models (mean ± std and confidence intervals)
|
|
||||||
- [ ] Generate overall metrics for each model
|
|
||||||
- [ ] Generate per-source metrics for each model
|
|
||||||
- [ ] Compare with Phase 1 baselines (ResNet18, SimpleCNN)
|
|
||||||
- [ ] Statistical significance testing vs baselines
|
|
||||||
- [ ] Generate Grad-CAM visualizations for top models (10-20 images each)
|
|
||||||
- [ ] Parameter count vs performance analysis
|
|
||||||
- [ ] Conclusions: Which architectures work best and why
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 4: Final Analysis on Best Models
|
|
||||||
|
|
||||||
### 4.1 Select Top Models
|
|
||||||
- [ ] Based on Phases 1-3 results, select top 3-4 models
|
|
||||||
- [ ] Document selection criteria (e.g., top AUC, balanced performance, efficiency)
|
|
||||||
|
|
||||||
### 4.2 Data Quantity Scaling (4A)
|
|
||||||
- [ ] For each selected model, create configs for different data sizes:
|
|
||||||
- [ ] `classifier/configs/phase4/p4a_<model>_20pct.json` (subsample: 0.2)
|
|
||||||
- [ ] `classifier/configs/phase4/p4a_<model>_50pct.json` (subsample: 0.5)
|
|
||||||
- [ ] `classifier/configs/phase4/p4a_<model>_100pct.json` (subsample: 1.0)
|
|
||||||
- [ ] In every 4A config, explicitly set the best Phase 2 preprocessing choices:
|
|
||||||
- image_size: best from Phase 2A
|
|
||||||
- face_crop: best from Phase 2B/E
|
|
||||||
- augment: best from Phase 2D/E
|
|
||||||
- [ ] Train each model with 5-fold stratified group CV at all three data sizes
|
|
||||||
- [ ] Compare how each model scales with more data
|
|
||||||
|
|
||||||
### 4.3 Full Dataset Evaluation (4B)
|
|
||||||
- [ ] For each selected model, create config for full dataset:
|
|
||||||
- `classifier/configs/phase4/p4b_<model>_full.json` (subsample: 1.0)
|
|
||||||
- [ ] In every 4B config, explicitly set the same best Phase 2 preprocessing choices used in 4A
|
|
||||||
- [ ] Train each model on full dataset with 5-fold stratified group CV
|
|
||||||
- [ ] Generate detailed per-source metrics
|
|
||||||
- [ ] Generate Grad-CAM visualizations (10-20 images each)
|
|
||||||
- [ ] Perform hard example analysis (false positives/negatives) with visualizations
|
|
||||||
- [ ] Generate confidence distribution histograms
|
|
||||||
- [ ] Cross-validation results (mean ± std with confidence intervals)
|
|
||||||
|
|
||||||
### 4.4 Analysis
|
|
||||||
- [ ] Use `classifier/notebooks/06_phase4_analysis.ipynb` for Phase 4 analysis
|
|
||||||
- [ ] Load data quantity scaling results
|
|
||||||
- [ ] Load full dataset evaluation results
|
|
||||||
- [ ] Generate comprehensive metrics comparison table
|
|
||||||
- [ ] Generate per-source metrics for final models
|
|
||||||
- [ ] Generate Grad-CAM galleries for final models
|
|
||||||
- [ ] Perform hard example analysis with visualizations
|
|
||||||
- [ ] Generate confidence distribution histograms
|
|
||||||
- [ ] Final model comparison and selection
|
|
||||||
- [ ] Conclusions and recommendations
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Notebooks and Analysis
|
|
||||||
|
|
||||||
This section is the consolidated notebook checklist for the notebooks referenced in the phase sections above; do not create duplicate notebooks for the same phase.
|
|
||||||
|
|
||||||
### 5.1 Exploratory Data Analysis
|
|
||||||
- [x] Create `classifier/notebooks/01_eda.ipynb`
|
|
||||||
- [x] Dataset overview (real vs fake distribution, sources)
|
|
||||||
- [x] Image resolution/aspect ratio analysis (identify potential shortcuts)
|
|
||||||
- [x] Color distribution analysis (identify potential shortcuts)
|
|
||||||
- [x] Sample visualization from each source
|
|
||||||
- [x] Statistical summary of the dataset
|
|
||||||
- [x] Data quality checks
|
|
||||||
|
|
||||||
### 5.2 Preprocessing Pipeline
|
|
||||||
- [x] Create `classifier/notebooks/02_preprocessing.ipynb`
|
|
||||||
- [x] Square crop and resize implementation demonstration
|
|
||||||
- [x] Face crop (MTCNN) demonstration and effectiveness analysis
|
|
||||||
- [x] Augmentation pipeline visualization (before/after examples)
|
|
||||||
- [x] Z-score normalization comparison (ImageNet vs real-image-only)
|
|
||||||
- [x] Data split verification (group-aware by basename, no overlap)
|
|
||||||
- [x] Preprocessing impact visualization
|
|
||||||
|
|
||||||
### 5.3 Phase 1 Analysis
|
|
||||||
- [x] Create `classifier/notebooks/03_phase1_analysis.ipynb`
|
|
||||||
- [x] Load Phase 1 training results
|
|
||||||
- [x] Generate 5-fold stratified group CV results (mean ± std with confidence intervals)
|
|
||||||
- [x] Generate per-source metrics for each model
|
|
||||||
- [x] Generate train/val/test performance curves
|
|
||||||
- [x] Generate confusion matrices
|
|
||||||
- [x] Perform statistical significance testing between models
|
|
||||||
- [x] Generate Grad-CAM visualizations (10-20 images each)
|
|
||||||
- [x] Document conclusions: Which baseline is better and why
|
|
||||||
|
|
||||||
### 5.4 Phase 2 Analysis
|
|
||||||
- [x] Create `classifier/notebooks/04_phase2_analysis.ipynb`
|
|
||||||
- [ ] Load all Phase 2 experiment results
|
|
||||||
- [ ] For each experiment (2A-2E):
|
|
||||||
- [ ] Generate 5-fold stratified group CV results (mean ± std with confidence intervals)
|
|
||||||
- [ ] Generate overall metrics
|
|
||||||
- [ ] Generate per-source metrics
|
|
||||||
- [ ] Calculate train/val gap
|
|
||||||
- [ ] Calculate pairwise source AUC variance (wiki-vs-source AUC variance)
|
|
||||||
- [ ] Perform statistical significance testing
|
|
||||||
- [ ] Generate comparison tables across all Phase 2 experiments
|
|
||||||
- [ ] Generate comparison visualizations (bar charts, heatmaps)
|
|
||||||
- [ ] For each model/condition, generate Grad-CAM visualizations (10-20 images)
|
|
||||||
- [ ] Organize visualizations by experiment, model, prediction type, and source
|
|
||||||
- [ ] Answer key analysis questions
|
|
||||||
- [ ] Generate comprehensive metrics comparison table
|
|
||||||
- [ ] Provide evidence-based conclusions for each experiment
|
|
||||||
- [ ] Provide recommendations for Phase 3
|
|
||||||
|
|
||||||
### 5.5 Phase 3 Analysis
|
|
||||||
- [ ] Create `classifier/notebooks/05_phase3_analysis.ipynb`
|
|
||||||
- [ ] Load Phase 3 training results
|
|
||||||
- [ ] Generate 5-fold stratified group CV results for each model (mean ± std with confidence intervals)
|
|
||||||
- [ ] Generate per-source metrics for each model
|
|
||||||
- [ ] Compare with Phase 1 baselines (ResNet18, SimpleCNN)
|
|
||||||
- [ ] Perform statistical significance testing vs baselines
|
|
||||||
- [ ] Generate Grad-CAM visualizations for top models (10-20 images each)
|
|
||||||
- [ ] Parameter count vs performance analysis
|
|
||||||
- [ ] Conclusions: Which architectures work best and why
|
|
||||||
|
|
||||||
### 5.6 Phase 4 Analysis
|
|
||||||
- [ ] Create `classifier/notebooks/06_phase4_analysis.ipynb`
|
|
||||||
- [ ] Load data quantity scaling results
|
|
||||||
- [ ] Load full dataset evaluation results
|
|
||||||
- [ ] Generate comprehensive metrics comparison table
|
|
||||||
- [ ] Generate per-source metrics for final models
|
|
||||||
- [ ] Generate Grad-CAM galleries for final models
|
|
||||||
- [ ] Perform hard example analysis with visualizations
|
|
||||||
- [ ] Generate confidence distribution histograms
|
|
||||||
- [ ] Final model comparison and selection
|
|
||||||
- [ ] Conclusions and recommendations
|
|
||||||
|
|
||||||
### 5.7 Grad-CAM Deep Dive (Optional)
|
|
||||||
- [ ] Create `classifier/notebooks/07_gradcam_deep_dive.ipynb`
|
|
||||||
- [ ] Load Grad-CAM results from all phases
|
|
||||||
- [ ] Comprehensive Grad-CAM analysis across all phases and models
|
|
||||||
- [ ] Feature visualization for different model architectures
|
|
||||||
- [ ] CNN vs EfficientNet vs ConvNeXt comparison
|
|
||||||
- [ ] What regions do different architectures focus on?
|
|
||||||
- [ ] Are there systematic differences in attention patterns?
|
|
||||||
- [ ] Evidence of shortcut removal analysis across phases
|
|
||||||
- [ ] Temporal analysis: does model attention change with different preprocessing?
|
|
||||||
- [ ] Generate visual explanations suitable for presentation
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Code Implementation Tasks
|
|
||||||
|
|
||||||
### Cross-Validation Implementation
|
|
||||||
- [x] Update `classifier/src/training/trainer.py` to support 5-fold stratified group CV by basename
|
|
||||||
- [x] Update `classifier/src/evaluation/evaluate.py` to support grouped CV splits
|
|
||||||
- [x] Implement metric aggregation across folds (mean ± std)
|
|
||||||
- [x] Ensure all metrics report confidence intervals
|
|
||||||
- [x] Reuse the same fold assignments for comparable experiments so paired statistical tests are valid
|
|
||||||
- [x] Rename `classifier/run_cv.py` to `classifier/run.py` (pipeline expects classifier/run.py)
|
|
||||||
- [x] Rename `classifier/run_cv.py` to `classifier/run.py` (pipeline expects classifier/run.py)
|
|
||||||
|
|
||||||
### Model Implementations
|
|
||||||
- [x] Implement ConvNeXt-Tiny in `classifier/src/models/convnext.py`
|
|
||||||
- [x] Implement MobileNetV3-Small in `classifier/src/models/mobilenet.py`
|
|
||||||
- [x] Register both models in `classifier/src/models/__init__.py`
|
|
||||||
|
|
||||||
### Normalization Implementation
|
|
||||||
- [ ] Implement function to calculate mean/std from real training images only
|
|
||||||
- [ ] Update `classifier/src/preprocessing/pipeline.py` to support custom normalization stats
|
|
||||||
- [ ] Test ImageNet normalization vs real-image-only normalization
|
|
||||||
|
|
||||||
### Evaluation Improvements
|
|
||||||
- [ ] Ensure test set uses `train=False` to disable augmentation
|
|
||||||
- [ ] Ensure diagnostic evaluation transforms never change the training data
|
|
||||||
- [ ] Verify CV fold assignments are identical across comparable experiments (same seed and basename grouping)
|
|
||||||
- [ ] Implement per-source metrics with detection rate and false alarm rate
|
|
||||||
- [ ] Implement pairwise AUC calculations
|
|
||||||
- [ ] Implement train/val gap calculations
|
|
||||||
- [ ] Implement pairwise source AUC variance calculations
|
|
||||||
|
|
||||||
### Grad-CAM Improvements
|
|
||||||
- [x] Ensure Grad-CAM works for all model types (CNN-based)
|
|
||||||
- [x] Implement Grad-CAM for ConvNeXt (last Conv2d found automatically by `find_conv()`)
|
|
||||||
- [x] Implement Grad-CAM for MobileNetV3 (last Conv2d found automatically by `find_conv()`)
|
|
||||||
- [ ] Organize Grad-CAM outputs by experiment, model, prediction type, source
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Final Report Preparation
|
|
||||||
- [ ] Compile results from all phases
|
|
||||||
- [ ] Create presentation slides (PDF format)
|
|
||||||
- [ ] Brief description of deep learning solutions (discriminative + generative)
|
|
||||||
- [ ] Description of implementation steps and improvements
|
|
||||||
- [ ] Motivate choices for architecture, training strategy, etc.
|
|
||||||
- [ ] Show intermediate results
|
|
||||||
- [ ] Interpret results and what changed
|
|
||||||
- [ ] What was decided to improve results
|
|
||||||
- [ ] Classification performance results
|
|
||||||
- [ ] Experimental setup
|
|
||||||
- [ ] Train/val/test splits
|
|
||||||
- [ ] Performance metrics chosen
|
|
||||||
- [ ] Data generation performance results
|
|
||||||
- [ ] Experimental setup
|
|
||||||
- [ ] Performance metrics chosen
|
|
||||||
- [ ] Discussion and conclusions
|
|
||||||
- [ ] Comments on performance
|
|
||||||
- [ ] Final remarks
|
|
||||||
- [ ] Fill auto-evaluation file
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Summary
|
|
||||||
|
|
||||||
Total tasks: ~150+
|
|
||||||
|
|
||||||
This implementation plan covers:
|
|
||||||
- ✅ All 4 phases with comprehensive experiments
|
|
||||||
- ✅ 5-fold stratified group cross-validation for all experiments
|
|
||||||
- ✅ 7 analysis notebooks for robust validation
|
|
||||||
- ✅ Shortcut analysis (resolution/ratio + color distribution + source holdout)
|
|
||||||
- ✅ Source holdout experiments to detect source-specific feature learning
|
|
||||||
- ✅ Grad-CAM visualizations for explainability
|
|
||||||
- ✅ Statistical analysis with confidence intervals
|
|
||||||
- ✅ Per-source metrics for all experiments
|
|
||||||
- ✅ Data quantity scaling analysis
|
|
||||||
- ✅ Full dataset evaluation on best models
|
|
||||||
- ✅ Comprehensive documentation and reporting
|
|
||||||
|
|
||||||
**Key Features:**
|
|
||||||
- Reproducible experiments with fixed seeds
|
|
||||||
- Stratified group CV keeps basename groups together while balancing class distribution
|
|
||||||
- Multiple shortcut analyses to prevent model cheating (resolution, color, source-specific)
|
|
||||||
- Source holdout experiments to test generalization to unseen sources
|
|
||||||
- Grad-CAM for explainability
|
|
||||||
- Statistical rigor with confidence intervals
|
|
||||||
- Per-source analysis to understand model behavior
|
|
||||||
- Clear progression from baselines -> preprocessing -> architectures -> final evaluation
|
|
||||||
@@ -1,449 +0,0 @@
|
|||||||
# Classifier Reorganization Plan (v2)
|
|
||||||
|
|
||||||
## Analysis of Current Phasing Issues
|
|
||||||
|
|
||||||
Your current phasing has several problems that make it difficult to present a rigorous, explainable report:
|
|
||||||
|
|
||||||
### Current Problems
|
|
||||||
|
|
||||||
1. **Inconsistent comparison conditions**:
|
|
||||||
- SimpleCNN uses lr=1e-3, ResNet uses lr=1e-4
|
|
||||||
- SimpleCNN trains 20 epochs (no ES), ResNet18 trains 15 epochs (with ES)
|
|
||||||
- Makes direct comparisons invalid
|
|
||||||
|
|
||||||
2. **No cross-validation**:
|
|
||||||
- Only a single 80/10/10 split
|
|
||||||
- Results may be split-dependent
|
|
||||||
- No confidence intervals on metrics
|
|
||||||
|
|
||||||
3. **Augmentation testing is incomplete**:
|
|
||||||
- Only tested on ResNet18 (Phase 3), not across architectures
|
|
||||||
- Performance drop could mean: (a) removing shortcuts (good) or (b) over-regularization (bad)
|
|
||||||
- No way to distinguish these cases
|
|
||||||
|
|
||||||
4. **Facecrop impact not generalized**:
|
|
||||||
- Only ResNet18 tested with facecrop
|
|
||||||
- Don't know if EfficientNet or ViT benefit similarly
|
|
||||||
|
|
||||||
5. **Full dataset only on one model**:
|
|
||||||
- Only ResNet18 tested on full dataset
|
|
||||||
- Don't know if data quantity helps all models equally
|
|
||||||
|
|
||||||
6. **Test set integrity**:
|
|
||||||
- Need to verify test set uses original images (no augmentation, no preprocessing or minimal if really necessary)
|
|
||||||
- Need to ensure same train/val/test splits across all model comparisons
|
|
||||||
- Need central config for shared parameters across phases
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Recommended Reorganization
|
|
||||||
|
|
||||||
I suggest reorganizing into **4 phases** with clear, isolated variables. All phases use **5-fold stratified cross-validation** as standard practice to ensure balanced class distribution across folds.
|
|
||||||
|
|
||||||
### Phase 1: Controlled Baseline Comparison
|
|
||||||
|
|
||||||
**Goal**: Compare simple architectures under identical conditions to establish baselines
|
|
||||||
|
|
||||||
**Fixed conditions for ALL models**:
|
|
||||||
- Data: 20% subsample
|
|
||||||
- Resolution: 128×128
|
|
||||||
- No face crop
|
|
||||||
- No augmentation
|
|
||||||
- Optimizer: AdamW (lr=1e-4, weight_decay=1e-4)
|
|
||||||
- Scheduler: CosineAnnealingLR (T_max=15)
|
|
||||||
- Epochs: 15 with early stopping (patience=5)
|
|
||||||
- Batch size: 32
|
|
||||||
- 5-fold stratified cross-validation (report mean ± std)
|
|
||||||
|
|
||||||
| Model | Params | Expected AUC (mean ± std) |
|
|
||||||
|-------|--------|---------------------------|
|
|
||||||
| SimpleCNN | ~400k | ? |
|
|
||||||
| ResNet18 | ~11.7M | ? |
|
|
||||||
|
|
||||||
**This gives you**: Clean, comparable baseline for simple architectures with confidence intervals
|
|
||||||
|
|
||||||
**These same 2 models will be used in Phase 2 for preprocessing experiments.**
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### Phase 2: Preprocessing Impact (Same 2 Models from Phase 1)
|
|
||||||
|
|
||||||
**Goal**: Test each preprocessing change on the SAME 2 models from Phase 1
|
|
||||||
|
|
||||||
**Experimental questions**:
|
|
||||||
- Does higher resolution improve performance?
|
|
||||||
- Does face cropping improve performance?
|
|
||||||
- Does augmentation improve or hurt performance?
|
|
||||||
- Does augmentation interact with face cropping?
|
|
||||||
- Is the model learning any shortcuts (e.g., resolution differences, aspect ratios, etc.)?
|
|
||||||
|
|
||||||
#### 2A: Shortcut Analysis
|
|
||||||
**Goal**: Establish whether the baseline model exploits geometry, colour, or source-specific shortcuts before drawing any conclusions from preprocessing experiments.
|
|
||||||
|
|
||||||
**Test 1: Resolution/Ratio Shortcuts (Letterboxing)**
|
|
||||||
- Train on original images (real=rectangular, fake=square); evaluate the same checkpoint under standard crop vs letterbox-padded real images to confirm geometry is or is not a discriminative cue
|
|
||||||
- Models: **ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
- Resolution: 224×224
|
|
||||||
- No facecrop, no augmentation
|
|
||||||
|
|
||||||
| Experiment | AUC | Train/Val Gap | Per-Source AUC Variance |
|
|
||||||
|------------|-----|---------------|-------------------------|
|
|
||||||
| Original images (standard eval) | ? | ? | ? |
|
|
||||||
| Matched geometry (letterboxed real images) | ? | ? | ? |
|
|
||||||
|
|
||||||
**Test 2: Color Distribution Shortcuts**
|
|
||||||
- Compare: Train with ImageNet normalization stats vs real-image-only normalization stats
|
|
||||||
- Models: **ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
- Resolution: 224×224
|
|
||||||
- No facecrop, no augmentation
|
|
||||||
- ImageNet stats: mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)
|
|
||||||
- Real-image stats: Calculate mean/std from real training images only, apply to all
|
|
||||||
|
|
||||||
| Experiment | AUC | Train/Val Gap | Per-Source AUC Variance |
|
|
||||||
|------------|-----|---------------|-------------------------|
|
|
||||||
| ImageNet normalization | ? | ? | ? |
|
|
||||||
| Real-image-only normalization | ? | ? | ? |
|
|
||||||
|
|
||||||
**Test 3: Source-Specific Feature Learning (Source Holdout)**
|
|
||||||
- Compare: Train on all sources vs train with one source held out
|
|
||||||
- Models: **ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
- Resolution: 224×224
|
|
||||||
- No facecrop, no augmentation
|
|
||||||
- Hold out each fake source (text2img, inpainting, insight) separately
|
|
||||||
|
|
||||||
| Experiment | Held-out Source | Train Sources | Held-out AUC | In-Source AUC | Δ (In-Source - Held-out) |
|
|
||||||
|------------|-----------------|---------------|--------------|---------------|--------------------------|
|
|
||||||
| Baseline | None | All | - | ? | - |
|
|
||||||
| Holdout text2img | text2img | wiki, inpainting, insight | ? | ? | ? |
|
|
||||||
| Holdout inpainting | inpainting | wiki, text2img, insight | ? | ? | ? |
|
|
||||||
| Holdout insight | insight | wiki, text2img, inpainting | ? | ? | ? |
|
|
||||||
|
|
||||||
**Interpretation**: If held-out source AUC is significantly lower than in-source AUC (Δ > 0.05-0.10), the model is learning source-specific features. If AUC drop under matched geometry is significant, the model exploits aspect-ratio as a shortcut — this must be known before interpreting resolution or facecrop results.
|
|
||||||
|
|
||||||
#### 2B: Resolution Impact (no facecrop, no augmentation)
|
|
||||||
- Test: 128×128 vs 224×224
|
|
||||||
- Models: **SimpleCNN, ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
|
|
||||||
| Model | 128×128 AUC | 224×224 AUC | Δ |
|
|
||||||
|-------|-------------|-------------|---|
|
|
||||||
| SimpleCNN | ? | ? | ? |
|
|
||||||
| ResNet18 | ? | ? | ? |
|
|
||||||
|
|
||||||
#### 2C: Facecrop Impact (224×224, no augmentation)
|
|
||||||
- Test: No facecrop vs MTCNN facecrop
|
|
||||||
- Models: **SimpleCNN, ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
|
|
||||||
| Model | No Facecrop AUC | Facecrop AUC | Δ |
|
|
||||||
|-------|-----------------|--------------|---|
|
|
||||||
| SimpleCNN | ? | ? | ? |
|
|
||||||
| ResNet18 | ? | ? | ? |
|
|
||||||
|
|
||||||
#### 2D: Augmentation Impact (224×224, without facecrop)
|
|
||||||
- Test: No augmentation vs augmentation
|
|
||||||
- Models: **SimpleCNN, ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
- **Verify test set has no augmentation** (code inspection of `get_transforms(train=False, ...)`)
|
|
||||||
- **Analyze shortcut removal**: Compare train/val gaps and per-source AUC balance
|
|
||||||
|
|
||||||
| Model | No Aug AUC | With Aug AUC | Δ | Train/Val Gap (No Aug) | Train/Val Gap (With Aug) |
|
|
||||||
|-------|------------|--------------|---|------------------------|--------------------------|
|
|
||||||
| SimpleCNN | ? | ? | ? | ? | ? |
|
|
||||||
| ResNet18 | ? | ? | ? | ? | ? |
|
|
||||||
|
|
||||||
**Experimental question**: Does augmentation without facecrop improve or hurt performance?
|
|
||||||
|
|
||||||
#### 2E: Augmentation + Facecrop Combined (224×224)
|
|
||||||
- Test: Facecrop only vs Facecrop + augmentation
|
|
||||||
- Models: **SimpleCNN, ResNet18**
|
|
||||||
- Data: 20% subsample
|
|
||||||
- 5-fold stratified CV (balanced class distribution)
|
|
||||||
- **Analyze shortcut removal**: Compare train/val gaps and per-source AUC balance
|
|
||||||
|
|
||||||
| Model | Facecrop Only AUC | Facecrop + Aug AUC | Δ | Train/Val Gap (Only) | Train/Val Gap (With Aug) |
|
|
||||||
|-------|-------------------|--------------------|---|----------------------|--------------------------|
|
|
||||||
| SimpleCNN | ? | ? | ? | ? | ? |
|
|
||||||
| ResNet18 | ? | ? | ? | ? | ? |
|
|
||||||
|
|
||||||
**Experimental question**: Does augmentation with facecrop improve or hurt performance compared to facecrop alone?
|
|
||||||
|
|
||||||
**This gives you**:
|
|
||||||
- Isolated impact of each preprocessing choice on SimpleCNN and ResNet18
|
|
||||||
- Verification that the model is not learning shortcuts
|
|
||||||
- Understanding of how augmentation interacts with face cropping
|
|
||||||
- Shortcut removal analysis through train/val gap and per-source AUC metrics
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### Phase 3: Extended Architecture Exploration
|
|
||||||
|
|
||||||
**Goal**: Test additional architectures to find the best performing models
|
|
||||||
|
|
||||||
**Fixed conditions** (based on best findings from Phase 2):
|
|
||||||
- Data: 20% subsample
|
|
||||||
- Resolution: Best from Phase 2A (likely 224×224)
|
|
||||||
- Facecrop: Best from Phase 2B/E (likely Yes)
|
|
||||||
- Augmentation: Best from Phase 2D/E (depends on experimental results)
|
|
||||||
- Optimizer: AdamW (lr=1e-4, weight_decay=1e-4)
|
|
||||||
- Scheduler: CosineAnnealingLR (T_max=15)
|
|
||||||
- Epochs: 15 with early stopping (patience=5)
|
|
||||||
- Batch size: 32
|
|
||||||
- 5-fold stratified cross-validation (balanced class distribution)
|
|
||||||
|
|
||||||
| Model | Params | Rationale |
|
|
||||||
|-------|--------|-----------|
|
|
||||||
| ResNet34 | ~21.8M | Deeper ResNet - test if more capacity helps |
|
|
||||||
| ResNet50 | ~25.6M | Even deeper with bottleneck blocks |
|
|
||||||
| EfficientNet-B0 | ~4.0M | Efficient compound scaling |
|
|
||||||
| ConvNeXt-Tiny | ~29M | Modern CNN, different architecture family |
|
|
||||||
| MobileNetV3-Small | ~2.5M | Lightweight efficiency comparison |
|
|
||||||
|
|
||||||
**This gives you**: Extended architecture exploration to identify top-performing models for Phase 4
|
|
||||||
- ResNet depth progression (18 -> 34 -> 50)
|
|
||||||
- Efficient architectures (EfficientNet-B0, MobileNetV3-Small)
|
|
||||||
- Modern CNN with different inductive bias (ConvNeXt-Tiny)
|
|
||||||
- Size range (2.5M to 29M parameters)
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### Phase 4: Final Analysis on Best Models
|
|
||||||
|
|
||||||
**Goal**: Comprehensive evaluation of top-performing models from Phases 1-3
|
|
||||||
|
|
||||||
**Select top 3-4 models** based on Phase 1-3 results (e.g., ResNet18, ResNet34, EfficientNet-B0, ConvNeXt-Tiny)
|
|
||||||
|
|
||||||
#### 4A: Data Quantity Scaling
|
|
||||||
Test how each best model scales with more data:
|
|
||||||
|
|
||||||
| Model | 20% Data AUC | 50% Data AUC | 100% Data AUC | Δ (100% - 20%) |
|
|
||||||
|-------|--------------|--------------|---------------|----------------|
|
|
||||||
| Model 1 | ? | ? | ? | ? |
|
|
||||||
| Model 2 | ? | ? | ? | ? |
|
|
||||||
| Model 3 | ? | ? | ? | ? |
|
|
||||||
| Model 4 | ? | ? | ? | ? |
|
|
||||||
|
|
||||||
**Fixed conditions**:
|
|
||||||
- Resolution: Best from Phase 2A
|
|
||||||
- Facecrop: Best from Phase 2B/E
|
|
||||||
- Augmentation: Best from Phase 2D/E
|
|
||||||
- 5-fold stratified cross-validation (balanced class distribution)
|
|
||||||
|
|
||||||
#### 4B: Comprehensive Evaluation on Full Dataset
|
|
||||||
- Train best models on **full dataset** (100%)
|
|
||||||
- Detailed per-source metrics (text2img, inpainting, insight)
|
|
||||||
- Grad-CAM visualizations for explainability
|
|
||||||
- Hard example analysis (false positives/negatives)
|
|
||||||
- Confidence distribution analysis
|
|
||||||
- Cross-validation results (mean ± std)
|
|
||||||
|
|
||||||
**This gives you**: Final, comprehensive evaluation of the best models with full explainability
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### Notebooks and Analysis
|
|
||||||
|
|
||||||
**Goal**: Use Jupyter notebooks for comprehensive analysis and validation of each phase
|
|
||||||
|
|
||||||
#### **01_eda.ipynb** - Exploratory Data Analysis
|
|
||||||
- Dataset overview (real vs fake distribution, sources)
|
|
||||||
- Image resolution/aspect ratio analysis (identify potential shortcuts)
|
|
||||||
- Color distribution analysis (identify potential shortcuts)
|
|
||||||
- Sample visualization from each source (text2img, inpainting, insight, wiki)
|
|
||||||
- Statistical summary of the dataset
|
|
||||||
- Data quality checks
|
|
||||||
|
|
||||||
#### **02_preprocessing.ipynb** - Preprocessing Pipeline
|
|
||||||
- Square crop and resize implementation demonstration
|
|
||||||
- Face crop (MTCNN) demonstration and effectiveness analysis
|
|
||||||
- Augmentation pipeline visualization (before/after examples)
|
|
||||||
- Z-score normalization comparison (ImageNet vs real-image-only)
|
|
||||||
- Data split verification (group-aware by basename, no overlap)
|
|
||||||
- Preprocessing impact visualization
|
|
||||||
|
|
||||||
#### **03_phase1_analysis.ipynb** - Phase 1: Architecture Baseline
|
|
||||||
- SimpleCNN vs ResNet18 comparison
|
|
||||||
- 5-fold stratified CV results (mean ± std with confidence intervals)
|
|
||||||
- Per-source metrics for each model (text2img, inpainting, insight)
|
|
||||||
- Train/val/test performance curves across epochs
|
|
||||||
- Confusion matrices for each model
|
|
||||||
- Statistical significance testing between models
|
|
||||||
- Grad-CAM visualizations for both models (10-20 images each)
|
|
||||||
- Conclusions: Which baseline is better and why
|
|
||||||
|
|
||||||
#### **04_phase2_analysis.ipynb** - Phase 2: Preprocessing Impact
|
|
||||||
- **2A**: Resolution impact (128×128 vs 224×224)
|
|
||||||
- **2B**: Facecrop impact
|
|
||||||
- **2C**: Shortcut analysis (resolution/ratio + color distribution)
|
|
||||||
- **2D**: Augmentation impact (without facecrop)
|
|
||||||
- **2E**: Augmentation + facecrop combined
|
|
||||||
|
|
||||||
For each experiment:
|
|
||||||
- 5-fold CV results (mean ± std with confidence intervals)
|
|
||||||
- Per-source metrics (text2img, inpainting, insight)
|
|
||||||
- Statistical significance testing vs baseline
|
|
||||||
- Comparison tables across all Phase 2 experiments
|
|
||||||
- Grad-CAM visualizations (10-20 images per condition)
|
|
||||||
- Analysis of train/val gap changes
|
|
||||||
- Analysis of per-source AUC variance changes
|
|
||||||
|
|
||||||
**Overall Phase 2 conclusions**:
|
|
||||||
- Which preprocessing choices work best and why
|
|
||||||
- Are shortcuts being learned (resolution, color distribution)?
|
|
||||||
- Does augmentation remove shortcuts or over-regularize?
|
|
||||||
- Recommendations for Phase 3 (best preprocessing settings)
|
|
||||||
|
|
||||||
#### **05_phase3_analysis.ipynb** - Phase 3: Extended Architecture Exploration
|
|
||||||
- ResNet34, ResNet50, EfficientNet-B0, ConvNeXt-Tiny, MobileNetV3-Small
|
|
||||||
- 5-fold CV results (mean ± std) for each model
|
|
||||||
- Per-source metrics for each model
|
|
||||||
- Comparison with Phase 1 baselines (ResNet18, SimpleCNN)
|
|
||||||
- Statistical significance testing vs baselines
|
|
||||||
- Grad-CAM visualizations for top models (10-20 images each)
|
|
||||||
- Parameter count vs performance analysis
|
|
||||||
- Conclusions: Which architectures work best and why
|
|
||||||
|
|
||||||
#### **06_phase4_analysis.ipynb** - Phase 4: Final Analysis
|
|
||||||
- **4A**: Data quantity scaling (20%, 50%, 100%) on top 3-4 models
|
|
||||||
- **4B**: Comprehensive evaluation on full dataset
|
|
||||||
- Detailed per-source metrics for final models
|
|
||||||
- Grad-CAM visualizations for final models (10-20 images each)
|
|
||||||
- Hard example analysis (false positives/negatives) with visualizations
|
|
||||||
- Confidence distribution analysis (histograms)
|
|
||||||
- Cross-validation results (mean ± std with confidence intervals)
|
|
||||||
- Final model comparison and selection
|
|
||||||
- Conclusions and recommendations
|
|
||||||
|
|
||||||
#### **07_gradcam_deep_dive.ipynb** - Grad-CAM Deep Dive (optional)
|
|
||||||
- Comprehensive Grad-CAM analysis across all phases and models
|
|
||||||
- Feature visualization for different model architectures (CNN vs EfficientNet vs ConvNeXt)
|
|
||||||
- Comparison of what different models focus on (face regions, backgrounds, artifacts)
|
|
||||||
- Evidence of shortcut removal (or lack thereof) across phases
|
|
||||||
- Temporal analysis: does model attention change with different preprocessing?
|
|
||||||
- Visual explanations suitable for presentation
|
|
||||||
|
|
||||||
**Notebook requirements**:
|
|
||||||
- Each notebook should be self-contained and reproducible
|
|
||||||
- Include statistical analysis with confidence intervals
|
|
||||||
- Generate publication-ready visualizations
|
|
||||||
- Address all experimental questions and hypotheses
|
|
||||||
- Provide clear conclusions for each phase
|
|
||||||
- Use consistent formatting and style across all notebooks
|
|
||||||
- Save all results (metrics, figures, tables) for easy reference
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Key Improvements
|
|
||||||
|
|
||||||
### 1. Stratified Cross-Validation Implementation
|
|
||||||
```python
|
|
||||||
# Use sklearn's StratifiedKFold to ensure balanced class distribution across folds
|
|
||||||
from sklearn.model_selection import StratifiedKFold
|
|
||||||
|
|
||||||
skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)
|
|
||||||
for fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):
|
|
||||||
# Train on train_idx, validate on val_idx
|
|
||||||
# Store metrics per fold
|
|
||||||
```
|
|
||||||
|
|
||||||
### 2. Augmentation Shortcut Removal Analysis (Phase 2D/2E)
|
|
||||||
Track these metrics with/without augmentation:
|
|
||||||
|
|
||||||
| Metric | Without Aug | With Aug | Interpretation |
|
|
||||||
|--------|-------------|----------|----------------|
|
|
||||||
| Train AUC | 0.99 | 0.95 | ↓ Expected |
|
|
||||||
| Val AUC | 0.90 | 0.89 | ↓ Slight |
|
|
||||||
| **Train/Val Gap** | **0.09** | **0.06** | **↓ Good!** |
|
|
||||||
| text2img AUC | 0.98 | 0.96 | ↓ Slight |
|
|
||||||
| InsightFace AUC | 0.82 | 0.85 | **↑ Good!** |
|
|
||||||
| **AUC Variance** | **0.08** | **0.06** | **↓ Good!** |
|
|
||||||
|
|
||||||
**Interpretation**: If train/val gap ↓ AND per-source AUC variance ↓, augmentation is removing shortcuts.
|
|
||||||
|
|
||||||
### 3. Consistent Hyperparameters
|
|
||||||
- Same lr for all models (1e-4 is safe for pretrained, may need adjustment for SimpleCNN)
|
|
||||||
- Same epochs, ES patience, batch size
|
|
||||||
- Only vary the architecture being tested
|
|
||||||
|
|
||||||
### 4. Test Set Integrity and Reproducibility
|
|
||||||
|
|
||||||
**Test set from original source**:
|
|
||||||
- Verify that test set uses original images with minimal preprocessing
|
|
||||||
- Test set should use `get_transforms(train=False, ...)` to disable augmentation
|
|
||||||
- Ensure test images are not preprocessed in a way that could affect model comparisons
|
|
||||||
|
|
||||||
**Reproducible splits across models**:
|
|
||||||
- The code already uses `cfg.get("seed", 42)` for reproducible splits
|
|
||||||
- All experiments should use the same seed (42) to ensure identical train/val/test splits
|
|
||||||
- This ensures fair comparison between models
|
|
||||||
|
|
||||||
**Central config for shared parameters**:
|
|
||||||
- Create a central config file (`classifier/configs/shared.json`) with parameters common across all phases
|
|
||||||
- This includes: seed, val_ratio, test_ratio, batch_size, optimizer settings, etc.
|
|
||||||
- Individual experiment configs can override these defaults
|
|
||||||
|
|
||||||
Example shared config:
|
|
||||||
```json
|
|
||||||
{
|
|
||||||
"seed": 42,
|
|
||||||
"val_ratio": 0.1,
|
|
||||||
"test_ratio": 0.1,
|
|
||||||
"batch_size": 32,
|
|
||||||
"optimizer": {
|
|
||||||
"type": "adamw",
|
|
||||||
"lr": 1e-4,
|
|
||||||
"weight_decay": 1e-4
|
|
||||||
},
|
|
||||||
"scheduler": {
|
|
||||||
"type": "cosine_annealing",
|
|
||||||
"T_max": 15
|
|
||||||
},
|
|
||||||
"early_stopping_patience": 5,
|
|
||||||
"num_workers": 4
|
|
||||||
}
|
|
||||||
```
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Summary Table for Report
|
|
||||||
|
|
||||||
| Phase | Variable Tested | Models | Data | Resolution | Facecrop | Augment | CV |
|
|
||||||
|-------|-----------------|--------|------|------------|----------|---------|----|
|
|
||||||
| 1 | Architecture Baseline | SimpleCNN, ResNet18 | 20% | 128 | No | No | 5-fold stratified |
|
|
||||||
| 2A | Shortcut Analysis | ResNet18 | 20% | 224 | No | No | 5-fold stratified |
|
|
||||||
| 2A-Holdout | Source Holdout | ResNet18 | 20% | 224 | No | No | 5-fold stratified |
|
|
||||||
| 2B | Resolution | SimpleCNN, ResNet18 | 20% | 128/224 | No | No | 5-fold stratified |
|
|
||||||
| 2C | Facecrop | SimpleCNN, ResNet18 | 20% | 224 | ± | No | 5-fold stratified |
|
|
||||||
| 2D | Augmentation (no facecrop) | SimpleCNN, ResNet18 | 20% | 224 | No | ± | 5-fold stratified |
|
|
||||||
| 2E | Augmentation + Facecrop | SimpleCNN, ResNet18 | 20% | 224 | Yes | ± | 5-fold stratified |
|
|
||||||
| 3 | Extended Architectures | ResNet34, ResNet50, EffNet-B0, ConvNeXt-Tiny, MobileNetV3-Small | 20% | Best | Best | Best | 5-fold stratified |
|
|
||||||
| 4A | Data Quantity | Top 3-4 models | 20/50/100% | Best | Best | Best | 5-fold stratified |
|
|
||||||
| 4B | Final Evaluation | Top 3-4 models | 100% | Best | Best | Best | 5-fold stratified |
|
|
||||||
|
|
||||||
This structure gives you:
|
|
||||||
- ✅ Identical comparison conditions across all phases
|
|
||||||
- ✅ 5-fold stratified cross-validation with confidence intervals (ensures balanced class distribution)
|
|
||||||
- ✅ Same 2 baseline models (SimpleCNN, ResNet18) tested across all preprocessing variations (Phase 2)
|
|
||||||
- ✅ Shortcut analysis to verify no bias (Phase 2C)
|
|
||||||
- ✅ Experimental questions about augmentation impact (Phase 2D/2E)
|
|
||||||
- ✅ Shortcut removal analysis via train/val gap and per-source AUC metrics
|
|
||||||
- ✅ Facecrop tested on baseline models (Phase 2B)
|
|
||||||
- ✅ Extended architecture exploration with proven models (Phase 3)
|
|
||||||
- ✅ Final comprehensive analysis on best models (Phase 4)
|
|
||||||
- ✅ Data quantity scaling on multiple best models (Phase 4A)
|
|
||||||
- ✅ Clear, isolated variables per phase
|
|
||||||
- ✅ Explainable progression for report
|
|
||||||
|
|
||||||
**Key Experimental Questions in Phase 2**:
|
|
||||||
- **2C (Shortcut Analysis)**: Is the model learning any shortcuts (e.g., resolution differences, aspect ratios, etc.)?
|
|
||||||
- **2D (Augmentation without facecrop)**: Does augmentation improve or hurt performance?
|
|
||||||
- **2E (Augmentation with facecrop)**: Does augmentation improve or hurt performance compared to facecrop alone?
|
|
||||||
Binary file not shown.
@@ -1,279 +0,0 @@
|
|||||||
# Generator Plan
|
|
||||||
|
|
||||||
The assignment rewards *iterative improvement with intermediate results*. This plan is structured around **model evolution as the spine**: each step has a *because* tied to an observed failure of the previous step. Pipeline ablations are honest but de-emphasized — they clear the table for the real story.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Standard Settings (Applied Everywhere Unless Noted)
|
|
||||||
|
|
||||||
| Setting | Value | Reason |
|
|
||||||
|---------|-------|--------|
|
|
||||||
| Batch size | 64 | Consistent across experiments |
|
|
||||||
| Mixed precision | float16 + GradScaler | Speed |
|
|
||||||
| EMA decay | 0.9999 | Sample from EMA weights for GANs |
|
|
||||||
| FID evaluation | Every 25 epochs | Objective quality tracking |
|
|
||||||
| FID n_real | 5000 | Held-out real images |
|
|
||||||
| Default epochs | 100 | Best-of-each in Phase 4 retrains to 200 |
|
|
||||||
|
|
||||||
Per-model optimizer/hyperparameters are listed inside each phase.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 1 — Pipeline Selection *(quick, one figure)*
|
|
||||||
|
|
||||||
**Goal**: Pick the data pipeline used for every downstream experiment. Don't dwell here — this is clearing the table, not the story.
|
|
||||||
|
|
||||||
Fixed model: **DCGAN at 64×64** (cheapest baseline, fast iteration). One variable per experiment.
|
|
||||||
|
|
||||||
| Experiment | Variable | Variants | Decision |
|
|
||||||
|---|---|---|---|
|
|
||||||
| 1A | Resolution | 64×64 vs 128×128 | Pick by FID — assumed transferable |
|
|
||||||
| 1B | Face crop + alignment | Full image vs MTCNN-aligned | Pick by FID — assumed transferable |
|
|
||||||
| 1C | Augmentation | H-flip only vs H-flip + rotation ±5° + mild color jitter | Per-family: validate inside Phase 2 for GAN, default to H-flip-only for VAE/DDPM |
|
|
||||||
| 1D | Combined dataset | Aligned only vs aligned + raw mixed | Pick by FID — expected to underperform aligned-only |
|
|
||||||
|
|
||||||
**Caveat on transferability**: Phase 1 uses DCGAN as a proxy to choose the pipeline cheaply, then assumes the choice transfers to VAE and DDPM. Resolution and alignment are largely architecture-invariant (more pixels help everyone; structural consistency helps any spatial prior). Augmentation is *not* — diffusion models benefit less from aug, and MSE-VAE may even be hurt by color jitter. So 1C is treated as an **indicative** result for GANs and re-checked per family rather than baked in globally.
|
|
||||||
|
|
||||||
**1D — combined dataset rationale**: Mixing aligned + raw doubles the variance the generator must model (face anywhere/any scale + face fixed) and dilutes the geometric prior. Hypothesis: combined < aligned-only. Cheap to test (one extra DCGAN run). Included for completeness so the report shows we considered it rather than asserting it.
|
|
||||||
|
|
||||||
**MTCNN alignment** (one-time preprocessing, cached to disk):
|
|
||||||
|
|
||||||
```python
|
|
||||||
from facenet_pytorch import MTCNN
|
|
||||||
from skimage.transform import SimilarityTransform, warp
|
|
||||||
import numpy as np
|
|
||||||
from PIL import Image
|
|
||||||
|
|
||||||
mtcnn = MTCNN(keep_all=False, device='cuda')
|
|
||||||
|
|
||||||
REF_LANDMARKS = np.array([ # reference positions in 128×128
|
|
||||||
[38.0, 51.0], # left eye
|
|
||||||
[90.0, 51.0], # right eye
|
|
||||||
[64.0, 71.0], # nose
|
|
||||||
[45.0, 95.0], # left mouth
|
|
||||||
[83.0, 95.0], # right mouth
|
|
||||||
], dtype=np.float32)
|
|
||||||
|
|
||||||
def align_face(img: Image.Image, out_size: int = 128):
|
|
||||||
boxes, _, landmarks = mtcnn.detect(img, landmarks=True)
|
|
||||||
if boxes is None:
|
|
||||||
return None
|
|
||||||
tform = SimilarityTransform()
|
|
||||||
tform.estimate(landmarks[0], REF_LANDMARKS)
|
|
||||||
aligned = warp(np.array(img), tform.inverse,
|
|
||||||
output_shape=(out_size, out_size),
|
|
||||||
order=3, preserve_range=True).astype(np.uint8)
|
|
||||||
return Image.fromarray(aligned)
|
|
||||||
```
|
|
||||||
|
|
||||||
**Augmentation philosophy** — only structure-preserving transforms (face-aligned crops are consistent by design):
|
|
||||||
|
|
||||||
| Transform | Apply? | Reason |
|
|
||||||
|---|---|---|
|
|
||||||
| Horizontal flip | Yes, p=0.5 | Faces are symmetric |
|
|
||||||
| Rotation | Yes, ±5° | Residual head tilt post-alignment |
|
|
||||||
| Color jitter | Yes, mild | brightness ±0.1, contrast ±0.1, saturation ±0.05 |
|
|
||||||
| Translation | No | Breaks alignment |
|
|
||||||
| Vertical flip | No | Meaningless for faces |
|
|
||||||
| Strong blur / noise | No | Teaches the model to generate blur |
|
|
||||||
|
|
||||||
**Output**: ~1 page in the report. Best pipeline carries forward to all phases.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 2 — GAN Evolution *(main spine)*
|
|
||||||
|
|
||||||
**Goal**: The richest narrative — each step has a clear *because* from observed failure. This is the strongest part of the storyline; keep it front and center.
|
|
||||||
|
|
||||||
Best pipeline from Phase 1 fixed throughout.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### 2.1 — DCGAN *(baseline)*
|
|
||||||
|
|
||||||
Simplest GAN baseline. BCE loss, no gradient penalty.
|
|
||||||
|
|
||||||
- Adam β1=0.5, β2=0.999, lr=2e-4
|
|
||||||
- ngf=ndf=64, latent_dim=100
|
|
||||||
- Resolution: 64×64
|
|
||||||
|
|
||||||
**Expected failure**: mode collapse, training instability, oscillating losses. Document these explicitly — they motivate 2.2.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### 2.2 — WGAN-GP
|
|
||||||
|
|
||||||
**Because**: DCGAN showed mode collapse and instability → Wasserstein loss + gradient penalty.
|
|
||||||
|
|
||||||
- Adam β1=0.0, β2=0.9, lr_g=lr_d=1e-4
|
|
||||||
- ngf=ndf=64, latent_dim=128, n_critic=2, gp_lambda=10
|
|
||||||
- Resolution: 64×64
|
|
||||||
|
|
||||||
**Expected**: more stable training, better diversity. Likely remaining issues: texture artifacts, limited global coherence at higher resolution.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### 2.3 — WGAN-GP + Spectral Norm + GroupNorm + Self-Attention
|
|
||||||
|
|
||||||
**Because**: WGAN-GP showed texture artifacts / limited coherence → principled Lipschitz constraint and long-range dependencies.
|
|
||||||
|
|
||||||
- Generator: BatchNorm → GroupNorm (no batch-size coupling)
|
|
||||||
- Critic: InstanceNorm → Spectral Normalization (principled Lipschitz constraint)
|
|
||||||
- Self-attention at 16×16 in both generator and critic
|
|
||||||
|
|
||||||
```python
|
|
||||||
class SelfAttention(nn.Module):
|
|
||||||
def __init__(self, in_ch):
|
|
||||||
super().__init__()
|
|
||||||
mid = max(in_ch // 8, 1)
|
|
||||||
self.q = nn.Conv2d(in_ch, mid, 1, bias=False)
|
|
||||||
self.k = nn.Conv2d(in_ch, mid, 1, bias=False)
|
|
||||||
self.v = nn.Conv2d(in_ch, in_ch, 1, bias=False)
|
|
||||||
self.gamma = nn.Parameter(torch.zeros(1))
|
|
||||||
self._mid = mid
|
|
||||||
|
|
||||||
def forward(self, x):
|
|
||||||
b, c, h, w = x.shape
|
|
||||||
q = self.q(x).view(b, self._mid, -1).transpose(-2, -1)
|
|
||||||
k = self.k(x).view(b, self._mid, -1)
|
|
||||||
v = self.v(x).view(b, c, -1)
|
|
||||||
attn = torch.softmax(q @ k * self._mid ** -0.5, dim=-1)
|
|
||||||
return x + self.gamma * (v @ attn.transpose(-2, -1)).view(b, c, h, w)
|
|
||||||
```
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### 2.4 — Scale to 128×128 *(if 2.3 looks coherent at 64×64)*
|
|
||||||
|
|
||||||
**Because**: 2.3 produces coherent samples at 64×64 → does the architecture hold up at higher resolution?
|
|
||||||
|
|
||||||
Same architecture as 2.3, retrained at 128×128. Add attention at 32×32 if memory permits.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
### Phase 2 Results
|
|
||||||
|
|
||||||
| Step | Model | FID @ 100ep ↓ | Main observed failure | Motivates next step |
|
|
||||||
|---|---|---|---|---|
|
|
||||||
| 2.1 | DCGAN | ? | ? | ? |
|
|
||||||
| 2.2 | WGAN-GP | ? | ? | ? |
|
|
||||||
| 2.3 | WGAN-GP + SN + Attn | ? | ? | ? |
|
|
||||||
| 2.4 | + 128×128 | ? | ? | — |
|
|
||||||
|
|
||||||
For each step: FID curve, 16-sample grid, one paragraph on what failed and why the next change addresses it.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 3 — VAE Track
|
|
||||||
|
|
||||||
**Goal**: A self-contained evolution story for the likelihood-based family. Every step motivated by a known limitation of the previous.
|
|
||||||
|
|
||||||
| Step | Model | Because |
|
|
||||||
|---|---|---|
|
|
||||||
| 3.1 | Vanilla VAE (MSE) | Baseline — expect blur |
|
|
||||||
| 3.2 | + Perceptual loss (VGG) | MSE blur is fundamental to pixel-space reconstruction |
|
|
||||||
| 3.3 | + PatchGAN discriminator (VQGAN-lite) | Perceptual loss still lacks local texture realism |
|
|
||||||
|
|
||||||
**3.1 — Vanilla VAE**: Adam lr=1e-3, latent_dim=256, β=1.0. Plain convolutional encoder/decoder, MSE reconstruction.
|
|
||||||
|
|
||||||
**3.2 — Perceptual loss**: VGG-16 feature matching at relu1_2, relu2_2, relu3_3.
|
|
||||||
|
|
||||||
**3.3 — Patch discriminator**: PatchGAN adversarial loss targeting local texture realism.
|
|
||||||
|
|
||||||
```
|
|
||||||
L = L_mse + λ_perc·L_vgg + λ_adv·L_adv + β·L_kl
|
|
||||||
λ_perc=0.1, λ_adv=0.1, β=0.0001
|
|
||||||
```
|
|
||||||
|
|
||||||
**Decoder fix** (applied from 3.1 onward): replace `ConvTranspose2d` with `Upsample(nearest) + Conv2d` — eliminates checkerboard artifacts.
|
|
||||||
|
|
||||||
| Step | Model | FID ↓ | Main observed failure |
|
|
||||||
|---|---|---|---|
|
|
||||||
| 3.1 | VAE MSE | ? | ? |
|
|
||||||
| 3.2 | + Perceptual | ? | ? |
|
|
||||||
| 3.3 | + PatchGAN | ? | ? |
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 4 — DDPM Track
|
|
||||||
|
|
||||||
**Goal**: A self-contained evolution story for the diffusion family.
|
|
||||||
|
|
||||||
| Step | Model | Because |
|
|
||||||
|---|---|---|
|
|
||||||
| 4.1 | DDPM linear + ε-pred | Baseline |
|
|
||||||
| 4.2 | + cosine schedule | Linear schedule wastes capacity at low timesteps |
|
|
||||||
| 4.3 | + v-prediction | ε-prediction is unstable across the full trajectory |
|
|
||||||
| 4.4 | + wider U-Net / more attention | If 4.3 still underfits |
|
|
||||||
|
|
||||||
**4.1 — Baseline**: AdamW lr=2e-4, base_ch=128, T=1000, attention at 8×8 and 16×16. DDIM sampling, 100 steps.
|
|
||||||
|
|
||||||
**4.2 — Cosine schedule**:
|
|
||||||
|
|
||||||
```python
|
|
||||||
def cosine_betas(T: int, s: float = 0.008):
|
|
||||||
t = torch.linspace(0, T, T + 1)
|
|
||||||
f = torch.cos((t / T + s) / (1 + s) * math.pi / 2) ** 2
|
|
||||||
alpha_bar = f / f[0]
|
|
||||||
betas = 1 - alpha_bar[1:] / alpha_bar[:-1]
|
|
||||||
return betas.clamp(0, 0.999)
|
|
||||||
```
|
|
||||||
|
|
||||||
**4.3 — v-prediction**: replaces ε target with `v = √ᾱ·ε − √(1−ᾱ)·x₀`.
|
|
||||||
|
|
||||||
**4.4 — Wider U-Net**: base_ch 128 → 192, attention at 8×8, 16×16, 32×32.
|
|
||||||
|
|
||||||
| Step | Model | FID ↓ | Main observed failure |
|
|
||||||
|---|---|---|---|
|
|
||||||
| 4.1 | DDPM linear + ε | ? | ? |
|
|
||||||
| 4.2 | + cosine | ? | ? |
|
|
||||||
| 4.3 | + v-pred | ? | ? |
|
|
||||||
| 4.4 | + wider | ? | ? |
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Phase 5 — Cross-Family Comparison
|
|
||||||
|
|
||||||
**Goal**: Side-by-side comparison of the best from each family (2.4, 3.3, 4.4) under identical conditions.
|
|
||||||
|
|
||||||
Best-of-each retrained for 200 epochs at the same resolution and pipeline.
|
|
||||||
|
|
||||||
### 5A — Quantitative
|
|
||||||
|
|
||||||
| Model | FID ↓ | IS ↑ | LPIPS diversity ↑ | Params | Train time |
|
|
||||||
|---|---|---|---|:---:|:---:|
|
|
||||||
| Best GAN (2.4) | ? | ? | ? | ? | ? |
|
|
||||||
| Best VAE (3.3) | ? | ? | ? | ? | ? |
|
|
||||||
| Best DDPM (4.4) | ? | ? | ? | ? | ? |
|
|
||||||
|
|
||||||
### 5B — Qualitative
|
|
||||||
|
|
||||||
- **Visual grids**: 16-image sample grids per finalist
|
|
||||||
- **Progression**: epoch 10 → 50 → 100 → 200 side by side
|
|
||||||
- **Latent interpolation**: smooth transitions between two latent codes (GAN, VAE)
|
|
||||||
- **Diversity**: average pairwise LPIPS distance across 100 generated images
|
|
||||||
- **Failure modes**: worst-generated images per model
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Compute Budget Notes
|
|
||||||
|
|
||||||
Three families × multiple steps is a lot of runs. If compute is tight:
|
|
||||||
|
|
||||||
- **Keep the GAN track complete** (2.1 → 2.4) — it carries the strongest narrative.
|
|
||||||
- **VAE and DDPM can drop the last step each** (stop at 3.2 and 4.3) without hurting the story.
|
|
||||||
- Phase 1 ablations can use 50 epochs instead of 100 — pipeline deltas show up early.
|
|
||||||
|
|
||||||
---
|
|
||||||
|
|
||||||
## Summary
|
|
||||||
|
|
||||||
| Phase | Purpose | Models | Output |
|
|
||||||
|---|---|---|---|
|
|
||||||
| 1 | Pipeline selection | DCGAN @ 64×64 across data variants | Best pipeline |
|
|
||||||
| 2 | GAN evolution (main spine) | DCGAN → WGAN-GP → +SN+Attn → 128×128 | GAN failure→fix narrative |
|
|
||||||
| 3 | VAE evolution | VAE → +Perceptual → +PatchGAN | VAE failure→fix narrative |
|
|
||||||
| 4 | DDPM evolution | DDPM → cosine → v-pred → wider | DDPM failure→fix narrative |
|
|
||||||
| 5 | Cross-family comparison | Best of each, retrained 200ep | Final FID + IS + qualitative |
|
|
||||||
|
|
||||||
**The narrative**: baseline fails in a specific way → fix targets that failure → new failure emerges → next fix targets that → repeat per family → compare families on equal footing.
|
|
||||||
@@ -0,0 +1,19 @@
|
|||||||
|
{
|
||||||
|
"extends": "../phase0/_base_phase0.json",
|
||||||
|
"run_name": "smoke",
|
||||||
|
"model": "vae",
|
||||||
|
"latent_dim": 32,
|
||||||
|
"ngf": 16,
|
||||||
|
"epochs": 1,
|
||||||
|
"batch_size": 8,
|
||||||
|
"lr": 0.001,
|
||||||
|
"beta_kl": 0.1,
|
||||||
|
"lambda_perceptual": 0,
|
||||||
|
"lambda_adversarial": 0,
|
||||||
|
"subsample": 1.0,
|
||||||
|
"sample_interval": 1,
|
||||||
|
"fid_interval": 999,
|
||||||
|
"fid_n_real": 64,
|
||||||
|
"num_workers": 0,
|
||||||
|
"ema_decay": 0.9
|
||||||
|
}
|
||||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because it is too large
Load Diff
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -0,0 +1,961 @@
|
|||||||
|
run,architecture,grid,grid_index,tile_index,row,col,source_path,score,mean,std,saturation,sharpness,exposure_score,contrast_score,detail_score,color_score
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.7545376674909341,0.5586341023445129,0.12475372850894928,0.3098646104335785,0.012272229418158531,0.7542684301733971,0.519807202120622,1.0,0.8154331853515223
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.7593448067256836,0.35332736372947693,0.14404259622097015,0.3831081986427307,0.008653096854686737,0.6041480116546154,0.6001774842540424,0.9921886318123451,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9525878772923821,0.5241501331329346,0.3510676324367523,0.3647458553314209,0.022909104824066162,0.8620308339595795,1.0,1.0,0.959857514030055
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.7095987265826474,0.49222832918167114,0.0913277193903923,0.29169315099716187,0.0034171135630458593,0.9617864713072777,0.3805321641266346,0.7670444106564817,0.7676135552556891
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9141213849186898,0.38839614391326904,0.248654305934906,0.4194767475128174,0.013700177893042564,0.7137379497289658,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9496999841967695,0.47258105874061584,0.33243075013160706,0.27019327878952026,0.03126365691423416,0.9768158085644245,1.0,1.0,0.7110349441829481
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.8351491647723474,0.45492032170295715,0.24781660735607147,0.021942120045423508,0.017632482573390007,0.9216260053217411,1.0,1.0,0.05774242117216712
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.8024060817141282,0.4370657205581665,0.2197197526693344,0.045618437230587006,0.01176534965634346,0.8658303767442703,0.9154989694555601,1.0,0.12004851902786054
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.6557122263077059,0.6841288805007935,0.2226022332906723,0.047703325748443604,0.02228841930627823,0.36209724843502045,0.9275093053778013,1.0,0.12553506775906212
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.7053433787861937,0.3807450532913208,0.19525142014026642,0.010970894247293472,0.018412042409181595,0.6898282915353775,0.8135475839177768,1.0,0.02887077433498282
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9597365334630013,0.4648057818412781,0.21918489038944244,0.42363959550857544,0.012437568977475166,0.952518068253994,0.9132703766226768,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9016706772148609,0.5735244154930115,0.23147985339164734,0.38165542483329773,0.033601272851228714,0.7077362015843391,0.964499389131864,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.9120497883150451,0.4331885576248169,0.3386186361312866,0.26836997270584106,0.01995547115802765,0.8537142425775528,1.0,1.0,0.7062367702785292
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.8598220698535443,0.5386441349983215,0.17184075713157654,0.3957240879535675,0.017268937081098557,0.8167370781302452,0.7160031547149023,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.7806122546362527,0.4119107127189636,0.20364314317703247,0.441266804933548,0.0013961864169687033,0.7872209772467613,0.8485130965709686,0.559568129963735,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0001.png,1,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0001.png,0.8230576974192731,0.44262832403182983,0.26020944118499756,0.020503897219896317,0.011849427595734596,0.8832135125994682,1.0,1.0,0.05395762426288504
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.8998476877808571,0.5348783135414124,0.20103688538074493,0.5457454323768616,0.01619734987616539,0.8285052701830864,0.8376536890864372,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.9812861617654561,0.46003857254981995,0.27224647998809814,0.4570789933204651,0.023630764335393906,0.9376205392181873,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.7420650539086349,0.5928292274475098,0.3344661593437195,0.02122393250465393,0.007521435152739286,0.647408664226532,1.0,0.9578583461869316,0.055852453959615606
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.9720733910014754,0.4618774652481079,0.2704009711742401,0.35229361057281494,0.025064971297979355,0.9433670789003372,1.0,1.0,0.9270884488758288
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.833084571145867,0.4547576606273651,0.31825003027915955,0.017098136246204376,0.030236052349209785,0.921117689460516,1.0,1.0,0.04499509538474836
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.7242124849244168,0.6618639230728149,0.1652834117412567,0.34986764192581177,0.03201981633901596,0.4316752403974533,0.6886808822552364,1.0,0.9207043208573994
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.7384208043400002,0.6048873662948608,0.3096829056739807,0.01394019927829504,0.027470922097563744,0.6097269803285599,1.0,1.0,0.036684734942881686
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.6106346278251241,0.21221114695072174,0.1588122397661209,0.4094204604625702,0.004881285130977631,0.16315983422100555,0.6617176656921705,0.8526855114046854,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.5587454412524638,0.23907819390296936,0.18229030072689056,0.017091788351535797,0.015121573582291603,0.24711935594677936,0.7595429196953773,1.0,0.04497839039877841
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.7776214455188818,0.4174554944038391,0.14888633787631989,0.49546483159065247,0.003931847400963306,0.8045484200119972,0.6203597411513329,0.8005959886795313,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.6441286732110473,0.21925386786460876,0.17080029845237732,0.5611382722854614,0.005940636619925499,0.1851683370769025,0.7116679102182388,0.9003111960900193,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.955621548742056,0.5273370146751404,0.30989617109298706,0.41411760449409485,0.015149565413594246,0.8520718291401863,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.9379227348064122,0.5395410060882568,0.254978746175766,0.36414748430252075,0.01970071718096733,0.8139343559741974,1.0,1.0,0.9582828534276862
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.7201142754209668,0.6283286809921265,0.17947918176651,0.21488603949546814,0.03531326353549957,0.5364728718996048,0.7478299240271251,1.0,0.565489577619653
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.8876708376564477,0.3642846345901489,0.2504919767379761,0.37025678157806396,0.018406979739665985,0.6383894830942154,1.0,1.0,0.9743599515212209
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0002.png,2,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0002.png,0.9283315275452638,0.47205090522766113,0.22538556158542633,0.2635980248451233,0.011152489110827446,0.975159078836441,0.9391065066059431,1.0,0.6936790127503244
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8586176541802989,0.528627336025238,0.2003055214881897,0.36663907766342163,0.004562776070088148,0.8480395749211311,0.8346063395341238,0.8363917125379082,0.9648396780616358
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.7858749721199275,0.26752373576164246,0.22805717587471008,0.5977087020874023,0.016115520149469376,0.3360116742551328,0.950238232811292,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.7536362484587651,0.5852681398391724,0.23954956233501434,0.0073167141526937485,0.03268556296825409,0.6710370630025864,0.9981231763958931,1.0,0.019254510928141445
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.5154432408321509,0.3417609930038452,0.14347687363624573,0.04783596843481064,0.0015792122576385736,0.5680031031370163,0.5978203068176906,0.5872543949069381,0.125884127460028
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8249723659142068,0.4270785450935364,0.18514028191566467,0.23600755631923676,0.015848493203520775,0.8346204534173012,0.7714178413152695,1.0,0.6210725166295704
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8648055293058095,0.3481628894805908,0.2968728244304657,0.35062047839164734,0.012981856241822243,0.5880090296268463,1.0,1.0,0.9226854694517035
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8721946236885333,0.3808496594429016,0.22183451056480408,0.4574553072452545,0.007325959857553244,0.6901551857590675,0.9243104606866837,0.9514197190192317,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.7095855738691413,0.27421942353248596,0.18987368047237396,0.6059479713439941,0.005044713616371155,0.35693569853901874,0.7911403353015583,0.8606510548678727,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.9354947239160538,0.5107810497283936,0.24028459191322327,0.28969162702560425,0.032173462212085724,0.9038092195987701,1.0,1.0,0.7623463869094849
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.7184575937296215,0.5963011384010315,0.1650904268026352,0.18018808960914612,0.022538598626852036,0.6365589424967766,0.6878767783443134,1.0,0.47417918318196345
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8459408931434155,0.4996097683906555,0.131460040807724,0.48573726415634155,0.013038482517004013,0.9387194737792015,0.5477501700321834,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.814499861270924,0.44499677419662476,0.16504618525505066,0.4865788519382477,0.0033741698134690523,0.8906149193644524,0.6876924385627111,0.7640306155710997,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.6293126838085683,0.34668371081352234,0.15933012962341309,0.013006241992115974,0.012032881379127502,0.5833865962922573,0.6638755400975546,1.0,0.03422695261083151
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.6962179163568899,0.7100358009338379,0.2800619900226593,0.1567537486553192,0.02437450736761093,0.2811381220817566,1.0,1.0,0.412509864882419
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8181977607309819,0.3186354637145996,0.21558161079883575,0.5210000276565552,0.017162248492240906,0.4957358241081239,0.8982567116618156,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0003.png,3,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0003.png,0.8527785818227757,0.5013974905014038,0.20327767729759216,0.217881977558136,0.006736705079674721,0.9331328421831131,0.8469903220733007,0.9309423550916126,0.5733736251529894
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.5895350809422575,0.36464041471481323,0.08890166878700256,0.23955386877059937,0.003430658020079136,0.6395012959837914,0.37042361994584405,0.7679874739630475,0.6304049178173667
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8943183845595309,0.5314095616340637,0.2107905000448227,0.3268676996231079,0.02474066987633705,0.8393451198935509,0.8782937501867613,1.0,0.8601781569029155
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8592290036380291,0.48304200172424316,0.12966470420360565,0.40858709812164307,0.018466269597411156,0.9904937446117401,0.540269600848357,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8956946209073067,0.381428062915802,0.23048464953899384,0.3853580951690674,0.02593258023262024,0.6919626966118813,0.9603527064124744,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.7436427799943237,0.3414962887763977,0.1721288114786148,0.5121854543685913,0.004504946526139975,0.5671759024262428,0.7172033811608951,0.8333159796727292,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8115268671265001,0.3547556400299072,0.20915274322032928,0.3197314143180847,0.007749638985842466,0.6086113750934601,0.8714697634180387,0.9651710270531206,0.8413984587318019
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8115414392444642,0.3684893846511841,0.2586754858493805,0.31895413994789124,0.003327572252601385,0.6515293270349503,1.0,0.7607187646181933,0.8393529998628717
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8835286594535176,0.5304690599441528,0.29684287309646606,0.20480328798294067,0.011063450947403908,0.8422841876745224,1.0,1.0,0.5389560210077387
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.9128316894173623,0.387020468711853,0.27525776624679565,0.6243563890457153,0.012326443567872047,0.7094389647245407,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.6316506314551138,0.2844827175140381,0.11516134440898895,0.30810052156448364,0.00884288176894188,0.38900849223136913,0.479838935037454,0.9975111053644723,0.8107908462223253
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8472280944599525,0.4859713912010193,0.24093836545944214,0.007159893400967121,0.022441085427999496,0.9813394024968147,1.0,1.0,0.01884182473938716
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.4586441533185386,0.15696296095848083,0.16091114282608032,0.7654194831848145,0.0007641658885404468,0.0,0.6704630951086681,0.43002089914375247,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.7715456023812294,0.40743064880371094,0.1606440544128418,0.22489489614963531,0.015978895127773285,0.7732207775115967,0.6693502267201742,1.0,0.5918286740779877
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8133123606443405,0.6550119519233704,0.22190885245800018,0.5648695230484009,0.038659438490867615,0.4530876502394676,0.9246202185750008,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.893380181863904,0.4470275938510895,0.1794334501028061,0.488193541765213,0.013053730130195618,0.8969612307846546,0.7476393754283588,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0004.png,4,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0004.png,0.8741027908889871,0.4827801585197449,0.15858162939548492,0.3254881203174591,0.016781821846961975,0.9913120046257973,0.6607567891478539,1.0,0.856547685045945
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.907768871125422,0.448178231716156,0.2107820361852646,0.3144480586051941,0.021893009543418884,0.9005569741129875,0.8782584841052692,1.0,0.8274948910663003
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8039472896997866,0.44701677560806274,0.2252752035856247,0.008296813815832138,0.021059446036815643,0.8969274237751961,0.9386466816067696,1.0,0.02183372056797931
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.9318601943552495,0.5526824593544006,0.25198668241500854,0.3896118402481079,0.032747358083724976,0.772867314517498,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8767036567000966,0.4646499752998352,0.1851481795310974,0.2778030037879944,0.028165467083454132,0.952031172811985,0.7714507480462393,1.0,0.7310605362841958
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.38463021956243315,0.23538875579833984,0.11072921752929688,0.01894732192158699,0.00228615989908576,0.23558986186981212,0.4613717397054037,0.6722501322727574,0.0498613734778605
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.7372267697273276,0.286445677280426,0.2164098471403122,0.46430760622024536,0.0038042094092816114,0.39514274150133144,0.9017076964179676,0.7926865534061516,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8786685809493066,0.6094201803207397,0.2474713772535324,0.38691121339797974,0.023118214681744576,0.5955619364976883,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8696450907737017,0.35869845747947693,0.22669222950935364,0.5497492551803589,0.013579826802015305,0.6209326796233654,0.9445509562889736,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8233137778937817,0.5261629819869995,0.13327325880527496,0.49656835198402405,0.021564697846770287,0.8557406812906265,0.5553052450219791,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.382277699169396,0.16975241899490356,0.07428576797246933,0.6958059668540955,0.001173350028693676,0.03047630935907375,0.3095240332186222,0.521110385584349,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.7702151579907561,0.3096632659435272,0.19988080859184265,0.3563356399536133,0.007513006683439016,0.4676977060735227,0.832836702466011,0.9575841207361948,0.9377253682989823
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.9107583697885275,0.3848089277744293,0.2420244812965393,0.39233115315437317,0.0095355324447155,0.7025278992950916,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.9117341909557581,0.38584980368614197,0.2577420771121979,0.3969506323337555,0.028145231306552887,0.7057806365191936,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.9243011623620987,0.4225029945373535,0.22256368398666382,0.5036839246749878,0.03184037283062935,0.8203218579292297,0.927348683277766,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.6619402099524138,0.26035311818122864,0.16354040801525116,0.531673789024353,0.004902512766420841,0.3136034943163396,0.6814183667302132,0.8537346065537919,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0005.png,5,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0005.png,0.8609141348583063,0.44484877586364746,0.17759594321250916,0.3676582872867584,0.0061057633720338345,0.8901524245738983,0.7399830967187881,0.9069808286923825,0.9675218086493642
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.875421778857708,0.5746668577194214,0.21133756637573242,0.3896825909614563,0.03409885615110397,0.7041660696268082,0.8805731932322185,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.7810890753741003,0.4248238503932953,0.24505583941936493,0.011037391610443592,0.006280225235968828,0.8275745324790478,1.0,0.9138394020840022,0.029045767395904188
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.7865731326373,0.5845127105712891,0.3116176724433899,0.0875362902879715,0.022526144981384277,0.6733977794647217,1.0,1.0,0.23035865865255656
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.8591934063749928,0.3876144587993622,0.21951551735401154,0.32111066579818726,0.00818733312189579,0.7112951837480068,0.9146479889750481,0.9786249774983253,0.8450280678899664
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.6532358039510788,0.3147972822189331,0.11174239218235016,0.70570969581604,0.005324860103428364,0.48374150693416607,0.46559330075979233,0.8737414465715652,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.7858239174281296,0.3296286463737488,0.24441830813884735,0.19455255568027496,0.01192161999642849,0.5300895199179649,1.0,1.0,0.5119804096849341
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.8589211200609018,0.47416725754737854,0.318678081035614,0.03645293414592743,0.0220384132117033,0.9817726798355579,1.0,1.0,0.09592877406823007
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.9243404397446858,0.4492695927619934,0.2036893665790558,0.3762975037097931,0.014060743153095245,0.9039674773812294,0.8487056940793991,1.0,0.9902565887099818
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.8971401409883248,0.5035743713378906,0.17587903141975403,0.378460556268692,0.016842877492308617,0.9263300895690918,0.7328292975823085,1.0,0.9959488322860316
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.7544976327801437,0.36918866634368896,0.14592470228672028,0.5229775905609131,0.006029176525771618,0.653714582324028,0.6080195928613346,0.9039095208981394,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.6670329091978492,0.24327167868614197,0.1800692081451416,0.388423889875412,0.004938778933137655,0.26022399589419376,0.7502883672714233,0.8555168009926564,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.7846233867108079,0.3194371163845062,0.20846299827098846,0.5017948150634766,0.005891444161534309,0.49824098870158207,0.8685958261291187,0.8982893690463906,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.6997036429714174,0.37598657608032227,0.12620475888252258,0.278771311044693,0.006379921920597553,0.6749580502510071,0.5258531620105108,0.9176758892065436,0.7336087132755079
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.9343138402229861,0.5160537958145142,0.24139896035194397,0.29922282695770264,0.009966598823666573,0.8873318880796432,1.0,1.0,0.7874284919939543
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.8053720891475677,0.394361674785614,0.1485264152288437,0.3891766667366028,0.012834908440709114,0.7323802337050438,0.6188600634535154,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0006.png,6,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0006.png,0.43927690061584435,0.15315912663936615,0.08104534447193146,0.5397039651870728,0.0032062034588307142,0.0,0.33768893529971444,0.75188088010372,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8493201643228531,0.4572209119796753,0.17145398259162903,0.2694404721260071,0.012126946821808815,0.9288153499364853,0.7143915941317877,1.0,0.7090538740158081
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.769081329384276,0.30254387855529785,0.2055985927581787,0.39737173914909363,0.006279023829847574,0.4454496204853059,0.856660803159078,0.9137928091638432,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8151628826815037,0.5378580689430237,0.1702156662940979,0.40093761682510376,0.004380045458674431,0.819193534553051,0.7092319428920746,0.826540957791864,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.6111208545718048,0.591734766960144,0.1315007209777832,0.0037906200159341097,0.02074606902897358,0.6508288532495499,0.5479196707407634,1.0,0.009975315831405552
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.646148418909625,0.7471302151679993,0.20146562159061432,0.2400357872247696,0.0406423881649971,0.1652180776000023,0.839440089960893,1.0,0.6316731242757094
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8904819957964021,0.40772008895874023,0.3016814589500427,0.32071495056152344,0.006617442704737186,0.7741252779960632,1.0,0.9265856223879277,0.843986712004009
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.7260528919725296,0.3227294385433197,0.1426870971918106,0.49903637170791626,0.008251594379544258,0.5085294954478741,0.5945295716325443,0.9805406873936162,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.7036572464083265,0.35407277941703796,0.22536706924438477,0.009349950589239597,0.0071435365825891495,0.6064774356782436,0.9390294551849365,0.9452576367197429,0.024605133129577888
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.7750198364946898,0.5274783372879028,0.16703392565250397,0.241154283285141,0.0050767576321959496,0.8516301959753036,0.6959746902187666,0.8621835615693362,0.6346165349608973
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8794440545141697,0.351406991481781,0.2869923710823059,0.5339280366897583,0.024457525461912155,0.5981468483805656,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8372685394396907,0.3613077402114868,0.21205411851406097,0.33813637495040894,0.009057862684130669,0.6290866881608963,0.8835588271419208,1.0,0.8898325656589708
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8015177220302192,0.390240341424942,0.15443000197410583,0.3613290786743164,0.011990748345851898,0.7195010669529438,0.6434583415587743,1.0,0.950865996511359
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.8176741555725273,0.40279310941696167,0.18400675058364868,0.27878618240356445,0.01006361935287714,0.7587284669280052,0.7666947940985362,1.0,0.7336478484304327
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.9672669764608145,0.48784127831459045,0.21969453990459442,0.4469306170940399,0.012913118116557598,0.9754960052669048,0.9153939162691435,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.9368669483810663,0.4143889248371124,0.23870186507701874,0.4027857184410095,0.010187076404690742,0.7949653901159763,0.9945911044875781,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0007.png,7,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0007.png,0.780505416539763,0.34109261631965637,0.19911116361618042,0.5398318767547607,0.004775059409439564,0.5659144259989262,0.8296298484007518,0.8473685368794388,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7970287408912766,0.5013858675956726,0.22536815702915192,0.011146023869514465,0.006544208154082298,0.9331691637635231,0.9390339876214664,0.9238721968459908,0.029331641761880172
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7108458863424235,0.43045708537101746,0.16398027539253235,0.005869795568287373,0.011441962793469429,0.8451783917844296,0.6832511474688848,1.0,0.015446830442861506
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7697381906155869,0.4496918022632599,0.11186768114566803,0.5492039918899536,0.004504089243710041,0.9052868820726871,0.46611533810695016,0.833270098246783,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.6989638672105051,0.35959097743034363,0.13119755685329437,0.5479761362075806,0.0037838509306311607,0.6237218044698238,0.5466564868887266,0.7914015192117597,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.8932776227593422,0.5262240171432495,0.22980892658233643,0.25169041752815247,0.01662326045334339,0.8555499464273453,0.9575371940930685,1.0,0.6623432040214539
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.6088977974402533,0.3927351236343384,0.14018476009368896,0.005774964112788439,0.00488344207406044,0.7272972613573074,0.5841031670570374,0.8527923112751862,0.015197273981022207
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.882893174082825,0.40290549397468567,0.22173728048801422,0.32426077127456665,0.014655661769211292,0.7590796686708927,0.9239053353667259,1.0,0.8533178191435964
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7554918011846511,0.5855017304420471,0.2542382478713989,0.011145839467644691,0.01503431424498558,0.6703070923686028,1.0,1.0,0.02933115649380182
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.5396268888817806,0.844924807548523,0.2291971743106842,0.007930399850010872,0.05406677722930908,0.0,0.9549882262945175,1.0,0.020869473289502293
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.8613160230219364,0.5235291719436646,0.16169969737529755,0.40834808349609375,0.019339991733431816,0.8639713376760483,0.6737487390637398,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7321886953701707,0.6097853779792786,0.31042152643203735,0.009784967638552189,0.01916937530040741,0.5944206938147545,1.0,1.0,0.025749914838295234
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.8331833370029926,0.33568674325942993,0.21478161215782166,0.40492361783981323,0.015957213938236237,0.5490210726857185,0.8949233839909236,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.5088473971517982,0.12006480991840363,0.10350232571363449,0.40002232789993286,0.006385215558111668,0.0,0.43125969047347706,0.9178779600390201,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.7030484216346921,0.2965622544288635,0.1651538461446762,0.3865181505680084,0.005337495356798172,0.42675704509019863,0.6881410256028175,0.8743160017071486,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.8323472074380046,0.39097630977630615,0.21305964887142181,0.2520219683647156,0.0095682917162776,0.7218009680509567,0.8877485369642576,1.0,0.6632157062229357
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0008.png,8,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0008.png,0.5792033701237205,0.18457084894180298,0.1619434654712677,0.517056941986084,0.0041741495952010155,0.07678390294313442,0.6747644394636154,0.8149554696067822,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8595949189116271,0.485787957906723,0.1843118816614151,0.298603355884552,0.005179741885513067,0.9819126315414906,0.7679661735892296,0.8670461265140358,0.7857983049593473
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.6028219508007169,0.19069211184978485,0.13923847675323486,0.437399297952652,0.01030984427779913,0.09591284953057777,0.5801603198051453,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.7336863054232182,0.5513945817947388,0.1982884407043457,0.006987376604229212,0.012560537084937096,0.7768919318914413,0.8262018362681072,1.0,0.018387833169024242
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.6670747307930264,0.6282001733779907,0.10933477431535721,0.3718968629837036,0.0056978557258844376,0.536874458193779,0.4555615596473217,0.890170128630621,0.9786759552202726
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.4287850607902928,0.2971642017364502,0.09910248965024948,0.008129455149173737,0.002492319094017148,0.42863813042640697,0.4129270402093729,0.6924260565837508,0.021393303024141413
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8743150602045812,0.4599578380584717,0.2946867346763611,0.10919828712940216,0.02124294824898243,0.937368243932724,1.0,1.0,0.2873639134984267
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8105629876452056,0.4211808145046234,0.1557983011007309,0.3064271807670593,0.012138865888118744,0.8161900453269482,0.6491595879197121,1.0,0.8063873178080508
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.804760357855182,0.5664476752281189,0.1747012436389923,0.2974855303764343,0.023043829947710037,0.7298510149121284,0.7279218484958013,1.0,0.7828566588853535
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.6948749497532846,0.2902379035949707,0.13822153210639954,0.44017231464385986,0.010719917714595795,0.40699344873428356,0.5759230504433315,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8288026563823223,0.34265631437301636,0.20604988932609558,0.4261782765388489,0.010843470692634583,0.5708009824156761,0.8585412055253983,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.9331365078687668,0.4384240508079529,0.21769116818904877,0.46462100744247437,0.013590224087238312,0.8700751587748528,0.9070465341210365,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.9031721539795399,0.44615936279296875,0.18791820108890533,0.521428644657135,0.009995403699576855,0.8942480087280273,0.7829925045371056,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8932308070361614,0.41031843423843384,0.20684581995010376,0.4193262457847595,0.024767670780420303,0.7822451069951057,0.8618575831254324,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.7570139667723568,0.2662753164768219,0.23317767679691315,0.5077286958694458,0.005107289180159569,0.33211036399006855,0.9715736533204715,0.863635046316779,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8732785806059837,0.35115379095077515,0.23525752127170563,0.41009122133255005,0.011141548864543438,0.5973555967211723,0.9802396719654402,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0009.png,9,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0009.png,0.8602951680751223,0.4846940040588379,0.1919434517621994,0.18940842151641846,0.009361225180327892,0.9853312373161316,0.7997643823424976,1.0,0.49844321451689066
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8469079197629502,0.5111610889434814,0.15641345083713531,0.3308650553226471,0.012622418813407421,0.9026215970516205,0.6517227118213972,1.0,0.870697514006966
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.7822690353141096,0.33351975679397583,0.18741919100284576,0.6403491497039795,0.007028179243206978,0.5422492399811745,0.7809132958451908,0.9412810982648001,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.6131865660610952,0.6606355309486389,0.15420351922512054,0.10077087581157684,0.018221400678157806,0.4355139657855034,0.6425146634380023,1.0,0.26518651529362325
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8051183959156577,0.46463191509246826,0.22866183519363403,0.009745027869939804,0.006425432860851288,0.9519747346639633,0.9527576466401418,0.9194078399872734,0.025644810184052114
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.7481512544941831,0.5874038934707642,0.26845401525497437,0.010588700883090496,0.008188189938664436,0.664362832903862,1.0,0.9786506170977449,0.027865002323922358
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8529746122658253,0.4173896312713623,0.1693374663591385,0.47458702325820923,0.010883152484893799,0.8043425977230072,0.7055727764964104,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8588580984426172,0.479427695274353,0.35124045610427856,0.02379973977804184,0.07146435976028442,0.9982115477323532,1.0,1.0,0.06263089415274169
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.7668434718721792,0.5449010133743286,0.14114394783973694,0.2565208673477173,0.00970851257443428,0.7971843332052231,0.5880997826655706,1.0,0.6750549140729403
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.7776098706220326,0.5767411589622498,0.15460778772830963,0.31678059697151184,0.020487023517489433,0.6976838782429695,0.6441991155346235,1.0,0.8336331499250311
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.9174275484328207,0.4640711843967438,0.21460025012493134,0.2890799343585968,0.01801297813653946,0.9502224512398243,0.8941677088538806,1.0,0.7607366693647284
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.7589390004741559,0.5512765049934387,0.17035728693008423,0.16556505858898163,0.008570555597543716,0.777260921895504,0.709822028875351,0.9898379474100473,0.4356975226025832
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8784318123352023,0.4552675485610962,0.2071256935596466,0.3053433895111084,0.005664038006216288,0.9227110892534256,0.8630237231651943,0.8887243331051369,0.8035352355555484
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.8947002198547125,0.43727806210517883,0.18780162930488586,0.5342007875442505,0.025740642100572586,0.8664939440786839,0.7825067887703578,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.9049478150904179,0.4964229464530945,0.17627546191215515,0.4802337884902954,0.017999855801463127,0.9486782923340797,0.7344810913006465,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.9813835850671718,0.4987140893936157,0.2715086340904236,0.37728437781333923,0.025151735171675682,0.9415184706449509,1.0,1.0,0.9928536258245769
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0010.png,10,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0010.png,0.716300159169756,0.23789291083812714,0.21230025589466095,0.4700593948364258,0.006222758442163467,0.2434153463691474,0.8845843995610874,0.9116009415627422,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.8617069907486439,0.39766180515289307,0.1911192387342453,0.3902971148490906,0.022736594080924988,0.7426931411027908,0.7963301613926888,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.9391638579141153,0.4860605299472809,0.2516518831253052,0.2402755320072174,0.09339642524719238,0.9810608439147472,1.0,1.0,0.6323040315979406
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.8260683723746457,0.3666582703590393,0.21132534742355347,0.47827720642089844,0.005301555152982473,0.6458070948719978,0.8805222809314728,0.8726782385344182,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.7919605132192373,0.390259712934494,0.14087362587451935,0.4148794710636139,0.010153334587812424,0.7195616029202938,0.5869734411438307,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.6491830306928028,0.2559933364391327,0.14960849285125732,0.364663302898407,0.0062568290159106255,0.29997917637228977,0.6233687202135723,0.9129304843970083,0.9596402707852815
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.6657208744436504,0.27518025040626526,0.12619151175022125,0.5309655666351318,0.009614264592528343,0.35993828251957904,0.525797965625922,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.7608880448498223,0.37905919551849365,0.17089742422103882,0.2328089475631714,0.015220238827168941,0.6845599859952927,0.7120726009209951,1.0,0.6126551251662404
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.7145841657723252,0.2859361171722412,0.17513622343540192,0.3232502341270447,0.013347038067877293,0.3935503661632539,0.7297342643141747,1.0,0.8506585108606439
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.8596613200479433,0.3479235768318176,0.26016852259635925,0.3381568491458893,0.013945198617875576,0.5872611775994301,1.0,1.0,0.8898864451207612
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.6983062481439671,0.2997108995914459,0.15603455901145935,0.36176618933677673,0.006386489141732454,0.4365965612232686,0.6501439958810806,0.9179265514136339,0.9520162877283598
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.8818235804557416,0.47700607776641846,0.15765540301799774,0.4270445704460144,0.007290821056813002,0.9906439930200577,0.6568975125749906,0.9502445151089086,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.774190147260302,0.5137311220169067,0.20070448517799377,0.012495584785938263,0.013682609423995018,0.8945902436971664,0.8362686882416408,1.0,0.03288311785773227
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.48064235309765546,0.14640486240386963,0.1200752854347229,0.8573397397994995,0.002828537719324231,0.0,0.5003136893113455,0.7221929852170071,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.5637109845914624,0.18502452969551086,0.15317346155643463,0.652617335319519,0.0038432846777141094,0.07820165529847156,0.6382227564851444,0.7951346442255104,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.717459922656417,0.22427037358283997,0.20576515793800354,0.6185629367828369,0.009624357335269451,0.200844917446375,0.8573548247416815,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0011.png,11,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0011.png,0.8127584176343619,0.34211820363998413,0.20422905683517456,0.370963454246521,0.0076253050938248634,0.5691193863749504,0.850954403479894,0.9612133528487059,0.9762196164382131
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.6864065705368491,0.2741782069206238,0.14817701280117035,0.5917714834213257,0.008119042962789536,0.3568068966269494,0.6174042200048765,0.9765729421892052,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9713950715959072,0.47971469163894653,0.21733003854751587,0.40844476222991943,0.010058732703328133,0.9991084113717079,0.9055418272813162,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.8162724675512627,0.4501749575138092,0.23194971680641174,0.010883957147598267,0.022110294550657272,0.9067967422306538,0.966457153360049,1.0,0.028641992493679647
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.8952369228397545,0.3791632056236267,0.247663214802742,0.35408759117126465,0.018146997317671776,0.6848850175738335,1.0,1.0,0.9318094504506964
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.8108695181944456,0.43630632758140564,0.23908624053001404,0.007535489741712809,0.02226063422858715,0.8634572736918926,0.9961926688750585,1.0,0.019830236162402128
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9931557374587933,0.4784943461418152,0.26595890522003174,0.5266944169998169,0.008175451308488846,0.9952948316931725,1.0,0.9782691518033664,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9205608076170871,0.5526833534240723,0.2633109390735626,0.351377010345459,0.016436833888292313,0.7728645205497742,1.0,1.0,0.9246763430143657
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.7959507714801258,0.3819081783294678,0.15686550736427307,0.6137540936470032,0.007817976176738739,0.6934630572795868,0.6536062806844711,0.9673198803636337,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.79812374677073,0.30458009243011475,0.25100991129875183,0.42348554730415344,0.004833739250898361,0.4518127888441087,1.0,0.8503196404699894,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9081090591847897,0.5377892255783081,0.2098291665315628,0.6009770035743713,0.01834665611386299,0.8194086700677872,0.8742881938815117,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9247215962723682,0.500356912612915,0.22976577281951904,0.2700507640838623,0.012169701978564262,0.9363846480846405,0.9573573867479961,1.0,0.7106599054838482
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9528578683341804,0.4682837128639221,0.22772490978240967,0.3272705674171448,0.02386489138007164,0.9633866026997566,0.9488537907600403,1.0,0.8612383353082758
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.862411005422473,0.4177171289920807,0.17664095759391785,0.5214425921440125,0.015117242932319641,0.8053660281002522,0.7360039899746578,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9598501920700073,0.45587438344955444,0.22597436606884003,0.4988783597946167,0.020012032240629196,0.9246074482798576,0.9415598586201668,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.8650946329102704,0.37269696593284607,0.2405259609222412,0.29308444261550903,0.028306839987635612,0.664678018540144,1.0,1.0,0.7712748489881817
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.7886823602020742,0.3391212821006775,0.17660492658615112,0.3878621459007263,0.014586799778044224,0.5597540065646172,0.7358538607756298,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.7800788427911904,0.31115302443504333,0.23668862879276276,0.4070294499397278,0.003460435662418604,0.47235320135951053,0.9862026199698448,0.7700483855695351,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.9391630170745662,0.5017827749252319,0.23649747669696808,0.34262484312057495,0.00630668830126524,0.9319288283586502,0.9854061529040337,0.9148634963821362,0.9016443240015131
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.7712428817574523,0.5294139981269836,0.21210214495658875,0.006183420307934284,0.025031905621290207,0.8455812558531761,0.8837589373191198,1.0,0.01627215870509022
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8566811236111742,0.6218043565750122,0.2461337447166443,0.3537108600139618,0.01103892270475626,0.5568613857030869,1.0,1.0,0.9308180526683205
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8376682562263388,0.3607521057128906,0.22964757680892944,0.28475600481033325,0.014505776576697826,0.6273503303527832,0.9568649033705394,1.0,0.7493579073956138
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.9891186870634555,0.4683932662010193,0.25315365195274353,0.42203789949417114,0.012284314259886742,0.9637289568781853,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.5655229001103006,0.5717395544052124,0.14466263353824615,0.024601956829428673,0.002023695269599557,0.7133138924837112,0.6027609730760257,0.6439565667756836,0.06474199165639125
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8070361636579038,0.2997651696205139,0.23822307586669922,0.3248429298400879,0.024911997839808464,0.4367661550641061,0.9925961494445801,1.0,0.8548498153686523
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8654121831059456,0.4027549624443054,0.1902635246515274,0.5719152688980103,0.012333137914538383,0.7586092576384544,0.7927646860480309,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.7781528002965323,0.3499597907066345,0.20828506350517273,0.29097089171409607,0.005918635055422783,0.5936243459582329,0.867854431271553,0.8994089447512873,0.7657128729318318
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.9213138908147812,0.5233176946640015,0.20953938364982605,0.47362884879112244,0.015652017667889595,0.8646322041749954,0.8730807652076086,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.4911369412626021,0.1367371678352356,0.11293863505125046,0.7546923160552979,0.003919710870832205,0.0,0.4705776460468769,0.799854589794156,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.5426214516948177,0.1607101410627365,0.16714097559452057,0.3070922791957855,0.0047724274918437,0.0022191908210517086,0.6964207316438358,0.8472353537227971,0.8081375768310145
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8714720599573349,0.4545878469944,0.16664092242717743,0.43773460388183594,0.007223552092909813,0.9205870218575001,0.694337176779906,0.9479792014644525,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.5526559902011956,0.8025230169296265,0.2603263556957245,0.00672850850969553,0.06973238289356232,0.0,1.0,1.0,0.017706601341304026
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.8802813161164522,0.4259483516216278,0.18476378917694092,0.5297917127609253,0.021040260791778564,0.8310885988175869,0.7698491215705872,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7407415405785639,0.5424097776412964,0.15462270379066467,0.39153799414634705,0.0018547483487054706,0.8049694448709488,0.6442612657944362,0.6238893095157935,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7587375699248361,0.4254619777202606,0.20579728484153748,0.006638244725763798,0.012493796646595001,0.8295686803758144,0.8574886868397396,1.0,0.017469065067799466
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7731683049350978,0.29915544390678406,0.1941680610179901,0.5130757689476013,0.013549610041081905,0.4348607622087003,0.8090335875749588,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.8925165824592114,0.36535102128982544,0.2630952000617981,0.44784021377563477,0.01910003088414669,0.6417219415307045,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.9000832740805651,0.5424793362617493,0.28022435307502747,0.27526605129241943,0.03489815443754196,0.8047520741820335,1.0,1.0,0.7243843455063669
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.8814360807208639,0.5581095814704895,0.2190711796283722,0.33142292499542236,0.021359482780098915,0.7559075579047203,0.9127965817848842,1.0,0.8721655920932168
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7512215759115,0.5901047587394714,0.32564619183540344,0.011260127648711205,0.014131704345345497,0.6559226289391518,1.0,1.0,0.029631914865029484
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.9090647824108601,0.576175332069397,0.23938332498073578,0.4051344394683838,0.023894552141427994,0.6994520872831345,0.9974305207530658,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7038594809605887,0.38715076446533203,0.1900695115327835,0.008407499641180038,0.014668822288513184,0.7098461389541626,0.7919562980532646,1.0,0.022124999055736942
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.8055793151259423,0.2726179361343384,0.2450747936964035,0.6575278043746948,0.015183239243924618,0.35193105041980755,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.6873617286133359,0.2956738770008087,0.18449264764785767,0.2508673071861267,0.006496814079582691,0.42398086562752735,0.768719365199407,0.9221003629566118,0.6601771241740176
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.800081877018276,0.3557407259941101,0.20855067670345306,0.268246054649353,0.01240632589906454,0.6116897687315941,0.8689611529310545,1.0,0.7059106701298764
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.7996633984148503,0.28442543745040894,0.2264116406440735,0.42042145133018494,0.01109558530151844,0.38882949203252803,0.9433818360169729,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.8768068260268161,0.5338320732116699,0.26701149344444275,0.1957617998123169,0.024476852267980576,0.8317747712135315,1.0,1.0,0.5151626310850445
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.9009054427393594,0.430291086435318,0.26297423243522644,0.4992072582244873,0.003762240521609783,0.8446596451103687,1.0,0.7900301968249951,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0014.png,14,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0014.png,0.8039119759084362,0.38075992465019226,0.1704365760087967,0.3392230272293091,0.01465622428804636,0.6898747645318508,0.7101524000366529,1.0,0.8926921769192344
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.7432192665724952,0.24435082077980042,0.22781600058078766,0.41463130712509155,0.006374833174049854,0.2635963149368764,0.949233335753282,0.9174814854617903,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.8307682323612664,0.45102840662002563,0.1656930297613144,0.255392462015152,0.016262967139482498,0.9094637706875801,0.6903876240054767,1.0,0.672085426355663
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.7878784725540563,0.5474376678466797,0.16663503646850586,0.23511230945587158,0.013947145082056522,0.789257287979126,0.6943126519521078,1.0,0.618716603831241
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.9247153528034686,0.40815919637680054,0.24393706023693085,0.3599008023738861,0.01962270960211754,0.7754974886775017,1.0,1.0,0.9471073746681213
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.5864899270236492,0.6555646061897278,0.1594225913286209,0.0045688822865486145,0.018462948501110077,0.4513606056571007,0.6642607972025871,1.0,0.012023374438285828
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.7942999303340912,0.3399207592010498,0.1804993748664856,0.5016100406646729,0.011573918163776398,0.5622523725032806,0.7520807286103567,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.8627199716866016,0.40566039085388184,0.18593068420886993,0.430484801530838,0.02167251892387867,0.7676887214183807,0.7747111842036247,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.8996455028653145,0.37295520305633545,0.24502159655094147,0.5237715244293213,0.015282982960343361,0.6654850095510483,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.975370334677006,0.4722314476966858,0.3008243143558502,0.3360551595687866,0.03294828534126282,0.9757232740521431,1.0,1.0,0.8843556830757543
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.9019431434571743,0.4112498164176941,0.21311715245246887,0.39880985021591187,0.03857170417904854,0.785155676305294,0.8879881352186203,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.9076573254834663,0.4421848952770233,0.3213356137275696,0.23587609827518463,0.011420232243835926,0.8818277977406979,1.0,1.0,0.6207265744083806
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.7160310596227647,0.3161490559577942,0.13571305572986603,0.4689984619617462,0.009724211879074574,0.48796579986810695,0.5654710655411085,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.8484421701219521,0.4870491921901703,0.2939058542251587,0.012795329093933105,0.0162888765335083,0.9779712744057178,1.0,1.0,0.03367191866824502
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.6716685353196115,0.41652441024780273,0.09809814393520355,0.3237360119819641,0.0028388802893459797,0.8016387820243835,0.4087422663966815,0.7230547589911194,0.8519368736367476
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.6985199927401385,0.622435450553894,0.21625640988349915,0.029722878709435463,0.017423739656805992,0.5548892170190811,0.9010683745145798,1.0,0.07821810186693542
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0015.png,15,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0015.png,0.8383523071727896,0.4447466731071472,0.15891675651073456,0.3717041611671448,0.006034497171640396,0.889833353459835,0.6621531521280607,0.9041241148796655,0.9781688451766968
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7467232157133128,0.31056469678878784,0.20241448283195496,0.2597951292991638,0.018338143825531006,0.4705146774649621,0.8433936784664791,1.0,0.6836713928925363
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.8761748644866441,0.4614783525466919,0.186544269323349,0.27957504987716675,0.014470174908638,0.9421198517084122,0.7772677888472875,1.0,0.7357238154662282
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.6506829364525222,0.5831090211868286,0.1323879361152649,0.18907222151756287,0.004423442296683788,0.6777843087911606,0.5516164004802704,0.8289158080776936,0.497558477677797
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.8203354813158512,0.404049813747406,0.15323102474212646,0.44978874921798706,0.017404763028025627,0.7626556679606438,0.6384626030921936,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.8569143503904343,0.5433372259140015,0.1730343997478485,0.4627038836479187,0.019491128623485565,0.8020711690187454,0.7209766656160355,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.8411121944847859,0.5319052934646606,0.1652536392211914,0.3374561071395874,0.022706329822540283,0.8377959579229355,0.6885568300882976,1.0,0.8880423872094405
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.884892939325226,0.44047823548316956,0.17845477163791656,0.37715286016464233,0.02034679800271988,0.8764944858849049,0.7435615484913191,1.0,0.9925075267490587
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7285709449969338,0.322512686252594,0.1645842343568802,0.3918088674545288,0.005507936701178551,0.5078521445393562,0.6857676431536674,0.8819400347561066,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7355873623919291,0.5233707427978516,0.18375299870967865,0.17006665468215942,0.0027751706074923277,0.8644664287567139,0.765637494623661,0.717698444644699,0.44754382811094584
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.9084798227015294,0.5499968528747559,0.2278805673122406,0.3527696132659912,0.01753823831677437,0.7812598347663879,0.9495023638010025,1.0,0.9283410875420821
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7982093503130805,0.34486597776412964,0.18317990005016327,0.4712831676006317,0.008358328603208065,0.5777061805129051,0.763249583542347,0.9836904843860194,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7055575221973029,0.34307751059532166,0.19303835928440094,0.10798183083534241,0.01715138927102089,0.5721172206103802,0.8043264970183372,1.0,0.28416271272458526
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7802756648118558,0.3692881166934967,0.1473008245229721,0.37985312938690186,0.011139638721942902,0.6540253646671772,0.6137534355123838,1.0,0.9996134983865839
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.8073003645986319,0.3883167803287506,0.15460270643234253,0.43305689096450806,0.012513567693531513,0.7134899385273457,0.6441779434680939,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.7134816550796753,0.35065758228302,0.223502978682518,0.013582335785031319,0.025194358080625534,0.5958049446344376,0.9312624111771584,1.0,0.035742988907977155
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0016.png,16,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0016.png,0.6053562955771398,0.272235631942749,0.21901960670948029,0.013670315966010094,0.005551657639443874,0.3507363498210908,0.9125816946228346,0.8838588195558328,0.03597451570002656
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.8000689573536971,0.5396853685379028,0.3299698531627655,0.015260775573551655,0.05055173859000206,0.8134832233190536,1.0,1.0,0.04015993571987277
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.6530644088957033,0.27247026562690735,0.1181761622428894,0.45654022693634033,0.008918961510062218,0.3514695800840856,0.4924006760120392,0.9996133282674634,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.7901423561428936,0.38881272077560425,0.18913355469703674,0.5581117868423462,0.0032739692833274603,0.7150397524237633,0.7880564779043198,0.7568539481778749,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.9169885468326117,0.3919905424118042,0.3154178261756897,0.3787267804145813,0.025505347177386284,0.7249704450368881,1.0,1.0,0.9966494221436349
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.8500145824528054,0.4204067289829254,0.164823979139328,0.37962836027145386,0.010730253532528877,0.8137710280716419,0.6867665797472,1.0,0.9990220007143522
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.8821932227595857,0.3858415484428406,0.28026506304740906,0.30518248677253723,0.020398907363414764,0.7057548388838768,1.0,1.0,0.8031118072961506
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.765615213662386,0.2896384596824646,0.1952633261680603,0.4197927713394165,0.01130150817334652,0.405120186507702,0.813597192366918,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.8775255475572729,0.40311914682388306,0.20109966397285461,0.424687922000885,0.008678479120135307,0.7597473338246346,0.837915266553561,0.992907069775257,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.1060203865223869,0.08488570153713226,0.06521442532539368,0.43964800238609314,0.00011834290489787236,0.0,0.2717267721891403,0.13413173859690072,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.7286023513266915,0.6903868913650513,0.21062985062599182,0.28513363003730774,0.029533270746469498,0.3425409644842148,0.8776243776082993,1.0,0.7503516579929151
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.8902436938725019,0.4110713601112366,0.21024318039417267,0.35988613963127136,0.03376898914575577,0.7845980003476143,0.8760132516423862,1.0,0.9470687885033456
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.7448399855155,0.35615774989128113,0.17505337297916412,0.276083767414093,0.006782068405300379,0.6129929684102535,0.7293890540798506,0.9325797770516233,0.7265362300370869
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.6767499036770002,0.2654770612716675,0.17257435619831085,0.4880240261554718,0.00479924026876688,0.329615816473961,0.7190598174929619,0.8485888539476931,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.9328115202486514,0.40833228826522827,0.24062082171440125,0.5199904441833496,0.01580057665705681,0.7760384008288383,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.9405414138773553,0.4455060660839081,0.24673490226268768,0.31129467487335205,0.014735187403857708,0.8922064565122128,1.0,1.0,0.8191965128246107
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0017.png,17,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0017.png,0.7440665274884055,0.4705606698989868,0.1124984622001648,0.25925108790397644,0.00462822150439024,0.9705020934343338,0.46874359250068665,0.8398274638253193,0.6822397050104643
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.9087680444121361,0.4924779534339905,0.17637290060520172,0.40666013956069946,0.01816350594162941,0.9610063955187798,0.7348870858550072,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8488528393650132,0.4742929935455322,0.27767375111579895,0.010648000054061413,0.012928320094943047,0.9821656048297882,1.0,1.0,0.028021052773845822
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.846041242287397,0.42204126715660095,0.1783185452222824,0.3279097080230713,0.008652274496853352,0.818878959864378,0.7429939384261768,0.9921653206381781,0.8629202842712402
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.9666911387914106,0.47083932161331177,0.23597531020641327,0.3301190137863159,0.030412841588258743,0.9713728800415993,0.9832304591933887,1.0,0.8687342468060945
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8631674908101559,0.39369702339172363,0.19526122510433197,0.6157183647155762,0.01187051273882389,0.7303031980991364,0.8135884379347166,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.7551792438289052,0.45202067494392395,0.1635371297597885,0.06837073713541031,0.015932418406009674,0.9125646091997623,0.6814047073324522,1.0,0.17992299246160606
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8884607032725685,0.452422559261322,0.20240382850170135,0.28198474645614624,0.016156919300556183,0.9138204976916313,0.8433492854237556,1.0,0.7420651222530165
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.6955662948021785,0.45507293939590454,0.07398809492588043,0.45716890692710876,0.0026384503580629826,0.9221029356122017,0.30828372885783517,0.7058011818446697,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8950493446698313,0.4675564169883728,0.24620762467384338,0.14367851614952087,0.010817540809512138,0.961113803088665,1.0,1.0,0.37810135828821284
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.9684420607984067,0.513661801815033,0.2601226568222046,0.4532388746738434,0.030447103083133698,0.894806869328022,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.6384765812863292,0.4098794758319855,0.10772261768579483,0.6401137113571167,0.0009622080833651125,0.7808733619749546,0.4488442403574785,0.4782452023463975,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8459881986750849,0.47413957118988037,0.1435789167881012,0.4443722367286682,0.005647978745400906,0.9816861599683762,0.5982454866170883,0.8880348187977822,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.9159632013816583,0.5285569429397583,0.30137500166893005,0.2824295163154602,0.018762830644845963,0.8482595533132553,1.0,1.0,0.743235569251211
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.8871443532407284,0.3699425458908081,0.23225857317447662,0.45730137825012207,0.010098276659846306,0.6560704559087753,0.9677440548936527,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.6971903395035225,0.31409573554992676,0.14680181443691254,0.3100327253341675,0.008484655991196632,0.48154917359352123,0.6116742268204689,0.9873679217265111,0.8158755929846513
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0018.png,18,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0018.png,0.7604554773749489,0.41306376457214355,0.2126333862543106,0.018788378685712814,0.009379632771015167,0.7908242642879486,0.8859724427262943,1.0,0.04944310180450741
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8521001825820171,0.41819798946380615,0.20216156542301178,0.3962128162384033,0.004431429319083691,0.8068687170743942,0.8423398559292158,0.8293504427237365,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.7765876641791117,0.4545985758304596,0.19768491387367249,0.008348237723112106,0.0128417257219553,0.9206205494701862,0.8236871411403021,1.0,0.021969046639768702
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8445852026343346,0.3910732865333557,0.18236319720745087,0.4763438105583191,0.01713361032307148,0.7221040204167366,0.7598466550310453,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.6105531284659445,0.23113161325454712,0.1169707402586937,0.529464840888977,0.008597764186561108,0.22228629142045986,0.48737808441122377,0.9906152628657575,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.7372971773147584,0.265078067779541,0.19102919101715088,0.38728511333465576,0.015489459037780762,0.3283689618110658,0.795954962571462,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8101943848948729,0.5640788078308105,0.21445924043655396,0.17972534894943237,0.026313647627830505,0.737253725528717,0.8935801684856415,1.0,0.4729614446037694
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8910518909584676,0.47009360790252686,0.168345108628273,0.5823169946670532,0.007576141972094774,0.9690425246953964,0.7014379526178043,0.9596309910580294,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.5742638274526449,0.23665672540664673,0.12246014177799225,0.41534799337387085,0.0038780360482633114,0.23955226689577114,0.5102505907416344,0.797291880645693,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8460940940501658,0.469973623752594,0.23579266667366028,0.027240904048085213,0.025629345327615738,0.9686675742268562,0.9824694444735845,1.0,0.07168658960022424
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.6573484642235072,0.35573288798332214,0.17677509784698486,0.007296023890376091,0.011754566803574562,0.6116652749478817,0.7365629076957703,1.0,0.019200062869410766
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8445696403125399,0.3968922197818756,0.21284671127796173,0.2696094810962677,0.02717634290456772,0.7402881868183613,0.8868612969915073,1.0,0.7094986344638624
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.7978115466570384,0.5200223922729492,0.22470614314079285,0.011272979900240898,0.01932649128139019,0.8749300241470337,0.9362755964199703,1.0,0.02966573657958131
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.8986510265618562,0.40813741087913513,0.22649267315864563,0.3366961181163788,0.01212324295192957,0.7754294089972973,0.9437194714943569,1.0,0.8860424160957336
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.810674570981979,0.48731565475463867,0.1195012554526329,0.4459676146507263,0.005300406366586685,0.9771385788917542,0.4979218977193038,0.8726257119946464,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.6552491658269183,0.6907745599746704,0.2888643145561218,0.007220800034701824,0.016408154740929604,0.34132950007915497,1.0,1.0,0.019002105354478483
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0019.png,19,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0019.png,0.7811032718733738,0.3114791512489319,0.19196613132953644,0.3778058886528015,0.013307714834809303,0.47337234765291225,0.7998588805397352,1.0,0.9942260227705303
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,1,0,0,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8456902621926641,0.47216540575027466,0.24522073566913605,0.007689158897846937,0.026696914806962013,0.9755168929696083,1.0,1.0,0.020234628678544572
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,2,0,1,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8580354317083937,0.4262220859527588,0.2029639184474945,0.3047005534172058,0.006931050680577755,0.8319440186023712,0.8456829935312271,0.9378831752987214,0.8018435616242258
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,3,0,2,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8584190677459302,0.38848093152046204,0.21223033964633942,0.3266233503818512,0.026523113250732422,0.7140029110014439,0.8842930818597476,1.0,0.859535132583819
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,4,0,3,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.7268820377360833,0.35411930084228516,0.23172320425510406,0.013277675956487656,0.02197645604610443,0.6066228151321411,0.9655133510629337,1.0,0.03494125251707278
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,5,1,0,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8663382722162887,0.45925769209861755,0.18883401155471802,0.25267890095710754,0.01563076861202717,0.9351802878081799,0.7868083814779918,1.0,0.6649444762029146
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.9824905775487424,0.49867671728134155,0.2486007660627365,0.3911525011062622,0.02533857524394989,0.9416352584958076,1.0,1.0,1.0
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,7,1,2,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.6942733257616821,0.6876616477966309,0.1817881017923355,0.2830265164375305,0.02869051694869995,0.35105735063552856,0.7574504241347313,1.0,0.7448066222040277
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,8,1,3,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.6995152900724683,0.3548176884651184,0.15287499129772186,0.2994121015071869,0.004450107458978891,0.608805276453495,0.6369791304071745,0.8303639223088802,0.7879265829136497
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,9,2,0,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.7313211287684622,0.5973894596099854,0.23988348245620728,0.007782702334225178,0.007397308945655823,0.6331579387187958,0.999514510234197,0.9537890989604284,0.020480795616382046
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,10,2,1,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.7233346639323587,0.4374307096004486,0.16841869056224823,0.006890693213790655,0.013099392876029015,0.8669709675014019,0.7017445440093677,1.0,0.018133403194185934
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,11,2,2,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8201521050857457,0.38383474946022034,0.19059307873249054,0.3378530442714691,0.007428276818245649,0.6994835920631886,0.7941378280520439,0.9548105410392259,0.8890869586091292
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,12,2,3,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.6955489445673791,0.6466243267059326,0.14668011665344238,0.29996973276138306,0.0123206852003932,0.47929897904396057,0.6111671527226766,1.0,0.789394033582587
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,13,3,0,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.8775790249438662,0.5596475601196289,0.22977720201015472,0.29140201210975647,0.013579258695244789,0.7511013746261597,0.9574050083756447,1.0,0.7668474002888328
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,14,3,1,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.5929520671727672,0.2791813015937805,0.15195143222808838,0.24747346341609955,0.003523734398186207,0.37244156748056423,0.6331309676170349,0.7743736527590557,0.6512459563581567
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,15,3,2,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.792331221856569,0.5539194345474243,0.1550215184688568,0.2985629439353943,0.016959497705101967,0.769001767039299,0.6459229936202368,1.0,0.7856919577247218
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,16,3,3,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.9279440619051456,0.43676942586898804,0.2630649507045746,0.3001309037208557,0.013238211162388325,0.8649044558405876,1.0,1.0,0.7898181676864624
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7628913777746106,0.39864563941955566,0.18986448645591736,0.27265122532844543,0.0035600236151367426,0.7457676231861115,0.7911020268996557,0.7768199962663973,0.7175032245485407
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7611940096475577,0.3698378801345825,0.33047229051589966,0.036659859120845795,0.009453600272536278,0.6557433754205704,1.0,1.0,0.09647331347590998
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.6789876241763587,0.4127175807952881,0.14407067000865936,0.27591875195503235,0.0017627556808292866,0.7897424399852753,0.6002944583694141,0.6122450313823883,0.7261019788290325
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.3668024896701126,0.18816962838172913,0.124603770673275,0.45887213945388794,0.00012343046546448022,0.08803008869290363,0.5191823778053125,0.13855499888259107,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.6909713083653636,0.3991561233997345,0.15284113585948944,0.4302128255367279,0.0010789327789098024,0.7473628856241703,0.6368380660812061,0.5028440914150027,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7453696670875408,0.4370042681694031,0.22190676629543304,0.03495150804519653,0.003577538998797536,0.8656383380293846,0.9246115262309711,0.777992239587426,0.0919776527505172
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.3164164923943671,0.7524522542953491,0.12219692021608353,0.0066132927313447,0.0009079689625650644,0.148586705327034,0.5091538342336814,0.46593528094985504,0.017403401924591316
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.6611469538155842,0.3651946783065796,0.1910868138074875,0.4680827260017395,0.0004319914150983095,0.6412333697080612,0.7961950575311979,0.31967370257522565,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.6577045627978575,0.4569193720817566,0.11670377105474472,0.39999139308929443,0.00046692381147295237,0.9278730377554893,0.48626571272810304,0.33385175061111927,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.8050765151565775,0.5087078809738159,0.16202805936336517,0.5184018015861511,0.002776605077087879,0.9102878719568253,0.6751169140140216,0.7178203174612937,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.6214485311374858,0.4847099781036377,0.11687489598989487,0.2653971314430237,0.0003867613268084824,0.9852813184261322,0.48697873329122865,0.30003396021065204,0.6984135037974307
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.5978538312501849,0.36028122901916504,0.16520895063877106,0.19928491115570068,0.0010631472105160356,0.6258788406848907,0.6883706276615461,0.4996555769496986,0.5244339767255282
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7444592952004337,0.5706098079681396,0.30288010835647583,0.07901345938444138,0.003807058557868004,0.7168443500995636,1.0,0.7928658669173511,0.20793015627484573
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7929060568101866,0.3936316668987274,0.22226789593696594,0.4606698453426361,0.0015577770536765456,0.7300989590585232,0.9261162330706915,0.5841659966856888,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.7538208311092871,0.35682809352874756,0.20990991592407227,0.5420700311660767,0.0018852085340768099,0.6150877922773361,0.8746246496836345,0.6276283940839837,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0001.png,1,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0001.png,0.4815692171557062,0.29302898049354553,0.09725406020879745,0.5837064981460571,0.00048568734200671315,0.4157155640423299,0.4052252508699894,0.3411478907280419,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.7842901587094131,0.35678768157958984,0.19774140417575836,0.25998321175575256,0.010966056026518345,0.6149615049362183,0.8239225173989932,1.0,0.6841663467256647
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.4907440327981547,0.6328915357589722,0.16684943437576294,0.037269722670316696,0.0008146517211571336,0.522213950753212,0.695205976565679,0.44322528787787097,0.09807821755346499
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.37545405831083667,0.18723610043525696,0.11006129533052444,0.5960727334022522,0.00028524798108264804,0.0851128138601781,0.45858873054385185,0.24937437995851092,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.6325249371206905,0.42501509189605713,0.09970459342002869,0.628302812576294,0.0007934708846732974,0.8281721621751785,0.4154358059167862,0.4377701867724045,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.7440589077324157,0.49713629484176636,0.19914981722831726,0.13435819745063782,0.001926447614096105,0.9464490786194801,0.8297909051179886,0.6326031281542193,0.35357420381746796
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.41519069063560454,0.2624385952949524,0.087938092648983,0.3576759696006775,0.0003284037229605019,0.32012061029672634,0.3664087193707625,0.27217603599299345,0.9412525515807302
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.8210262310424181,0.3983412981033325,0.20485129952430725,0.4503845274448395,0.003403151873499155,0.7448165565729141,0.8535470813512802,0.7660685586606393,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.8010968356912126,0.4575977027416229,0.21487721800804138,0.382781445980072,0.0007064203964546323,0.9299928210675716,0.8953217417001724,0.4140098674435574,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.6748834936183764,0.54515540599823,0.15268632769584656,0.09364062547683716,0.0044912416487932205,0.7963893562555313,0.6361930320660274,0.8325814893136819,0.24642269862325566
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.5992740329296645,0.32350432872772217,0.1739739179611206,0.40015169978141785,0.00041875772876664996,0.5109510272741318,0.7248913248380026,0.3140853091840972,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.7786330575070187,0.5310583710670471,0.22053061425685883,0.5818937420845032,0.0006699684308841825,0.8404425904154778,0.9188775594035785,0.4033480502452076,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.6203773310599053,0.4487788677215576,0.1615300327539444,0.15300306677818298,0.0005074574728496373,0.9024339616298676,0.673041803141435,0.34935461686429553,0.402639649416271
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.8597139660502846,0.4811953902244568,0.2116411030292511,0.4360826909542084,0.0015644605737179518,0.9962644055485725,0.8818379292885463,0.5851330623965957,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.7826422527086921,0.5284242033958435,0.16135649383068085,0.4578281044960022,0.002633698284626007,0.848674364387989,0.6723187242945036,0.7053773044157775,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.528052307992831,0.29147517681121826,0.16669034957885742,0.3019697070121765,0.000406812148867175,0.4108599275350572,0.6945431232452393,0.30893129680521614,0.794657123716254
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0002.png,2,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0002.png,0.8911951477944692,0.48296621441841125,0.22318270802497864,0.4076688289642334,0.0021687771659344435,0.9907305799424648,0.9299279501040777,0.6599903551220259,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.6693602051072369,0.5813428163528442,0.18909676373004913,0.1714693009853363,0.0020001232624053955,0.6833036988973618,0.787903182208538,0.6412515615460737,0.45123500259299026
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7268840588970904,0.587814211845398,0.24263732135295868,0.2517995834350586,0.0011371364817023277,0.6630805879831314,1.0,0.5142612403743011,0.6626304827238384
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.8637723605071763,0.48292356729507446,0.2961854338645935,0.07765182852745056,0.007090715691447258,0.9908638522028923,1.0,0.9434446690787333,0.20434691717750147
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.43503496895626226,0.25032204389572144,0.1109653189778328,0.6130366325378418,0.000280270614894107,0.2822563871741296,0.46235549574097,0.24660561632692937,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7546294376160023,0.49435877799987793,0.19711707532405853,0.3324585258960724,0.0005419884109869599,0.9551288187503815,0.8213211471835773,0.36184327677051326,0.8748908576212431
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.5357790140129729,0.26767510175704956,0.15501296520233154,0.37654876708984375,0.0005646682111546397,0.33648469299078,0.6458873550097148,0.3697189135725972,0.9909178081311677
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.8380298500220915,0.5373179316520691,0.23175550997257233,0.25657814741134644,0.003974109888076782,0.8208814635872841,0.9656479582190514,0.8031607032712692,0.6752056510824906
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.6574755640956861,0.5453730821609497,0.1507633477449417,0.10026170313358307,0.00335856806486845,0.7957091182470322,0.6281806156039238,0.7629266234454134,0.2638465871936396
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.6226501882459492,0.48558998107910156,0.15544915199279785,0.030146734789013863,0.0010011194972321391,0.9825313091278076,0.6477047999699911,0.48671731450483724,0.07933351260266806
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.712641067811896,0.6001120209693909,0.29811814427375793,0.021973062306642532,0.005163596943020821,0.6246499344706535,1.0,0.8662900409775747,0.05782384817537509
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7871948550681898,0.5719524621963501,0.2298862785100937,0.25806668400764465,0.0030074352398514748,0.712648555636406,0.957859493792057,0.7366960493675858,0.6791228526516965
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.641793152703201,0.3506262004375458,0.14696261286735535,0.30296817421913147,0.0019819033332169056,0.5957068763673306,0.6123442202806473,0.6391404934366022,0.7972846689977143
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7330849346428717,0.4735666513442993,0.1410803645849228,0.5469887256622314,0.0008459041127935052,0.9798957854509354,0.5878348524371784,0.45106297310575066,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7550583002372345,0.4771563410758972,0.18608535826206207,0.13383366167545319,0.00245774257928133,0.9911135658621788,0.7753556594252586,0.6891538226954028,0.3521938465143505
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.8596169245685386,0.5084947347640991,0.2006111443042755,0.33566582202911377,0.004127745982259512,0.9109539538621902,0.8358797679344814,0.8122685657563895,0.883331110602931
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0003.png,3,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0003.png,0.7448875964259782,0.44367218017578125,0.1698082685470581,0.38628077507019043,0.0009114266140386462,0.8864755630493164,0.7075344522794088,0.46673836730944257,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.5491046296075472,0.26951032876968384,0.15532337129116058,0.2970895767211914,0.0012788069434463978,0.3422197774052621,0.6471807137131691,0.540049123738822,0.7818146755820826
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.6771935376525813,0.4508514702320099,0.16540850698947906,0.18462368845939636,0.0010625174036249518,0.9089108444750309,0.6892021124561628,0.49952751525174105,0.4858518117352536
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.5863239928853504,0.530262291431427,0.16597238183021545,0.06857383996248245,0.0006445770268328488,0.8429303392767906,0.6915515909592311,0.3956431711068873,0.18045747358548014
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.6644517674317987,0.5887230038642883,0.14212986826896667,0.365724116563797,0.0015117988223209977,0.660240612924099,0.5922077844540279,0.5774098614569212,0.9624318856942026
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.577825087048279,0.3304774761199951,0.13406091928482056,0.5189406275749207,0.0006644886452704668,0.5327421128749847,0.5585871636867523,0.4017052163190312,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.5340191994946679,0.5783567428588867,0.15847237408161163,0.038343675434589386,0.0008497473900206387,0.692635178565979,0.6603015586733818,0.45201006786840314,0.10090440903839312
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.734094921701522,0.5298677682876587,0.17953215539455414,0.2108716368675232,0.0024959566071629524,0.8441632241010666,0.748050647477309,0.6927678248054223,0.5549253601776926
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.7664406340031571,0.37575316429138184,0.2246083915233612,0.3282003700733185,0.0017875334015116096,0.6742286384105682,0.9358682980140051,0.6154351016610587,0.8636851844034696
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.7961962926053767,0.5303654670715332,0.20402401685714722,0.29360610246658325,0.002466082340106368,0.8426079154014587,0.8501000702381134,0.6899470048120466,0.772647638069956
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.8352073635832489,0.44569772481918335,0.2513512670993805,0.32576867938041687,0.0013684495352208614,0.892805390059948,1.0,0.5550913872393476,0.857285998369518
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.5400806479551098,0.2550522983074188,0.16469964385032654,0.2961341142654419,0.0011295280419290066,0.29703843221068393,0.6862485160430273,0.512798073496867,0.7793003006985313
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.5783194705666215,0.5446428060531616,0.16710570454597473,0.009729161858558655,0.001088706310838461,0.7979912310838699,0.6962737689415615,0.5047980477224547,0.025603057522522777
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.7412741169977423,0.46240657567977905,0.20525386929512024,0.18740150332450867,0.0011095014633610845,0.9450205489993095,0.8552244553963344,0.5089053522038146,0.4931618508539702
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.515069844719576,0.2516993284225464,0.14234349131584167,0.6540426015853882,0.0006744695128872991,0.28656040132045757,0.5930978804826736,0.4046894407145464,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.7832001334272339,0.446008563041687,0.15524061024188995,0.39141198992729187,0.0024048658087849617,0.8937767595052719,0.6468358760078748,0.6840653710931596,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0004.png,4,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0004.png,0.6134309337101822,0.5973632335662842,0.23161309957504272,0.014294463209807873,0.001131614437326789,0.6332398951053619,0.9650545815626781,0.5132001577709633,0.037617008446862825
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.644904072994684,0.2747288644313812,0.23279830813407898,0.09033690392971039,0.0046846866607666016,0.35852770134806644,0.9699929505586624,0.8427542929595396,0.23772869455186943
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.5855007666264935,0.264739453792572,0.15723487734794617,0.33775514364242554,0.001902763033285737,0.3273107931017877,0.6551453222831091,0.6297581328192157,0.8888293253748041
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7529947394891399,0.32671022415161133,0.22624780237674713,0.24029314517974854,0.005378385540097952,0.5209694504737854,0.9426991765697798,0.8761663762730992,0.6323503820519698
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.5716404923462326,0.5789431929588318,0.16808243095874786,0.019546212628483772,0.0015727293211966753,0.6908025220036507,0.7003434623281162,0.5863243471944676,0.05143740165390466
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7570057447702384,0.4401888847351074,0.13464120030403137,0.2912818491458893,0.004712626338005066,0.8755902647972107,0.5610050012667974,0.8441899506264245,0.7665311819628665
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7553748180949609,0.5430905818939209,0.14055100083351135,0.39219382405281067,0.0032531137112528086,0.8028419315814972,0.5856291701396307,0.7553339503144901,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.8412784121143784,0.4979163408279419,0.20769041776657104,0.41737836599349976,0.001625870238058269,0.9440114349126816,0.865376740694046,0.5938478377294403,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7359442257742106,0.3912610113620758,0.2000763863325119,0.30441156029701233,0.0016378737054765224,0.7226906605064869,0.8336516097187996,0.5955163467785843,0.8010830534131903
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7167930857325282,0.3137442469596863,0.2280748337507248,0.42076218128204346,0.00133905082475394,0.48045077174901973,0.9503118072946867,0.550257248077665,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.5511520051437163,0.5719832181930542,0.14227381348609924,0.0146960923448205,0.001783914165571332,0.7125524431467056,0.5928075561920803,0.6149716650343952,0.03867392722321184
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.780349072794403,0.46945565938949585,0.16047142446041107,0.320072740316391,0.0021062190644443035,0.9670489355921745,0.6686309352517128,0.6532024351390671,0.8422966850431342
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7313276683393423,0.5984414219856262,0.22421672940254211,0.21991801261901855,0.002586671616882086,0.6298705562949181,0.9342363725105922,0.7011433914975697,0.578731612155312
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.8137034995707862,0.38323140144348145,0.21401239931583405,0.3945028781890869,0.0031493939459323883,0.6975981295108795,0.8917183304826419,0.7476342462909196,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7622411817917836,0.4944821000099182,0.18153981864452362,0.5074140429496765,0.0006443510064855218,0.9547434374690056,0.7564159110188484,0.3955735089817095,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.4738959535164575,0.23492412269115448,0.08718881756067276,0.47410720586776733,0.0015203093644231558,0.23413788340985786,0.3632867398361365,0.5786742661706369,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0005.png,5,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0005.png,0.7739360061952082,0.49200910329818726,0.14076735079288483,0.4266963601112366,0.001963086659088731,0.9624715521931648,0.5865306283036869,0.6369414081846105,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.6886887093937326,0.618424654006958,0.16508378088474274,0.37317416071891785,0.002161461627110839,0.5674229562282562,0.6878490870197614,0.6592060266474391,0.9820372650497838
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.6215088972502838,0.3120153248310089,0.17621754109859467,0.1688176989555359,0.003435678780078888,0.47504789009690296,0.7342397545774778,0.768336153883137,0.4442571025145681
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.5863781539448947,0.3427355885505676,0.14734266698360443,0.15361320972442627,0.0023734630085527897,0.5710487142205238,0.6139277790983518,0.6809936505478337,0.4042452887484902
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.8813695711258026,0.5256774425506592,0.2570291757583618,0.23378221690654755,0.006646084599196911,0.8572579920291901,1.0,0.9276388778999494,0.6152163602803883
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.6397514414174822,0.3012111485004425,0.23707817494869232,0.02548210322856903,0.003985346294939518,0.44128483906388294,0.9878257289528847,0.8038381842152248,0.06705816639097113
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.5426015126872726,0.4136943817138672,0.09375893324613571,0.17664359509944916,0.0009314829367212951,0.792794942855835,0.3906622218588988,0.47134651346070094,0.464851566051182
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.5638626486351662,0.5996423363685608,0.15750648081302643,0.016911007463932037,0.0024653279688209295,0.6261176988482475,0.6562770033876102,0.6898753611251114,0.04450265122087378
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.5679844338807323,0.2921523153781891,0.1789000779390335,0.7935611009597778,0.00034796964609995484,0.412975985556841,0.7454169914126396,0.2818661631595525,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.8938122158966854,0.507004976272583,0.2635866701602936,0.3133477568626404,0.0036343636456876993,0.9156094491481781,1.0,0.7817579085100215,0.8245993601648431
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.3085464418033443,0.16295188665390015,0.10362079739570618,0.22306188941001892,0.0005168374627828598,0.009224645793438069,0.43175332248210907,0.35280922200374326,0.5870049721316287
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.6872706688027737,0.3059653043746948,0.17968067526817322,0.7144078612327576,0.0026106303557753563,0.45614157617092144,0.7486694802840551,0.7033094074651228,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.5738284744150083,0.5599633455276489,0.10249464213848114,0.32234734296798706,0.0005765077657997608,0.7501145452260971,0.4270610089103381,0.3737337437994891,0.8482824814947028
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.7090984465637497,0.5006695985794067,0.14522510766983032,0.2923628091812134,0.0012006573379039764,0.9354075044393539,0.6051046152909597,0.526153754397759,0.769375813634772
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.8038935426327425,0.5164090394973755,0.29033198952674866,0.053713321685791016,0.0051851216703653336,0.8862217515707016,1.0,0.8672975607211948,0.1413508465415553
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.7181829007275435,0.38627299666404724,0.26293039321899414,0.055565256625413895,0.003004607744514942,0.7071031145751476,1.0,0.7364732496956589,0.14622435954056287
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0006.png,6,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0006.png,0.6351663887406743,0.5496786236763,0.21318873763084412,0.04602416977286339,0.0008969003683887422,0.7822543010115623,0.8882864067951839,0.4633469638479757,0.12111623624437734
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.49145303566091636,0.6972036361694336,0.11492282897233963,0.3439701497554779,0.0008937310194596648,0.32123863697052,0.47884512071808183,0.46260087064553634,0.9051846046196786
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.5948701061700659,0.2820001244544983,0.1753661334514618,0.4176972508430481,0.00082223309436813,0.38125038892030727,0.7306922227144241,0.44514929071858583,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.779811782295266,0.5394842028617859,0.22282609343528748,0.1679912507534027,0.003358659101650119,0.8141118660569191,0.9284420559803646,0.7629330794414776,0.44208223882474396
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.8356298410764676,0.4959571659564972,0.1870853751897812,0.46435844898223877,0.0022345317993313074,0.9501338563859463,0.7795223966240883,0.6669318606938288,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.6866426246021369,0.4887099862098694,0.15112730860710144,0.03334802761673927,0.0034734182991087437,0.9727812930941582,0.629697119196256,0.7709416232125675,0.08775796741247177
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.7501512832969368,0.6107369065284729,0.265173077583313,0.31289374828338623,0.0016473501455038786,0.5914471670985222,1.0,0.5968257722220689,0.8234046007457533
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.6926693028853235,0.4306119680404663,0.14235356450080872,0.4665117859840393,0.0008181483717635274,0.8456624001264572,0.5931398520867031,0.4441145088855018,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.780331358591498,0.49369826912879944,0.19463300704956055,0.3320404291152954,0.0009487842326052487,0.9571929089725018,0.8109708627065023,0.4752545465901945,0.8737906029349879
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.7412980596204741,0.5599890947341919,0.23228563368320465,0.1631484031677246,0.0020427361596375704,0.7500340789556503,0.9678568070133527,0.6461204334753172,0.4293379030729595
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.5848569237447105,0.5629514455795288,0.14128856360912323,0.1776966154575348,0.0008974630618467927,0.7407767325639725,0.5887023483713468,0.46347919450245695,0.4676226722566705
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.6769448149378027,0.42911311984062195,0.1025148332118988,0.37989217042922974,0.0015718723880127072,0.8409784995019436,0.4271451383829117,0.5862011515063902,0.9997162379716572
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.5928555971898726,0.2886362075805664,0.12610505521297455,0.5038477182388306,0.002155001275241375,0.40198814868927013,0.5254377300540607,0.6585113342674933,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.705932004298399,0.3796853721141815,0.13530710339546204,0.27446818351745605,0.005693902261555195,0.6865167878568172,0.5637795974810919,0.8900015387079113,0.7222846934669896
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.6454435321261678,0.5916168093681335,0.2060631811618805,0.25747019052505493,0.0005466698785312474,0.6511974707245827,0.8585965881745021,0.3634893780493667,0.6775531329606709
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.7025162935303909,0.34502357244491577,0.22238174080848694,0.385220468044281,0.0006732209585607052,0.5781986638903618,0.926590586702029,0.40431807341069476,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0007.png,7,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0007.png,0.69653711076485,0.34394538402557373,0.2546344995498657,0.045665543526411057,0.004338566213846207,0.5748293250799179,1.0,0.8242497631849549,0.12017248296423962
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.8105690646442799,0.4463285803794861,0.3430221974849701,0.2921926975250244,0.0011007608845829964,0.894776813685894,1.0,0.5071871913250612,0.7689281513816432
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.5326737479032394,0.32279324531555176,0.14871588349342346,0.17151790857315063,0.0010938644409179688,0.5087288916110992,0.6196495145559311,0.5058231538338626,0.45136291729776484
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.7686796767295756,0.3862152099609375,0.18215151131153107,0.44435128569602966,0.0027512123342603445,0.7069225311279297,0.7589646304647129,0.7156541130071316,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.3463194143455768,0.14977681636810303,0.08675044029951096,0.4799230396747589,0.0005133366212248802,0.0,0.361460167914629,0.3515254558847522,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.6172478515948816,0.3447533845901489,0.21430222690105438,0.02608208730816841,0.0022015373688191175,0.5773543268442154,0.89292594542106,0.6634728365430352,0.06863707186360109
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.5791884485377288,0.27958357334136963,0.16799579560756683,0.4554741382598877,0.0007579150842502713,0.3736986666917802,0.6999824816981952,0.4283364160829445,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.6587629447323052,0.4124422073364258,0.13005979359149933,0.35146406292915344,0.0009845802560448647,0.7888818979263306,0.5419158066312473,0.4831512762035695,0.9249054287609301
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.6650145149524358,0.5780848860740662,0.21857582032680511,0.2481795698404312,0.0004908926784992218,0.6934847310185432,0.9107325846950214,0.3431348022580479,0.6531041311590295
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.7461360353781388,0.47371283173561096,0.21052345633506775,0.01775806024670601,0.002892367774620652,0.9803525991737843,0.8771810680627823,0.7274646983338763,0.0467317374913316
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.8140877091931568,0.4614091217517853,0.1792517602443695,0.413550466299057,0.0019031744450330734,0.941903505474329,0.7468823343515396,0.6298078289815847,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.8313572661426499,0.5087400078773499,0.1966741532087326,0.41763734817504883,0.0020757941529154778,0.9101874753832817,0.8194756383697193,0.6498333280669988,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.8969707852348567,0.5056782960891724,0.27795565128326416,0.30253463983535767,0.004029630217701197,0.9197553247213364,1.0,0.8064904778495743,0.7961437890404149
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.5610418519256445,0.23521874845027924,0.15664102137088776,0.2628627419471741,0.0033715497702360153,0.23505858890712272,0.6526709223786991,0.7638455595061588,0.6917440577557212
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.7780147358500503,0.5677659511566162,0.15973970293998718,0.44748347997665405,0.004679420031607151,0.7257314026355743,0.66558209558328,0.8424827455375763,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.7674222094829416,0.3114550709724426,0.2783783972263336,0.36815696954727173,0.0028075119480490685,0.4732970967888833,1.0,0.7204318435525723,0.9688341303875572
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0008.png,8,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0008.png,0.8529371883565194,0.3769072890281677,0.24687594175338745,0.4722301959991455,0.0038951332680881023,0.6778352782130241,1.0,0.7983464195704487,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.5712428976257962,0.3694494962692261,0.16936977207660675,0.17636814713478088,0.0005779281491413713,0.6545296758413315,0.7057073836525282,0.37421109731879365,0.46412670298626546
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.6577941527633568,0.5941555500030518,0.2226116806268692,0.09721886366605759,0.0016176450299099088,0.6432639062404633,0.9275486692786217,0.5926980514841188,0.25583911491067785
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.854410449641763,0.5369629859924316,0.18464846909046173,0.45143336057662964,0.006131654605269432,0.8219906687736511,0.7693686212102573,0.9080106505863622,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.6084897739914267,0.31131839752197266,0.15703009068965912,0.24219462275505066,0.0025626434944570065,0.47286999225616466,0.6542920445402464,0.6989520895651965,0.6373542704080281
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.5408721315950623,0.31839150190353394,0.14789365231990814,0.4145904779434204,0.0002516356180422008,0.49497344344854366,0.6162235513329506,0.23005213264245628,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.8156097048087454,0.5646195411682129,0.2560921013355255,0.21898874640464783,0.004517565947026014,0.7355639338493347,1.0,0.8339903937663362,0.5762861747490732
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.5631951406908619,0.6449512839317322,0.15662221610546112,0.19430014491081238,0.0015391242923215032,0.48452723771333694,0.652592567106088,0.5814470944893811,0.5113161708179274
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.7548856286470811,0.4512518048286438,0.2291537970304489,0.18452197313308716,0.0010176377836614847,0.9101618900895119,0.9548074876268705,0.4902287774343175,0.4855841398239136
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.8116160473244478,0.4504586160182953,0.17013178765773773,0.42625564336776733,0.002647263929247856,0.9076831750571728,0.7088824485739073,0.7065854409404951,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.6137708078794866,0.32876715064048767,0.14678682386875153,0.3277944028377533,0.0014673632103949785,0.527397345751524,0.6116117661197981,0.5707021875285386,0.862616849573035
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.4190401273271829,0.30692440271377563,0.09289928525686264,0.32397937774658203,0.0001359904563287273,0.45913875848054897,0.3870803552369277,0.14915118693210372,0.8525773098594264
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.6768054256060771,0.3119733929634094,0.18533702194690704,0.6837745308876038,0.001750380382873118,0.47491685301065456,0.772237591445446,0.6106363690769879,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.8635888584858199,0.43105971813201904,0.2154752016067505,0.35374802350997925,0.003954203799366951,0.8470616191625595,0.8978133400281271,0.8019559721092254,0.9309158513420507
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.694967044903313,0.49476802349090576,0.19738459587097168,0.02971147745847702,0.0016809384105727077,0.9538499265909195,0.8224358161290487,0.6014124292043264,0.07818809857493952
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.6915902852276714,0.5798747539520264,0.19535939395427704,0.5465143918991089,0.0005483985878527164,0.6878913938999176,0.8139974748094877,0.3640944984593994,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0009.png,9,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0009.png,0.42422370668623893,0.3005353808403015,0.09679935872554779,0.3312526345252991,0.0001530500449007377,0.43917306512594234,0.4033306613564491,0.16285987939982444,0.8717174592771028
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.6704839497870161,0.3670973777770996,0.2225990742444992,0.08863921463489532,0.0020986509043723345,0.6471793055534363,0.9274961426854134,0.6523686065747684,0.2332610911444614
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.7110441686551522,0.4131958484649658,0.22258344292640686,0.030851446092128754,0.0029616323299705982,0.7912370264530182,0.927431012193362,0.7330622186258022,0.08118801603191778
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.7778628373692659,0.4315928518772125,0.20991018414497375,0.3180709183216095,0.0012855345848947763,0.8487276621162891,0.874625767270724,0.5412099947574748,0.8370287324252882
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.8697790173664266,0.46241772174835205,0.24504254758358002,0.21154966950416565,0.004106417298316956,0.9450553804636002,1.0,0.8110238189554411,0.5567096565899096
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.7342496433881629,0.5698586106300354,0.2511926591396332,0.2065809667110443,0.0013242514105513692,0.7191918417811394,1.0,0.5477878896610034,0.5436341229238009
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.9116364613990301,0.4356265366077423,0.24881285429000854,0.2984734773635864,0.007039450109004974,0.8613329268991947,1.0,0.9416724216903715,0.785456519377859
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.8231386988671866,0.4277927279472351,0.224237322807312,0.2878873348236084,0.0027156274300068617,0.8368522748351097,0.9343221783638,0.7125865019085679,0.7575982495358116
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.8947368091013533,0.553561806678772,0.23800767958164215,0.3868780732154846,0.005131193436682224,0.7701193541288376,0.991698664923509,0.8647656135425973,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.8414846618714866,0.4451272487640381,0.21738509833812714,0.21961691975593567,0.005094381049275398,0.891022652387619,0.9057712430755298,0.863022415420796,0.5779392625156202
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.7604586984687636,0.4707384705543518,0.13267041742801666,0.3729167580604553,0.0018588616512715816,0.9710577204823494,0.5527934059500694,0.6243975083764854,0.9813598896327772
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.702649330090087,0.3778398633003235,0.16029849648475647,0.44661325216293335,0.0016141319647431374,0.6807495728135109,0.6679104020198187,0.5922053505603527,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.6554004684663309,0.46097445487976074,0.13492952287197113,0.3510313034057617,0.00031255162321031094,0.9405451714992523,0.5622063452998798,0.26404010096042607,0.9237665879098993
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.8230012619314453,0.4891759157180786,0.21386070549488068,0.27418527007102966,0.001857157563790679,0.9713252633810043,0.8910862728953362,0.6241870935557045,0.7215401843974465
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.6772429345465579,0.32535064220428467,0.16394241154193878,0.29684972763061523,0.003930022940039635,0.5167207568883896,0.683093381424745,0.8004846759516044,0.7811834937647769
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.602573691819113,0.3625960946083069,0.20693211257457733,0.20560306310653687,0.0003676803898997605,0.633112795650959,0.8622171357274055,0.29126243419104053,0.5410606923856234
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.7360590490770321,0.5022114515304565,0.13738122582435608,0.31711965799331665,0.0019885075744241476,0.9305892139673233,0.5724217742681503,0.6399077609624287,0.8345254157718859
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.4648751330592361,0.28529566526412964,0.13811016082763672,0.2921530604362488,0.00026479666121304035,0.39154895395040523,0.5754590034484863,0.23779667740630275,0.7688238432532862
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.8049133016519397,0.4476465880870819,0.19417859613895416,0.5627347230911255,0.0014633375685662031,0.898895587772131,0.809077483912309,0.5700855205864311,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.6713688384008736,0.43556979298591614,0.1388402283191681,0.36805039644241333,0.0005855443887412548,0.8611556030809879,0.5785009513298671,0.3767552834013944,0.9685536748484561
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.5506484372543811,0.6144756078720093,0.16258235275745392,0.06982122361660004,0.0015547135844826698,0.579763725399971,0.6774264698227247,0.5837214774609212,0.18374006214894748
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.6041100160693248,0.45493897795677185,0.14019645750522614,0.013147925026714802,0.0015891734510660172,0.921684306114912,0.5841519062717756,0.5886767277921459,0.03459980270188106
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.7031050427669342,0.4581500291824341,0.14369626343250275,0.34553366899490356,0.0007651936030015349,0.9317188411951065,0.5987344309687614,0.43029676711072123,0.9092991289339567
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.5305453795474399,0.36061716079711914,0.1600128561258316,0.03550371155142784,0.0011344437953084707,0.6269286274909973,0.6667202338576317,0.5137443926480977,0.09343081987217852
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.6470495442546956,0.5235108137130737,0.11557435989379883,0.31011852622032166,0.0009877150878310204,0.8640287071466446,0.4815598328908285,0.483831098099623,0.8161013847903201
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.5573886537148018,0.2964869737625122,0.1312144547700882,0.43036088347435,0.0008897316292859614,0.42652179300785076,0.5467268948753675,0.46165618939934555,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.6713137542325004,0.3092658817768097,0.17399290204048157,0.52632075548172,0.0021276024635881186,0.4664558805525304,0.7249704251686733,0.6555434500645567,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.7565717680673133,0.4527081251144409,0.23915113508701324,0.06164640933275223,0.0019511771388351917,0.9147128909826279,0.9964630628625553,0.6355394918664773,0.16222739298092692
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.72620806898064,0.30232295393943787,0.21369117498397827,0.4805850684642792,0.0026034843176603317,0.44475923106074344,0.8903798957665762,0.7026653237297766,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.4302927062891502,0.2578611373901367,0.11351554095745087,0.26962342858314514,0.0005392988678067923,0.30581605434417736,0.4729814206560453,0.3608926521303148,0.7095353383766977
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.8544969694761424,0.5137399435043335,0.272903710603714,0.2453012764453888,0.0032786643132567406,0.8945626765489578,1.0,0.7571948611320479,0.6455296748562863
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.7296261564147445,0.5814797282218933,0.21666400134563446,0.47181329131126404,0.0007124610710889101,0.6828758493065834,0.902766672273477,0.4157335997629055,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0011.png,11,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0011.png,0.6919343159523184,0.5126971006393433,0.11076440662145615,0.36805665493011475,0.0013702674768865108,0.8978215605020523,0.46151836092273396,0.5553872713677694,0.9685701445529336
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.5078178767336159,0.24047045409679413,0.14800973236560822,0.5971057415008545,0.0006247545825317502,0.25147016905248176,0.6167072181900343,0.3894586422434443,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.8566788121880696,0.3972838521003723,0.21543057262897491,0.31016236543655396,0.007904613390564919,0.7415120378136635,0.8976273859540622,0.970017889541712,0.8162167511488262
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7778795758783926,0.36013343930244446,0.19296599924564362,0.460585355758667,0.0038602144923061132,0.6254169978201389,0.8040249968568485,0.7961879099011858,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7703438290057437,0.4393177628517151,0.1719665229320526,0.47794777154922485,0.0014897305518388748,0.8728680089116096,0.7165271788835526,0.5741010906687805,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.5346120931962608,0.32668235898017883,0.1604345291852951,0.24328777194023132,0.00044998788507655263,0.5208823718130589,0.6684772049387296,0.32707829340884764,0.6402309787900824
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7873896881318827,0.36831429600715637,0.19295509159564972,0.5768885016441345,0.003981470130383968,0.6509821750223637,0.8039795483152072,0.8036046845224458,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7535257534750005,0.3514583110809326,0.2061678171157837,0.31702980399131775,0.003383974079042673,0.5983072221279144,0.8590325713157654,0.7647218870444533,0.8342889578718888
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.5892814420612532,0.6538997888565063,0.19269070029258728,0.010017551481723785,0.004443009849637747,0.45656315982341766,0.8028779178857803,0.829979288443885,0.026361977583483645
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.4695530190393054,0.6301388144493103,0.1491997241973877,0.01777954399585724,0.0009132509003393352,0.5308162048459053,0.6216655174891155,0.4671610451512117,0.046788273673308525
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.862918082682789,0.4042009115219116,0.223291277885437,0.3862045407295227,0.004253116436302662,0.7631278485059738,0.9303803245226543,0.8194625230968022,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.6146944652047452,0.25229376554489136,0.24218730628490448,0.027086559683084488,0.005241296254098415,0.2884180173277856,1.0,0.8699079878944523,0.07128042021864339
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7231291620990632,0.36052820086479187,0.20815779268741608,0.38499292731285095,0.0010635966900736094,0.6266506277024746,0.867324136197567,0.49974693171620294,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.7508608869554393,0.40931612253189087,0.21477314829826355,0.11391445994377136,0.004171059001237154,0.779112882912159,0.8948881179094315,0.8147774100825257,0.299774894588872
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.6976476591683968,0.4514197111129761,0.11528734862804413,0.4469712972640991,0.0011745275696739554,0.9106865972280502,0.48036395261685055,0.5213299768597062,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.6744626719700663,0.6959607601165771,0.30818822979927063,0.06821224093437195,0.011736618354916573,0.3251226246356964,1.0,1.0,0.17950589719571566
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0012.png,12,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0012.png,0.5618920928029587,0.36674097180366516,0.1596677601337433,0.3316039443016052,0.00013746441982220858,0.6460655368864536,0.6652823338905971,0.150365751066313,0.8726419586884347
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.901039089333058,0.4923376142978668,0.19744431972503662,0.40606826543807983,0.005098136607557535,0.9614449553191662,0.822684665520986,0.8632008123240492,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.6933806681511858,0.4069077670574188,0.14582058787345886,0.2976754605770111,0.002063881605863571,0.7715867720544338,0.6075857828060787,0.6485017216505318,0.7833564752026608
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.5395459549083385,0.32183897495269775,0.110214464366436,0.1911945641040802,0.0025558515917509794,0.5057467967271805,0.45922693486015004,0.6983291878800096,0.5031435897475794
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.723074104012511,0.3421921133995056,0.22335664927959442,0.28294485807418823,0.0020375922322273254,0.569350354373455,0.9306527053316435,0.6455377053394187,0.7445917317741796
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.5367896621620449,0.3666684031486511,0.14386968314647675,0.021912086755037308,0.0018093109829351306,0.6458387598395348,0.5994570131103198,0.6182056893898744,0.057663386197466596
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.6076564844905471,0.39446014165878296,0.11971428990364075,0.2656380236148834,0.0012408719630911946,0.7326879426836967,0.4988095412651698,0.5334004988842591,0.6990474305654827
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.8343826234291025,0.5092136859893799,0.19723199307918549,0.35452017188072205,0.002586779184639454,0.9087072312831879,0.8217999711632729,0.7011531583374119,0.9329478207387422
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.6710300517814379,0.48831427097320557,0.1621699333190918,0.230656698346138,0.0004833596758544445,0.9740179032087326,0.6757080554962158,0.34025426981749035,0.6069913114372053
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.6590352918713756,0.517359733581543,0.1500525325536728,0.12782591581344604,0.001856833929196,0.8832508325576782,0.6252188856403034,0.6241471122582726,0.33638398898275274
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.8748851288948316,0.41417235136032104,0.24328172206878662,0.4486241936683655,0.0031329863704741,0.7942885980010033,1.0,0.7463941979781228,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.7929012096471728,0.35832899808883667,0.2961188554763794,0.23017236590385437,0.005124319810420275,0.6197781190276146,1.0,0.864441044328415,0.6057167523785641
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.5636168949905913,0.29055821895599365,0.15234433114528656,0.36359018087387085,0.0007606055587530136,0.4079944342374803,0.634768046438694,0.4290628438764233,0.9568162654575548
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.36775282343044063,0.1998092383146286,0.12935252487659454,0.5506104230880737,5.827743007102981e-05,0.12440386973321449,0.5389688536524773,0.0749640256589325,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.8159729418601845,0.4391941428184509,0.17043495178222656,0.3373600244522095,0.004481633193790913,0.8724816963076591,0.7101456324259441,0.8320652501411362,0.8877895380321302
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.549960424729375,0.27629488706588745,0.1815863698720932,0.22253744304180145,0.001086855074390769,0.3634215220808984,0.7566098744670551,0.5044291129939537,0.5856248501100039
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.5942974888530487,0.36874687671661377,0.15621457993984222,0.24076762795448303,0.0007759156287647784,0.652333989739418,0.6508940830826759,0.4331568554659721,0.63359902093285
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.5387822689491824,0.2524991035461426,0.18542851507663727,0.3116719424724579,0.0006233080057427287,0.28905969858169567,0.772618812819322,0.3890012687379432,0.8201893222959418
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.7198230709665303,0.3443449139595032,0.16611486673355103,0.4446253180503845,0.003281830810010433,0.5760778561234474,0.6921452780564626,0.7574245228502292,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.5861243864685488,0.4104350805282593,0.17223306000232697,0.053177669644355774,0.0008837472996674478,0.7826096266508102,0.7176377500096958,0.4602359523378687,0.1399412359061994
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.7026045992435812,0.4182705879211426,0.1475747972726822,0.449776828289032,0.0010848540114238858,0.8070955872535706,0.6148949886361759,0.5040297059066293,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.8174458341929662,0.47356078028678894,0.20256386697292328,0.31072568893432617,0.0016019660979509354,0.9798774383962154,0.8440161123871803,0.5904915669881173,0.8176991814061215
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.6525167953097006,0.5464790463447571,0.15615415573120117,0.3283963203430176,0.0005369861610233784,0.7922529801726341,0.6506423155466716,0.3600723205708707,0.864200843007941
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.475906712902198,0.27857527136802673,0.16176266968250275,0.2586418390274048,0.0002717165043577552,0.37054772302508365,0.6740111236770948,0.2417743844702756,0.6806364184931705
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.7448647886768397,0.5238845348358154,0.14353224635124207,0.5292755961418152,0.001874864799901843,0.8628608286380768,0.5980510264635086,0.6263649285854561,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.42361614253163604,0.2379615753889084,0.11408873647451401,0.7327514290809631,0.000254342972766608,0.24362992309033882,0.47536973531047505,0.2316649800455675,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.8229937356749648,0.5276996493339539,0.2556387782096863,0.04487079754471779,0.009175095707178116,0.8509385958313942,1.0,1.0,0.11808104617030997
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.8186204181910673,0.5005805492401123,0.19446156919002533,0.2576698660850525,0.003496234305202961,0.935685783624649,0.8102565382917722,0.7725037294881654,0.6780785949606645
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.6206419705199367,0.6099585294723511,0.17693457007408142,0.4085603952407837,0.0003549609100446105,0.5938795953989029,0.7372273753086727,0.2852395172306562,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.7515491505031063,0.408803254365921,0.19457894563674927,0.268169105052948,0.0023318559397011995,0.7775101698935032,0.8107456068197887,0.6768647672412943,0.7057081711919684
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.4916191941412861,0.32793372869491577,0.15197166800498962,0.25718048214912415,0.00016335748659912497,0.5247929021716118,0.6332152833541235,0.17079250843564528,0.6767907424976951
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.785378946669914,0.44530364871025085,0.18246668577194214,0.37700197100639343,0.0014243132900446653,0.8915739022195339,0.7602778573830923,0.56402740514641,0.9921104500168249
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0014.png,14,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0014.png,0.7220782932132207,0.49459734559059143,0.16165126860141754,0.04513781517744064,0.005105002783238888,0.9543832950294018,0.6735469525059065,0.8635266413198172,0.11878372415115959
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.49267767844368904,0.26461225748062134,0.10888151824474335,0.530407726764679,0.0007791270036250353,0.3269133046269418,0.45367299268643063,0.43400715699870906,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.6535987390741125,0.38890165090560913,0.1634555608034134,0.562164306640625,0.00047942387755028903,0.7153176590800285,0.6810648366808891,0.3387359613833489,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.7770498358844099,0.5924196243286133,0.2051314115524292,0.29809170961380005,0.004507353529334068,0.6486886739730835,0.8547142148017883,0.8334447565649513,0.7844518674047369
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.7364499821764017,0.507427453994751,0.2007940262556076,0.24778585135936737,0.0008555855602025986,0.9142892062664032,0.8366417760650318,0.4534419319720413,0.652068029893072
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.6852039982572852,0.5161172151565552,0.2076721489429474,0.035691022872924805,0.001539762131869793,0.8871337026357651,0.8653006205956142,0.5815405585100253,0.0939237444024337
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.7639204222231437,0.3230203092098236,0.25909748673439026,0.39188843965530396,0.0020271935500204563,0.5094384662806988,1.0,0.6443555293557363,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.55172954958825,0.33216267824172974,0.13162927329540253,0.30552881956100464,0.0007302718004211783,0.5380083695054054,0.5484553053975105,0.4207478628468624,0.8040232093710649
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.8992375074204981,0.47605395317077637,0.24727554619312286,0.2988958954811096,0.003047177568078041,0.9876686036586761,1.0,0.7398068176898293,0.78656814600292
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.8105393603618943,0.5278458595275879,0.17563526332378387,0.4416234493255615,0.0030937432311475277,0.8504816889762878,0.7318135971824329,0.7434030980571122,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.5673177064627702,0.2839888632297516,0.2154962718486786,0.038999177515506744,0.0022184662520885468,0.3874651975929738,0.8979011327028275,0.6652535807874247,0.10262941451449144
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.6972677894313167,0.45136356353759766,0.11399919539690018,0.51524418592453,0.0012023432645946741,0.9105111360549927,0.4749966474870841,0.5264618174747747,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.5773141983542143,0.30510616302490234,0.13242319226264954,0.42579060792922974,0.0010796735296025872,0.45345675945281994,0.5517633010943731,0.5029927207602253,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.699472676715646,0.33534538745880127,0.3038898706436157,0.04096516594290733,0.0053672464564442635,0.547954335808754,1.0,0.8756636629295939,0.10780306827080877
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.6687533008226656,0.549109935760498,0.17978478968143463,0.2452004849910736,0.0008339454652741551,0.7840314507484436,0.749103290339311,0.44809285347313543,0.6452644341870358
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.5001676585997658,0.7235910296440125,0.18783149123191833,0.09244005382061005,0.0018966748612001538,0.2387780323624611,0.7826312134663265,0.6290215596877654,0.24326329952792117
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0015.png,15,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0015.png,0.7794252905169496,0.6585040092468262,0.2980923652648926,0.3004041910171509,0.006252675782889128,0.4421749711036682,1.0,0.9127687898742108,0.790537344781976
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.669791365061986,0.3839704096317291,0.15095727145671844,0.3666802942752838,0.0010923427762463689,0.6999075300991535,0.6289886310696602,0.5055211811475511,0.9649481428296942
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.7309898147693907,0.30789387226104736,0.2669786214828491,0.33853423595428467,0.0019451710395514965,0.4621683508157731,1.0,0.6348294971181857,0.8908795683007491
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.7416314075033364,0.5628526210784912,0.15569724142551422,0.2841411828994751,0.004829203709959984,0.741085559129715,0.6487385059396427,0.8500927789309457,0.7477399549986187
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.5527672409555219,0.30994611978530884,0.1987924575805664,0.04623492807149887,0.0015415763482451439,0.46858162432909023,0.8283019065856934,0.5818062087167174,0.12167086334604967
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.6319872787055848,0.6928229331970215,0.16405747830867767,0.3855248689651489,0.00263785058632493,0.33492833375930786,0.683572826286157,0.7057477227677812,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.683221088684577,0.3864246606826782,0.17891882359981537,0.4304468631744385,0.0006239291396923363,0.7075770646333694,0.7454950983325641,0.3891977591791877,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.8626804363835222,0.4085250496864319,0.22069710493087769,0.4125608205795288,0.004179567098617554,0.7766407802700996,0.9195712705453237,0.8152672845555807,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.768739298874748,0.3191677927970886,0.21781431138515472,0.39663171768188477,0.003746184054762125,0.49739935249090206,0.907559630771478,0.789006415584136,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.7654111514188239,0.48503461480140686,0.13741451501846313,0.24301594495773315,0.004084571730345488,0.9842668287456036,0.5725604792435964,0.8097424493128875,0.6395156446256136
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.8500383615145443,0.46861934661865234,0.21266357600688934,0.4016011357307434,0.0015259786741808057,0.9644354581832886,0.886098233362039,0.5795130162037839,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.6339279065090718,0.6159341931343079,0.120671346783638,0.3001565933227539,0.0034333812072873116,0.5752056464552879,0.5027972782651584,0.7681766532305617,0.789885771901984
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.5379116318812451,0.2847849130630493,0.1818782538175583,0.259038507938385,0.0005518654943443835,0.3899528533220292,0.7578260575731596,0.365303664021725,0.6816802840483815
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.7278664289481981,0.49318361282348633,0.13439776003360748,0.25679725408554077,0.0023985039442777634,0.9588012099266052,0.5599906668066978,0.6834461151567115,0.6757822475935283
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.7518577104260051,0.5031198263168335,0.22078515589237213,0.09038673341274261,0.002054859884083271,0.9277505427598953,0.9199381495515506,0.6474885160680603,0.2378598247703753
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.6176375731289576,0.6915732622146606,0.21648606657981873,0.14225973188877106,0.003274590242654085,0.3388335555791855,0.9020252774159114,0.7568990636236543,0.37436771549676595
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0016.png,16,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0016.png,0.6381204072071143,0.40429770946502686,0.2282719761133194,0.08345244824886322,0.0005459538660943508,0.7634303420782089,0.9511332338054975,0.3632383142171727,0.2196117059180611
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7330340193082298,0.44999629259109497,0.15897077322006226,0.290324866771698,0.001608322374522686,0.9062384143471718,0.6623782217502594,0.5913884295396364,0.764012807293942
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7493258456548904,0.3934113383293152,0.218108668923378,0.1590171456336975,0.0036135895643383265,0.72941043227911,0.908786120514075,0.7803878156355857,0.4184661727202566
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7277463575575079,0.5667400360107422,0.19566664099693298,0.2357901781797409,0.002420980017632246,0.7289373874664307,0.8152776708205541,0.6856270789492174,0.620500468894055
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.8442343241537713,0.5667943954467773,0.20920652151107788,0.4415706396102905,0.004956512711942196,0.7287675142288208,0.8716938396294912,0.8563836719851109,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7572466236647808,0.3393931984901428,0.23652216792106628,0.34284985065460205,0.0019239889224991202,0.5606037452816963,0.9855090330044429,0.6323092912611217,0.902236449091058
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.5058852530493072,0.3131287097930908,0.1750391572713852,0.01810610294342041,0.0013108111452311277,0.4785272181034089,0.729329821964105,0.545523980521338,0.047647639324790554
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7587574649410238,0.38972359895706177,0.22408561408519745,0.2010810673236847,0.002994079142808914,0.717886246740818,0.9336900586883228,0.7356418711590978,0.5291607034833807
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7203480477279417,0.6116021871566772,0.22884266078472137,0.2182397097349167,0.002425859682261944,0.5887431651353836,0.9535110866030058,0.6860980734547784,0.5743150256182018
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7693278962293529,0.39024218916893005,0.21887783706188202,0.2064918577671051,0.0038167431484907866,0.7195068411529064,0.9119909877578418,0.7934744148027691,0.5433996257029081
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.8223124454914866,0.5108766555786133,0.1812843233346939,0.5001590251922607,0.0025589726865291595,0.9035104513168335,0.7553513472278913,0.6986156237122767,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.8081663883209768,0.5444992184638977,0.299907386302948,0.11830835044384003,0.005641008727252483,0.7984399423003197,1.0,0.8877349639279861,0.31133776432589483
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.29199558537059184,0.1553468108177185,0.1307705044746399,0.21876616775989532,0.0001606192090548575,0.0,0.5448771019776663,0.16870955422512265,0.5757004414734087
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7673766576782413,0.4192410111427307,0.2286262959241867,0.301588773727417,0.0009612302528694272,0.8101281598210335,0.952609566350778,0.4780285586845545,0.7936546677037289
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.7429386403578417,0.4979734420776367,0.18141232430934906,0.3531453609466553,0.0005787754198536277,0.9438329935073853,0.7558846846222878,0.3744954093389354,0.9293298972280402
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.8019134311974198,0.5926350355148315,0.25538361072540283,0.2356218695640564,0.0049897548742592335,0.6480155140161514,1.0,0.858000577079682,0.6200575514843589
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0017.png,17,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0017.png,0.6096563057127198,0.40314289927482605,0.13152976334095,0.3560163378715515,0.0004025343805551529,0.7598215602338314,0.5480406805872917,0.3070594740683449,0.9368850996619776
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5274350260880056,0.7606945633888245,0.2167380005121231,0.006245528347790241,0.005217238329350948,0.12282948940992355,0.9030750021338463,0.8687933539503565,0.016435600915237478
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5611384300687533,0.5851519107818604,0.186855286359787,0.06946872174739838,0.0006421997677534819,0.6714002788066864,0.7785636931657791,0.39490949851742585,0.18281242565104835
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5027399799980116,0.24377702176570892,0.12413065135478973,0.41888266801834106,0.0009527546353638172,0.2618031930178405,0.5172110473116239,0.4761428315966892,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.701852157413254,0.3440812826156616,0.20657773315906525,0.4211845397949219,0.000989489839412272,0.5752540081739426,0.8607405548294386,0.48421515404895854,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.7919807781812807,0.5442374348640442,0.2296043187379837,0.3509364724159241,0.0010982006788253784,0.7992580160498619,0.9566846614082655,0.5066816801712792,0.9235170326734844
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5913033942804842,0.5401217937469482,0.18515099585056305,0.05190901458263397,0.0006044276524335146,0.8121193945407867,0.7714624827106794,0.3829537224475974,0.13660266995429993
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5293162970361664,0.36316603422164917,0.11052382737398148,0.3260396122932434,0.0003606978280004114,0.6348938569426537,0.4605159473915895,0.2879740351121374,0.8579989797190616
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.32881135429198544,0.2679606080055237,0.10405325889587402,0.09524674713611603,0.0002681588230188936,0.3373769000172616,0.4335552453994751,0.23973724192662202,0.2506493345687264
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.48401709014159555,0.31321465969085693,0.10235138982534409,0.5903451442718506,0.00028593913884833455,0.47879581153392803,0.42646412427226704,0.24975643759894797,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.7497123503821691,0.5217852592468262,0.1822250932455063,0.2570897340774536,0.0019762986339628696,0.8694210648536682,0.7592712218562763,0.6384874984070795,0.6765519317827726
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.8603093020178384,0.4828222990036011,0.17146454751491547,0.3768714666366577,0.003912905231118202,0.9911803156137466,0.7144356146454811,0.7994378812813472,0.9917670174648887
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.6689275453660899,0.49466878175735474,0.13417480885982513,0.3098253309726715,0.0005673858104273677,0.9541600570082664,0.5590617035826048,0.3706461777458329,0.8153298183491355
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.6845976240354965,0.29342490434646606,0.21649761497974396,0.2544783353805542,0.0032315305434167385,0.41695282608270656,0.9020733957489332,0.7537511319747222,0.6696798299488268
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5515584404814073,0.3145527243614197,0.11079806089401245,0.5300061702728271,0.0009373151697218418,0.4829772636294366,0.46165858705838525,0.47267074110024326,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.735421847009936,0.4081607460975647,0.2019743174314499,0.26274245977401733,0.0015729529550299048,0.7755023315548897,0.8415596559643745,0.5863564875839679,0.6914275257210982
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0018.png,18,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0018.png,0.5850395742895304,0.32857680320739746,0.15541428327560425,0.3503539562225342,0.0005884931888431311,0.5268025100231171,0.6475595136483511,0.3777334115588848,0.9219840953224584
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.754086217799101,0.4806838035583496,0.13950765132904053,0.6133281588554382,0.0011747470125555992,0.9978631138801575,0.5812818805376689,0.5213708778950122,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.5238737471738721,0.3813861012458801,0.15772220492362976,0.08222027868032455,0.0005007764557376504,0.6918315663933754,0.6571758538484573,0.34686459175214535,0.2163691544219067
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.7090063149066516,0.3247353136539459,0.2364426553249359,0.6989353895187378,0.0007869044784456491,0.514797855168581,0.9851777305205663,0.4360545567996295,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.5299444007019457,0.3487287163734436,0.16774962842464447,0.12115742266178131,0.0006014780374243855,0.5897772386670113,0.6989567851026853,0.3819955806076503,0.3188353227941613
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.8302624139458757,0.43707728385925293,0.20593030750751495,0.2957208454608917,0.0036906166933476925,0.8658665120601654,0.858042947947979,0.7854306530460062,0.7782127512128729
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.6638430119034886,0.42382070422172546,0.1807899922132492,0.3043328523635864,0.00034735805820673704,0.8244397006928921,0.7532916342218717,0.2815688893526804,0.8008759272725958
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.8322740903244322,0.3492036461830139,0.23632539808750153,0.4227602779865265,0.0045924559235572815,0.5912613943219185,0.9846891586979231,0.837955697673919,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.49783016370285926,0.32152074575424194,0.14801765978336334,0.17539429664611816,0.000561376684345305,0.5047523304820061,0.6167402490973473,0.3685911961907636,0.46156393854241623
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.7557016893010863,0.3780074715614319,0.21937623620033264,0.23636355996131897,0.0029883957467973232,0.6812733486294746,0.9140676508347194,0.7351919368557559,0.6220093683192605
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.631646326560458,0.41959908604621887,0.1433974951505661,0.03301307186484337,0.00366822793148458,0.811247143894434,0.5974895631273588,0.7839753548711914,0.08687650490748255
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.7663876234821072,0.6059151887893677,0.23965834081172943,0.25959116220474243,0.002918113488703966,0.606515035033226,0.9985764200488727,0.72955996540137,0.6831346373809011
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.6887109147196473,0.3580505847930908,0.18587960302829742,0.21208718419075012,0.0031526745297014713,0.6189080774784088,0.7744983459512393,0.7478814494091957,0.5581241689230266
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.5795848264405808,0.3445149064064026,0.1488630324602127,0.3806297779083252,0.0003484373155515641,0.5766090825200081,0.6202626352508863,0.28209324443725003,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.6501918496500859,0.5018806457519531,0.13531097769737244,0.2845851182937622,0.0005281693302094936,0.9316229820251465,0.5637957404057186,0.35692000806157614,0.7489082060362163
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.5229491535368559,0.31096333265304565,0.1533501297235489,0.06971646845340729,0.002067034365609288,0.4717604145407678,0.6389588738481204,0.6488548336806408,0.1834643906668613
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0019.png,19,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0019.png,0.8301971035540783,0.5700551271438599,0.2567411959171295,0.26293596625328064,0.004695931449532509,0.7185777276754379,1.0,0.8433330890269229,0.6919367532981069
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.589652327018466,0.3462028503417969,0.1663162112236023,0.29602617025375366,0.0005406136624515057,0.5818839073181152,0.6929842134316763,0.36135782066818767,0.779016237509878
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,2,0,1,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.8521574947762125,0.4242672324180603,0.24203713238239288,0.3305864930152893,0.0025268220342695713,0.8258351013064384,1.0,0.6956491843550885,0.869964455303393
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,3,0,2,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6657745843308033,0.37476271390914917,0.14559060335159302,0.5543177127838135,0.001220663427375257,0.6711334809660912,0.6066275139649709,0.5297851434059384,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,4,0,3,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6328670982991692,0.4675992727279663,0.12815552949905396,0.23358049988746643,0.0005607681814581156,0.9612477272748947,0.5339813729127248,0.3683821573597445,0.6146855260196485
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,5,1,0,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.543310276021975,0.20284998416900635,0.17186301946640015,0.49160513281822205,0.0013571144081652164,0.13390620052814495,0.716095914443334,0.5532385661221255,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6861760545257476,0.4424625039100647,0.151298388838768,0.5425307154655457,0.00045469129690900445,0.8826953247189522,0.6304099534948667,0.3289778842464079,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,7,1,2,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6114919930067569,0.5190298557281494,0.165119469165802,0.10441093146800995,0.0006650473806075752,0.8780317008495331,0.6879977881908417,0.40187321970277284,0.27476560912634196
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,8,1,3,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.7524267149206839,0.6192562580108643,0.27952858805656433,0.08354795724153519,0.009028132073581219,0.5648241937160492,1.0,1.0,0.219863045372461
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,9,2,0,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.7143901928813832,0.5092427730560303,0.19194747507572174,0.056271523237228394,0.0027862004935741425,0.9086163341999054,0.7997811461488407,0.7186340215745718,0.14808295588744314
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,10,2,1,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.7741820627846651,0.4214365482330322,0.18114128708839417,0.4445401132106781,0.0017503680428490043,0.8169892132282257,0.754755362868309,0.6106347598228185,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,11,2,2,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6104409068238793,0.3252027630805969,0.1900467723608017,0.20989099144935608,0.0012820684351027012,0.5162586346268654,0.7918615515033405,0.540612575335024,0.5523447143404108
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,12,2,3,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.8843762984261755,0.4633253812789917,0.22993764281272888,0.2457098364830017,0.0050809611566364765,0.9478918164968491,0.9580735117197037,0.8623839001348359,0.6466048328500045
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,13,3,0,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.7913443506970944,0.4147002398967743,0.2117040902376175,0.48919057846069336,0.0013479535700753331,0.7959382496774197,0.882100375990073,0.5517310519873866,1.0
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,14,3,1,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.6740970508808081,0.40151113271713257,0.20314693450927734,0.20272400975227356,0.0008615495753474534,0.7547222897410393,0.8464455604553223,0.45489624157348235,0.5334842361901936
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,15,3,2,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.7807915132229295,0.4497499465942383,0.1673498898744583,0.28309381008148193,0.0032194943632930517,0.9054685831069946,0.697291207810243,0.7528640773465882,0.7449837107407419
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.9072376066109756,0.4530244469642639,0.20702789723873138,0.5726377367973328,0.0058115217834711075,0.9157013967633247,0.8626162384947141,0.8949692641342557,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.9116318971480539,0.47325778007507324,0.18837468326091766,0.5833568572998047,0.006709013134241104,0.9789305627346039,0.784894513587157,0.9299374970061028,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.7192884382614351,0.5028549432754517,0.14797186851501465,0.06983056664466858,0.006251979153603315,0.9285783022642136,0.6165494521458944,0.9127416583146606,0.18376464906491732
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.9414321175531336,0.44176924228668213,0.2915038764476776,0.32242608070373535,0.016569411382079124,0.8805288821458817,1.0,1.0,0.8484896860624614
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.7033952318524059,0.41508862376213074,0.16504724323749542,0.0201161690056324,0.010562589392066002,0.7971519492566586,0.6876968468228977,1.0,0.05293728685692737
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.8426398285253904,0.4743870496749878,0.1499159187078476,0.32832086086273193,0.006537627428770065,0.9824595302343369,0.6246496612826984,0.9236269250241749,0.8640022654282419
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.7806650745241511,0.39060813188552856,0.1571740061044693,0.5212979316711426,0.005286953411996365,0.7206504121422768,0.6548916921019554,0.8720097730035261,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.5968226582950592,0.40812015533447266,0.13728170096874237,0.27247071266174316,0.0004832566191907972,0.775375485420227,0.5720070873697599,0.3402146311031843,0.7170281912151136
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.7410167763584936,0.3909832239151001,0.13519898056983948,0.33453497290611267,0.005780580919235945,0.7218225747346878,0.5633290857076645,0.8936719977882371,0.8803551918581912
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.9190536325552077,0.473049521446228,0.20476196706295013,0.35123834013938904,0.006544474977999926,0.9782797545194626,0.8531748627622923,0.9238821366310577,0.9243114214194448
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.9488322170723044,0.4315989315509796,0.248799666762352,0.4641268253326416,0.008127662353217602,0.8487466610968113,1.0,0.976832874973044,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.8799134305683936,0.4651832580566406,0.1976875215768814,0.3536776900291443,0.004412890411913395,0.953697681427002,0.8236980065703392,0.8283404387360108,0.9307307632345903
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.8113368996941543,0.46745967864990234,0.13910476863384247,0.37097805738449097,0.004107636399567127,0.9608114957809448,0.5796032026410103,0.8110951332210748,0.9762580457486604
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.5545529907275188,0.31655794382095337,0.12142748385667801,0.015188761055469513,0.016303734853863716,0.4892435744404794,0.5059478494028251,1.0,0.03997042383018293
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.868531068767372,0.4173423647880554,0.21599550545215607,0.27177149057388306,0.011116456240415573,0.8041948899626732,0.899981272717317,1.0,0.715188133089166
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.6307223916339769,0.4268700182437897,0.13489581644535065,0.025649476796388626,0.004040693864226341,0.8339688070118427,0.5620659018556278,0.8071487420059084,0.06749862314839113
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0001.png,1,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0001.png,0.7902129892781813,0.4355955123901367,0.1367034614086151,0.35813868045806885,0.005427463911473751,0.8612359762191772,0.5695977558692297,0.8783693515675809,0.9424702117317602
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.6304812259510906,0.33236002922058105,0.18878239393234253,0.058813244104385376,0.004608551971614361,0.5386250913143158,0.7865933080514272,0.8387998075585464,0.15477169501154045
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.8425815588958561,0.4546288251876831,0.17951244115829468,0.2576083838939667,0.007623513229191303,0.9207150787115097,0.7479685048262279,0.9611558555055605,0.677916799720965
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.8772944478550168,0.4919036030769348,0.17881891131401062,0.3103680908679962,0.007893288508057594,0.9628012403845787,0.745078797141711,0.969666866070631,0.8167581338631479
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.9044754948467016,0.495199054479599,0.17497968673706055,0.4956547021865845,0.012232928536832333,0.9525029547512531,0.7290820280710857,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.9809664762009381,0.4803164303302765,0.2304077297449112,0.3651939034461975,0.00880263838917017,0.999011155217886,0.9600322072704633,0.996391917007948,0.9610365880163092
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7071087435499203,0.4330425262451172,0.16347913444042206,0.06416179239749908,0.005596732720732689,0.8532578945159912,0.6811630601684253,0.8858217353191722,0.1688468220986818
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.8555049569674948,0.449088990688324,0.2006777971982956,0.3323103189468384,0.0040863435715436935,0.9034030959010124,0.8361574883262317,0.8098466231970156,0.8745008393337852
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7415669932961465,0.2989102005958557,0.16907094419002533,0.5158606767654419,0.012127527967095375,0.4340943768620492,0.7044622674584389,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.8859900348173476,0.45077216625213623,0.19072438776493073,0.5155342817306519,0.005931638181209564,0.9086630195379257,0.7946849490205448,0.8999425769992261,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7654749847596892,0.40607333183288574,0.1474865823984146,0.5101568698883057,0.003949855454266071,0.7689791619777679,0.6145274266600609,0.8016920326733621,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.5673722327183176,0.5984125137329102,0.151397243142128,0.020615320652723312,0.0028502552304416895,0.6299608945846558,0.6308218464255333,0.7239991353672698,0.054250843822956085
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7141013479144596,0.46762901544570923,0.1497366726398468,0.06541077792644501,0.0048440187238156796,0.9613406732678413,0.6239028026660284,0.8508330448638606,0.17213362612222372
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7457764393807502,0.37381619215011597,0.1717122197151184,0.32549604773521423,0.004068453796207905,0.6681756004691124,0.7154675821463268,0.8087928103815037,0.8565685466716164
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.7084868128476377,0.5493718385696411,0.17679990828037262,0.014074400067329407,0.008502026088535786,0.7832130044698715,0.7366662845015526,0.9878693676764269,0.037037894914024753
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.8427606736554911,0.4617164433002472,0.18360716104507446,0.20366114377975464,0.009909335523843765,0.9428638853132725,0.7650298376878103,1.0,0.5359503783677754
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0002.png,2,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0002.png,0.9178483268540156,0.536697268486023,0.22056271135807037,0.36808985471725464,0.014156797900795937,0.8228210359811783,0.9190112973252933,1.0,0.968657512413828
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.9536914945433015,0.498268723487854,0.30504122376441956,0.3060733377933502,0.009919430129230022,0.9429102391004562,1.0,1.0,0.8054561520877638
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7266573437739036,0.5359104871749878,0.10346105694770813,0.36362937092781067,0.004349157214164734,0.8252797275781631,0.43108773728211724,0.8248367789562083,0.9569193971784491
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7061203039577924,0.5266486406326294,0.10883384943008423,0.32638436555862427,0.0030483838636428118,0.8542229980230331,0.45347437262535095,0.7399006359605439,0.8589062251542744
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7895840741209501,0.5018264055252075,0.20512638986110687,0.03487221151590347,0.007571837864816189,0.9317924827337265,0.8546932910879453,0.9594919812937365,0.09176897767343019
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.9323930676415157,0.49939748644828796,0.20283041894435883,0.4333275854587555,0.008512135595083237,0.9393828548491001,0.8451267456014951,0.9881607500253486,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.8487480713247222,0.4462805986404419,0.18289321660995483,0.3283664882183075,0.005658821202814579,0.8946268707513809,0.7620550692081451,0.8885005548974988,0.8641223374165986
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7565190136365337,0.5335605144500732,0.13098092377185822,0.3108885884284973,0.005489956587553024,0.8326233923435211,0.5457538490494093,0.8811466463033072,0.8181278642855192
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.8773888449712125,0.4858204126358032,0.18043741583824158,0.3711032271385193,0.004694167524576187,0.9818112105131149,0.7518225659926733,0.8432423841749788,0.9765874398382086
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.911135788358636,0.46730929613113403,0.20176656544208527,0.3590331971645355,0.006346751004457474,0.9603415504097939,0.8406940226753553,0.9164059438936246,0.9448242030645672
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.8144658939226678,0.5640316009521484,0.17852573096752167,0.30422383546829224,0.009316966868937016,0.7374012470245361,0.7438572123646736,1.0,0.8005890407060322
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7078882067920497,0.35600370168685913,0.12722991406917572,0.35334473848342896,0.0059937662445008755,0.6125115677714348,0.5301246419548988,0.9024766305227635,0.929854574956392
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.8690387486617965,0.47725147008895874,0.1504698097705841,0.6051380634307861,0.006824813317507505,0.9914108440279961,0.6269575407107671,0.9341129329606703,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7498135830352671,0.5311260223388672,0.14735093712806702,0.2371155023574829,0.005460228770971298,0.84023118019104,0.6139622380336126,0.8798293318123019,0.6239881640986392
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.8355375508359855,0.409909188747406,0.19638465344905853,0.4088955521583557,0.004317310638725758,0.7809662148356438,0.8182693893710773,0.8230674782958767,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.6865378049167581,0.5553699135780334,0.14904765784740448,0.24349263310432434,0.002564128255471587,0.7644690200686455,0.6210319076975187,0.6990880540776496,0.640770087116643
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0003.png,3,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0003.png,0.7776005223325867,0.3858782649040222,0.15233322978019714,0.36954575777053833,0.00639363843947649,0.7058695778250694,0.6347217907508215,0.9181991452963226,0.9724888362382588
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.751717228955762,0.4057711362838745,0.15115566551685333,0.2785658836364746,0.005684683099389076,0.7680348008871078,0.6298152729868889,0.8896079582745552,0.733068114832828
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8777861785151743,0.4648604989051819,0.16930532455444336,0.40875375270843506,0.006477939430624247,0.9526890590786934,0.7054388523101807,0.921391220394048,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8876508531209669,0.5201554894447327,0.20319196581840515,0.30731022357940674,0.011419958434998989,0.8745140954852104,0.8466331909100215,1.0,0.8087111146826493
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.7079242838738254,0.46224045753479004,0.14711743593215942,0.3013780117034912,0.0010017311433330178,0.9445014297962189,0.6129893163839977,0.48684822159984514,0.7931000307986611
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.7192835657353446,0.3992374539375305,0.16077767312526703,0.3582368791103363,0.0017490917816758156,0.7476170435547829,0.6699069713552793,0.6104682675066675,0.9427286292377272
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8414871148192152,0.46537327766418457,0.15654537081718445,0.39580249786376953,0.004594666883349419,0.9542914927005768,0.6522723784049352,0.8380718139502463,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.7882313431032064,0.3851439952850342,0.1550694853067398,0.4065999686717987,0.006801780313253403,0.7035749852657318,0.6461228554447492,0.9332879635602486,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.7816518509291351,0.47819840908050537,0.19116143882274628,0.12679964303970337,0.003567876061424613,0.9943700283765793,0.7965059950947762,0.7773462128566458,0.3336832711571141
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8471137948138149,0.4735890030860901,0.2496320754289627,0.007914397865533829,0.013905524276196957,0.9799656346440315,1.0,1.0,0.02082736280403639
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.9088907594743527,0.5216933488845825,0.20587143301963806,0.35628542304039,0.010535245761275291,0.8697082847356796,0.8577976375818253,1.0,0.9375932185273421
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.9232369151554609,0.4883054494857788,0.18905481696128845,0.366585373878479,0.00988788716495037,0.9740454703569412,0.7877284040053686,1.0,0.9646983523117868
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.7969719933448098,0.3662157654762268,0.19133475422859192,0.4213753938674927,0.004987785592675209,0.6444242671132088,0.797228142619133,0.8579050817004292,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.5753192979551361,0.5942530632019043,0.14895622432231903,0.021611548960208893,0.0031919418834149837,0.6429591774940491,0.6206509346763294,0.7508215588578657,0.05687249726370761
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8111403674137487,0.4501085877418518,0.25330835580825806,0.029312465339899063,0.00619141198694706,0.9065893366932869,1.0,0.9103714256126844,0.0771380666839449
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.8729110772931653,0.4482382833957672,0.18137820065021515,0.46612393856048584,0.00602794298902154,0.9007446356117725,0.7557425027092298,0.9038597431874587,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0004.png,4,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0004.png,0.5888805252214239,0.6784073710441589,0.1568884402513504,0.10217706114053726,0.00739689264446497,0.3799769654870033,0.6537018343806267,0.9537753392436912,0.2688870030014138
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.7075234999925205,0.32262033224105835,0.15193438529968262,0.5581142902374268,0.00504356250166893,0.5081885382533073,0.6330599387486776,0.8605958275677,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.7701049625045512,0.5432980060577393,0.14001405239105225,0.36417579650878906,0.00468368548899889,0.8021937310695648,0.5833918849627178,0.8427026952392723,0.9583573592336554
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.8785445677214547,0.4341217279434204,0.19101892411708832,0.3363805413246155,0.011151997372508049,0.8566303998231888,0.7959121838212013,1.0,0.8852119508542512
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.6503917992804622,0.3763148784637451,0.11481962352991104,0.3270104229450226,0.002573625883087516,0.6759839951992035,0.478415098041296,0.6999560384779507,0.8605537445921647
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.8383769623150951,0.39000558853149414,0.189178004860878,0.34522801637649536,0.014195497147738934,0.7187674641609192,0.788241686920325,1.0,0.9084947799381456
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.7800974337479498,0.5595617294311523,0.18316693603992462,0.30911609530448914,0.004172032233327627,0.7513695955276489,0.7631955668330193,0.8148334949413831,0.8134634086960241
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.7971830388100948,0.46023017168045044,0.1789522022008896,0.19412867724895477,0.005064303055405617,0.9382192865014076,0.74563417583704,0.8615890363569451,0.5108649401288283
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.8397108393533159,0.48148584365844727,0.13454072177410126,0.3850395977497101,0.005734128877520561,0.9953567385673523,0.5605863407254219,0.8917116622619345,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.811642204952828,0.4740508794784546,0.15674322843551636,0.24152350425720215,0.0060266852378845215,0.9814089983701706,0.6530967851479849,0.9038089781307855,0.6355881690979004
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.888365919652738,0.49271297454833984,0.20480704307556152,0.23883134126663208,0.01040099747478962,0.960271954536438,0.8533626794815063,1.0,0.6285035296490318
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.726495551012229,0.5586234927177429,0.13965901732444763,0.34385189414024353,0.003311986569315195,0.7543015852570534,0.5819125721851985,0.759601171738882,0.9048734056322199
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.8400304105193177,0.4717482924461365,0.1704636961221695,0.2562922239303589,0.006823756266385317,0.9742134138941765,0.7102654005090396,0.934075132271792,0.6744532208693654
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.9148354340344668,0.4920879900455475,0.180934339761734,0.42669612169265747,0.009165910072624683,0.9622250311076641,0.7538930823405584,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.8263135050336518,0.327253133058548,0.30818116664886475,0.30674052238464355,0.008707558736205101,0.5226660408079624,1.0,0.9937276305577201,0.8072119010122198
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.890989962494687,0.4554205536842346,0.2528993785381317,0.1622174233198166,0.015306985005736351,0.9231892302632332,1.0,1.0,0.4268879561047805
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0005.png,5,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0005.png,0.7909312176123338,0.5113800764083862,0.12112748622894287,0.3675304651260376,0.005823063664138317,0.901937261223793,0.504697859287262,0.8954514651107465,0.9671854345422042
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8884573838428447,0.44965916872024536,0.20923547446727753,0.26690584421157837,0.018853653222322464,0.9051849022507668,0.8718144769469898,1.0,0.7023838005567852
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.5424075909532144,0.5810010433197021,0.11326755583286285,0.01616492122411728,0.003269416745752096,0.6843717396259308,0.47194814930359524,0.7565229372698733,0.042539266379255994
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.9199470886522805,0.4832928776741028,0.1857834905385971,0.4970867931842804,0.0076880743727087975,0.9897097572684288,0.7740978772441547,0.9632191931940217,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.79763440980382,0.49713900685310364,0.1525484025478363,0.28422439098358154,0.00469512352719903,0.9464406035840511,0.6356183439493179,0.8432915479906348,0.747958923641004
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8936389699578285,0.5909118056297302,0.23809503018856049,0.38532236218452454,0.015070254914462566,0.653400607407093,0.9920626257856687,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8748202401360399,0.48604580760002136,0.1775204986333847,0.27508848905563354,0.00958376843482256,0.9811068512499332,0.7396687443057697,1.0,0.7239170764621935
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8148307377442228,0.48512446880340576,0.1678411066532135,0.4708274006843567,0.001983568537980318,0.983986034989357,0.6993379443883896,0.6393341757235952,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.797819862632375,0.5439020395278931,0.17033889889717102,0.2401711493730545,0.011211000382900238,0.8003061264753342,0.709745412071546,1.0,0.6320293404554066
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8381420406071763,0.38817107677459717,0.19225603342056274,0.33924275636672974,0.009868860244750977,0.7130346149206161,0.8010668059190115,1.0,0.8927440957019204
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8582291157476377,0.43478697538375854,0.16556254029273987,0.4932303726673126,0.008055641315877438,0.8587092980742455,0.6898439178864162,0.9746526038377571,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8232310895245083,0.4738370180130005,0.12807868421077728,0.3554766774177551,0.006293745711445808,0.9807406812906265,0.533661184211572,0.9143631551515713,0.9354649405730397
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8282308311292355,0.48941248655319214,0.15173371136188507,0.4112844169139862,0.003754726145416498,0.9705859795212746,0.6322237973411878,0.7895515922819869,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8399940205578029,0.5373072028160095,0.18343955278396606,0.3080950975418091,0.007943334989249706,0.8209149911999702,0.764331469933192,0.9712143853843498,0.8107765724784449
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.6663175725832784,0.37575361132621765,0.13051962852478027,0.5184023380279541,0.001697147381491959,0.6742300353944302,0.5438317855199178,0.6035961052358961,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.6062260132684663,0.2930038571357727,0.10691956430673599,0.3616059422492981,0.00427300576120615,0.4156370535492898,0.4454981846114,0.8205850163610929,0.9515945848665739
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0006.png,6,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0006.png,0.8229765754788053,0.4177233874797821,0.15528073906898499,0.42389506101608276,0.0072549739852547646,0.8053855858743191,0.6470030794541042,0.9490399035211134,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.7181384994452124,0.31943070888519287,0.14263805747032166,0.40797895193099976,0.007634199224412441,0.49822096526622783,0.5943252394596736,0.9614985521097674,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.8478331998343883,0.5331958532333374,0.180244579911232,0.53031986951828,0.005684364587068558,0.8337629586458206,0.7510190829634666,0.8895943494064086,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.9324784831090485,0.4094802737236023,0.2649616003036499,0.38844987750053406,0.008730137720704079,0.7796258553862572,1.0,0.9943629059726856,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.6096716730115982,0.2992134690284729,0.11796226352453232,0.3512931764125824,0.003491076175123453,0.4350420907139779,0.49150943135221803,0.7721514291260382,0.9244557274015326
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.8223307981891066,0.42776134610176086,0.29406997561454773,0.05397149175405502,0.014682739041745663,0.8367542065680027,1.0,1.0,0.14203024145803952
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.9140250849488534,0.5122768878936768,0.2016274780035019,0.36036747694015503,0.011847620829939842,0.8991347253322601,0.840114491681258,1.0,0.9483354656319869
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.9180771350860596,0.48781657218933105,0.18032413721084595,0.4911941885948181,0.00930072646588087,0.9755732119083405,0.7513505717118582,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.9374442373713449,0.48250946402549744,0.21537818014621735,0.5147197842597961,0.005516034550964832,0.9921579249203205,0.8974090839425723,0.8822965388499081,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.7860941951616304,0.38020795583724976,0.17712736129760742,0.34638020396232605,0.005601859651505947,0.6881498619914055,0.7380306720733643,0.886044028249336,0.9115268525324369
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.868548129333967,0.40274500846862793,0.21631170809268951,0.31362342834472656,0.008476955816149712,0.7585781514644623,0.9012987837195396,0.9871453082549713,0.8253248114334909
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.9170337303688652,0.543903648853302,0.2270985096693039,0.362444669008255,0.012053943239152431,0.8003010973334312,0.9462437902887663,1.0,0.9538017605480394
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.765771490363299,0.4606860280036926,0.1381261646747589,0.33797234296798706,0.002700167940929532,0.9396438375115395,0.5755256861448288,0.7112419915371537,0.8894009025473344
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.8089450322475654,0.429851233959198,0.16050851345062256,0.3250102698802948,0.006134443450719118,0.8432851061224937,0.6687854727109274,0.9081213240528488,0.8552901838955126
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.7632670428910165,0.4325971007347107,0.1387975960969925,0.3940131366252899,0.003009276930242777,0.8518659397959709,0.5783233170708021,0.7368410633239384,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.6310432369302732,0.2640632390975952,0.15765050053596497,0.2662172019481659,0.006583734415471554,0.32519762217998516,0.6568770855665207,0.9253403479808132,0.7005715840741208
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0007.png,7,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0007.png,0.6948630072001871,0.33744144439697266,0.13102002441883087,0.5216690301895142,0.0050093019381165504,0.5545045137405396,0.5459167684117954,0.8589464902179466,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8578294727559153,0.3566327393054962,0.24090887606143951,0.31283190846443176,0.010827157646417618,0.6144773103296757,1.0,1.0,0.8232418643800836
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.9062446802854538,0.41842663288116455,0.21117576956748962,0.5985947847366333,0.014399413019418716,0.8075832277536392,0.8798990398645401,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.830003917554375,0.4168418049812317,0.1585189402103424,0.43012213706970215,0.007720976136624813,0.802630640566349,0.6604955842097601,0.9642642004861689,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7277176810031172,0.5425536632537842,0.20449933409690857,0.04221971705555916,0.004954543896019459,0.8045198023319244,0.8520805587371191,0.8562875795892602,0.11110451856726095
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8718254566192627,0.4160996079444885,0.18538565933704376,0.40254637598991394,0.012268205173313618,0.8003112748265266,0.7724402472376823,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7404042784705271,0.44251206517219543,0.13541799783706665,0.31520071625709534,0.0028918397147208452,0.8828502036631107,0.5642416576544445,0.7274215388424707,0.8294755690976193
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7926763694929448,0.35941803455352783,0.18378563225269318,0.31917446851730347,0.008972741663455963,0.6231813579797745,0.7657734677195549,1.0,0.8399328118876407
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8511700443923473,0.42353320121765137,0.16328613460063934,0.54993736743927,0.009047940373420715,0.8235412538051605,0.6803588941693306,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.9835371665656567,0.483590304851532,0.2652520537376404,0.346821129322052,0.009857337921857834,0.9887802973389626,1.0,1.0,0.9126871824264526
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8313710134730616,0.35413259267807007,0.2063194364309311,0.535241961479187,0.007772427052259445,0.606664352118969,0.8596643184622129,0.9658896491948282,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8246953509217139,0.422816663980484,0.19060008227825165,0.4336584806442261,0.0033205871004611254,0.8213020749390125,0.7941670094927152,0.7602185023687822,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7477972348932284,0.43886902928352356,0.11274916678667068,0.3504166603088379,0.004413885995745659,0.8714657165110111,0.4697881949444612,0.8283947821808131,0.9221491060758892
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7676579830016416,0.3923119306564331,0.1518050581216812,0.44908827543258667,0.004639924503862858,0.7259747833013535,0.6325210755070051,0.8404369014365359,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.7482811951208406,0.4355027973651886,0.1117275282740593,0.44914543628692627,0.003944254480302334,0.8609462417662144,0.4655313678085804,0.8013516489936086,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.9661234380884615,0.47174549102783203,0.2849469780921936,0.42810940742492676,0.005822984501719475,0.9742046594619751,1.0,0.8954481609994762,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.8949746314436198,0.3679729402065277,0.27674049139022827,0.4891512989997864,0.010152001865208149,0.6499154381453991,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.8200183843348151,0.4506993293762207,0.1702098548412323,0.38663724064826965,0.003035563975572586,0.9084354043006897,0.7092077285051346,0.7389017779722711,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.7428060335911393,0.3004440665245056,0.1944645643234253,0.32973453402519226,0.007331442553550005,0.43888770788908016,0.8102690180142721,0.9516025885039153,0.8677224579610322
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.9174913689494133,0.46987205743789673,0.1815890520811081,0.41412973403930664,0.012643668800592422,0.9683501794934273,0.7566210503379505,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.8970126920280906,0.4145393967628479,0.21817629039287567,0.6078044772148132,0.007067584432661533,0.7954356148838997,0.909067876636982,0.9426465782873044,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.8587507868744532,0.38062766194343567,0.20873036980628967,0.4191056787967682,0.0077125681564211845,0.6894614435732365,0.8697098741928737,0.9639975661784804,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.7221774383361201,0.4957490563392639,0.13767169415950775,0.28004205226898193,0.0018016818212345243,0.9507841989398003,0.573632058997949,0.6172385823418417,0.7369527691288998
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.7535079380139714,0.33035600185394287,0.20295388996601105,0.296772837638855,0.005737125873565674,0.5323625057935715,0.8456412081917127,0.8918386042648266,0.7809811516811973
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.9408562304132185,0.46214669942855835,0.23004117608070374,0.33930355310440063,0.007119272835552692,0.9442084357142448,0.9585049003362656,0.9444264661221557,0.8929040871168438
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.9571427084505558,0.4932240843772888,0.21563223004341125,0.4287053942680359,0.01311987079679966,0.9586747363209724,0.8984676251808803,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.7850668812464726,0.37262359261512756,0.15030664205551147,0.3745507001876831,0.01575438305735588,0.6644487269222736,0.6262776752312978,1.0,0.9856597373360082
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.8073081995703673,0.5828477144241333,0.17797638475894928,0.33251887559890747,0.010230002924799919,0.6786008924245834,0.7415682698289554,1.0,0.8750496726287038
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.6022668950316687,0.5781677961349487,0.128550186753273,0.07408773899078369,0.004217946901917458,0.6932256370782852,0.5356257781386375,0.8174652413546036,0.19496773418627286
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.8587818834930658,0.3847428858280182,0.19846834242343903,0.5150179266929626,0.009994670748710632,0.7023215182125568,0.8269514267643293,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.9715768993578436,0.46528637409210205,0.23166707158088684,0.4172598123550415,0.008339861407876015,0.9540199190378189,0.9652794649203619,0.9831483366815578,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.6447668988382608,0.5440042614936829,0.15740476548671722,0.06123896688222885,0.0029906013514846563,0.7999866828322411,0.6558531895279884,0.7353666429172484,0.1611551760058654
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0009.png,9,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0009.png,0.6264041364138251,0.3067668080329895,0.11905036866664886,0.41130155324935913,0.0033173896372318268,0.4586462751030923,0.49604320277770364,0.7599891721983453,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.9316919521281594,0.46314090490341187,0.2222105711698532,0.30332648754119873,0.018684761598706245,0.9473153278231621,0.9258773798743885,1.0,0.7982275987926283
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.8362612498826102,0.4938051700592041,0.22587327659130096,0.04271706938743591,0.015329504385590553,0.9568588435649872,0.941138652463754,1.0,0.1124133404932524
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.7165602362825156,0.44519293308258057,0.2061324268579483,0.016990389674901962,0.0030403914861381054,0.8912279158830643,0.858885111908118,0.7392783807150094,0.0447115517760578
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.7544185508750695,0.38513433933258057,0.18245159089565277,0.3918326199054718,0.0021797525696456432,0.7035448104143143,0.7602149620652199,0.661162476524837,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.6353039675322418,0.38145485520362854,0.1089044138789177,0.33215081691741943,0.0020047901198267937,0.6920464225113392,0.45376839116215706,0.6417894354301092,0.8740810971511037
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.7871885440680236,0.3831210136413574,0.15443271398544312,0.5156314373016357,0.006988164037466049,0.6972531676292419,0.643469641606013,0.9398868051897886,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.6827521596496043,0.2784041464328766,0.15173983573913574,0.33458614349365234,0.009204437956213951,0.37001295760273945,0.6322493155797323,1.0,0.8804898512990851
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.7846924976592785,0.4086574912071228,0.23411597311496735,0.02262553758919239,0.013685686513781548,0.7770546600222588,0.975483221312364,1.0,0.059540888392611555
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.9533158849336598,0.462105929851532,0.23926198482513428,0.30656903982162476,0.016961678862571716,0.9440810307860374,0.9969249367713928,1.0,0.8067606311095388
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.9361020717769861,0.4754411280155182,0.19230081140995026,0.3937605917453766,0.011753564700484276,0.9857535250484943,0.8012533808747928,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.8594096478365186,0.36091628670692444,0.24943320453166962,0.44628000259399414,0.005559524521231651,0.6278633959591389,1.0,0.8842025161951075,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.8705840738196122,0.5035823583602905,0.21619310975074768,0.1835429072380066,0.009106960147619247,0.9263051301240921,0.9008046239614487,1.0,0.4830076506263331
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.8898332814907813,0.49297034740448,0.17341305315494537,0.3887375295162201,0.007017411291599274,0.9594676643610001,0.7225543881456058,0.9409066629551981,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.7507977860676607,0.4421009421348572,0.16425229609012604,0.3041188418865204,0.002022879896685481,0.8815654441714287,0.6843845670421919,0.6438634857303186,0.8003127418066326
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.861679406269971,0.45241543650627136,0.17910058796405792,0.38011497259140015,0.004921246320009232,0.913798239082098,0.7462524498502414,0.8546567983610766,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0010.png,10,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0010.png,0.851069178362064,0.395182728767395,0.23430998623371124,0.3024459183216095,0.0053139738738536835,0.7349460273981094,0.9762916093071302,0.8732453625783748,0.7959103113726566
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8150182591137793,0.4300612807273865,0.2012411504983902,0.23933394253253937,0.005098999477922916,0.8439415022730827,0.8385047937432926,0.8632417824998787,0.6298261645593142
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.567511856040738,0.2784450650215149,0.12670008838176727,0.2576598823070526,0.0036924059968441725,0.37014082819223415,0.5279170349240303,0.7855465953070356,0.6780523218606648
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.6365915862482319,0.5922831296920776,0.08095132559537888,0.3849925994873047,0.0033549717627465725,0.6491152197122574,0.3372971899807453,0.762671453361324,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.9557315949350595,0.4665307104587555,0.21468724310398102,0.3871467709541321,0.01106947846710682,0.9579084701836109,0.894530179599921,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.6630120269093387,0.5909801125526428,0.1772090196609497,0.019194001331925392,0.007339530624449253,0.6531871482729912,0.7383709152539571,0.9518721135125032,0.05051052982085629
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.759279356697286,0.41999003291130066,0.19927017390727997,0.07469053566455841,0.0072203692980110645,0.8124688528478146,0.8302923912803333,0.9478715091018534,0.19655404122252212
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8259643635075342,0.44743263721466064,0.24041514098644257,0.0164572075009346,0.011462138034403324,0.8982269912958145,1.0,1.0,0.04330844079193316
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.9654331909590645,0.5166597962379456,0.2601720690727234,0.379497766494751,0.009406311437487602,0.8854381367564201,1.0,1.0,0.9986783328809236
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8087891925676537,0.41896116733551025,0.19532984495162964,0.22636893391609192,0.006710141897201538,0.8092536479234695,0.8138743539651235,0.9299785355052104,0.5957077208318209
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8723431497812272,0.6161673069000244,0.25088974833488464,0.44107890129089355,0.015555943362414837,0.5744771659374237,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.902430422800152,0.49903982877731323,0.2022586166858673,0.2975577116012573,0.012624170631170273,0.9405005350708961,0.8427442361911138,1.0,0.7830466094769929
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8566085423686003,0.5279251337051392,0.19167496263980865,0.2835931181907654,0.009432444348931313,0.8502339571714401,0.7986456776658695,1.0,0.7462976794493825
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.9167166218161583,0.5052434802055359,0.19230590760707855,0.4406413435935974,0.010759645141661167,0.9211141243577003,0.801274615029494,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.7332594111351545,0.35893142223358154,0.15458421409130096,0.3551349639892578,0.004895415157079697,0.6216606944799423,0.6441008920470874,0.8533843238830381,0.9345656947085732
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.8619380719959736,0.5050806999206543,0.1483609825372696,0.4822181761264801,0.009327592328190804,0.9216228127479553,0.6181707605719566,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0011.png,11,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0011.png,0.763304197082394,0.6085044145584106,0.1783471703529358,0.28080257773399353,0.014795580878853798,0.5984237045049667,0.7431132098038992,1.0,0.7389541519315619
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7158460548313181,0.397758424282074,0.1910116821527481,0.027980558574199677,0.007986839860677719,0.7429950758814812,0.7958820089697838,0.9725518881760709,0.07363304887947283
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.9387205943465232,0.4146353006362915,0.2460840493440628,0.48197445273399353,0.016116084530949593,0.795735314488411,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7778074048745092,0.5069218873977661,0.12979258596897125,0.32808905839920044,0.004732152447104454,0.9158691018819809,0.5408024415373802,0.8451884120276448,0.8633922589452643
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7244336965262637,0.40517404675483704,0.2190302014350891,0.019041884690523148,0.004889964126050472,0.7661688961088657,0.9126258393128713,0.8531149698770916,0.05011022286979776
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.8636746160262206,0.47610896825790405,0.14643068611621857,0.7238438129425049,0.0069098807871341705,0.9878405258059502,0.6101278588175774,0.9371364025566497,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.8681429023413282,0.47012829780578613,0.15831895172595978,0.32806396484375,0.009349946863949299,0.9691509306430817,0.6596622988581657,1.0,0.8633262232730263
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7536923471477148,0.5303459167480469,0.11438284069299698,0.4298040270805359,0.0044739628210663795,0.8426690101623535,0.4765951695541541,0.8316523729310502,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7794650786113853,0.48692944645881653,0.19055089354515076,0.06873984634876251,0.005521905142813921,0.9783454798161983,0.7939620564381282,0.8825546714423036,0.18089433249674344
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.8002301176699335,0.4839397370815277,0.13295488059520721,0.2905888855457306,0.0057431962341070175,0.9876883216202259,0.5539786691466968,0.8920955256345889,0.7647075935413963
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.6416018173452989,0.3573581278324127,0.0971079096198082,0.45106086134910583,0.003059644717723131,0.6167441494762897,0.4046162900825342,0.7407747419106067,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7302155502281252,0.40242958068847656,0.20257100462913513,0.08602330088615417,0.00509538222104311,0.7575924396514893,0.8440458526213964,0.8630699856279528,0.22637710759514257
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.8919718374170396,0.4909346401691437,0.1829661875963211,0.3153865933418274,0.008791543543338776,0.965829249471426,0.7623591149846713,0.9960824807284825,0.8299647193205983
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.6881742365541984,0.5913218259811401,0.10964275151491165,0.4025637209415436,0.004297134466469288,0.6521192938089371,0.45684479797879857,0.8219400360715108,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.9243399069958896,0.4931982755661011,0.22578468918800354,0.4697073698043823,0.004226100631058216,0.9587553888559341,0.9407695382833481,0.8179297154164198,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7163034117498465,0.45544472336769104,0.16119147837162018,0.026964709162712097,0.006150629371404648,0.9232647605240345,0.6716311598817508,0.9087626859397405,0.07095976095450551
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0012.png,12,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0012.png,0.7900618837822942,0.5382877588272095,0.15630966424942017,0.48000404238700867,0.003877816488966346,0.8178507536649704,0.6512902677059174,0.7972783094841118,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.843622784695209,0.4460427165031433,0.19060862064361572,0.4360131621360779,0.003164730966091156,0.8938834890723228,0.7942025860150655,0.7487878486759701,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8530858941376209,0.592255175113678,0.2125052809715271,0.36149024963378906,0.011180900037288666,0.6492025777697563,0.8854386707146963,1.0,0.9512901306152344
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.7295526614400535,0.30937492847442627,0.16713692247867584,0.4565824270248413,0.006504006218165159,0.4667966514825822,0.6964038436611494,0.922370051587736,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8402884434908628,0.34234514832496643,0.21547189354896545,0.5588728189468384,0.011818223632872105,0.5698285885155201,0.8977995564540228,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8848915930837393,0.3572176992893219,0.2689514458179474,0.4199599027633667,0.009353730827569962,0.6163053102791309,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.9490443464956785,0.46286529302597046,0.23512785136699677,0.30703574419021606,0.01412055641412735,0.9464540407061577,0.9796993806958199,1.0,0.8079888005005685
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.7436631292051705,0.5985981225967407,0.2016238123178482,0.18397708237171173,0.006461880635470152,0.6293808668851852,0.8400992179910343,0.9207862849086214,0.48415021676766246
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.27824470352082153,0.7856162786483765,0.09011327475309372,0.013086659833788872,0.0015830990159884095,0.04494912922382355,0.3754719781378905,0.5878103381432469,0.034438578509970716
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8600906051828257,0.5038070678710938,0.1652703583240509,0.3225787281990051,0.008715537376701832,0.925602912902832,0.6886264930168788,0.9939522996292207,0.8488913899973819
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8061984225290415,0.5071687698364258,0.15667252242565155,0.3790510892868042,0.003112172707915306,0.9150975942611694,0.6528021767735481,0.7448122449479723,0.9975028665442216
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.7468581363673962,0.5641802549362183,0.17713198065757751,0.25274860858917236,0.004233876243233681,0.7369367033243179,0.738049919406573,0.8183718477885592,0.6651279173399273
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.7940011328063342,0.3370037078857422,0.1865503191947937,0.5364981889724731,0.008216256275773048,0.5531365871429443,0.7772929966449738,0.9794890306798354,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8412809632718563,0.3839397430419922,0.1850699633359909,0.6466249823570251,0.009766172617673874,0.6998116970062256,0.7711248472332954,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8293791332221887,0.3835112452507019,0.1854138821363449,0.5246545076370239,0.007351785898208618,0.6984726414084435,0.7725578422347705,0.9522799525168982,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.6318264134533996,0.3000733256340027,0.09789157658815384,0.5086930990219116,0.006247645244002342,0.4377291426062585,0.40788156911730766,0.9125727997453189,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0013.png,13,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0013.png,0.8530633766204119,0.3916539251804352,0.1887102574110031,0.4534149765968323,0.010742193087935448,0.7239185161888599,0.786292739212513,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.7805764975034436,0.49535948038101196,0.12522120773792267,0.34557345509529114,0.004057674668729305,0.9520016238093376,0.5217550322413445,0.8081557052340752,0.9094038291981346
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8115707012395226,0.3908045291900635,0.18456260859966278,0.44036346673965454,0.004988769069314003,0.7212641537189484,0.7690108691652616,0.8579527774970385,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.7546108777407563,0.3149700164794922,0.17627683281898499,0.4901430606842041,0.007462111301720142,0.4842813014984132,0.7344868034124374,0.9559217850700042,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8348472186858512,0.5052899122238159,0.1425306797027588,0.4295963644981384,0.0064827632158994675,0.9209690243005753,0.5938778320948284,0.9215726470689206,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.9116570405579624,0.4579988121986389,0.19427739083766937,0.40064263343811035,0.007517970632761717,0.9312462881207466,0.8094891284902891,0.9577456622986068,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.7831956819188084,0.4145781397819519,0.18005575239658356,0.3199993968009949,0.0034972601570189,0.7955566868185997,0.7502323016524315,0.7725737360347411,0.8421036757920918
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8497177962522485,0.528252363204956,0.17157141864299774,0.4179542660713196,0.006493085995316505,0.8492113649845123,0.7148809110124906,0.9219604538125904,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8477325943624341,0.5478042960166931,0.18221351504325867,0.377159059047699,0.006951847113668919,0.788111574947834,0.7592229793469112,0.9386146085365172,0.9925238395992079
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8293601733246289,0.5703660845756531,0.22234542667865753,0.2892942428588867,0.005642659030854702,0.7176059857010841,0.9264392778277397,0.8878059936025265,0.7613006391023335
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8754360514517048,0.47541603446006775,0.16772903501987457,0.3482781648635864,0.006721084006130695,0.9856751076877117,0.6988709792494774,0.9303760095923187,0.9165214864831221
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8031599724763319,0.3450366258621216,0.2989216148853302,0.20187661051750183,0.016666218638420105,0.5782394558191299,1.0,1.0,0.5312542382039522
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.6114059155028135,0.29910266399383545,0.12111715227365494,0.3780674338340759,0.002819701097905636,0.4346958249807359,0.504654801140229,0.7214543309280816,0.9949142995633576
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.5813320235069901,0.279643177986145,0.13355864584445953,0.3075016736984253,0.0028423594776540995,0.3738849312067033,0.5564943576852481,0.7233439888864207,0.8092149307853297
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8970121815800667,0.4527817368507385,0.17802344262599945,0.6306769847869873,0.00922947097569704,0.9149429276585579,0.7417643442749977,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.6383918591053747,0.40350714325904846,0.18181803822517395,0.02632799744606018,0.002464060438796878,0.7609598226845264,0.7575751592715582,0.6897549357909446,0.06928420380542152
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0014.png,14,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0014.png,0.8100685693414514,0.3897586762905121,0.17422953248023987,0.3983655273914337,0.0061195967718958855,0.7179958634078503,0.7259563853343328,0.9075315788751861,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7531626727458968,0.36451369524002075,0.15716034173965454,0.3906942903995514,0.0050295512191951275,0.6391052976250648,0.6548347572485607,0.8599226251352363,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.8558606616787459,0.47998446226119995,0.15786714851856232,0.41584277153015137,0.004520841874182224,0.9999514445662498,0.6577797854940097,0.8341651706426719,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7269447186931878,0.3516874313354492,0.2197750359773636,0.4817739427089691,0.0010169134475290775,0.5990232229232788,0.9157293165723484,0.4900758273779987,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.8546108287016478,0.5095452070236206,0.20612335205078125,0.2846146821975708,0.004811981692910194,0.9076712280511856,0.8588473002115886,0.8492294774214139,0.748986005783081
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7799806835576515,0.443111777305603,0.13125254213809967,0.5569332838058472,0.003954778891056776,0.8847243040800095,0.546885592242082,0.8019908586440961,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.8395220767706633,0.4117112457752228,0.16283422708511353,0.4288371801376343,0.009580913931131363,0.7865976430475712,0.678475946187973,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.8496681485185789,0.49299055337905884,0.25172892212867737,0.04769085347652435,0.007971653714776039,0.9594045206904411,1.0,0.9720858216512737,0.12550224599085355
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7035404018017976,0.412480890750885,0.11283215880393982,0.2936195135116577,0.004623625427484512,0.7890027835965157,0.47013399501641595,0.8395877146953706,0.7726829302938361
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.8972006485925356,0.5154476761817932,0.2248903065919876,0.334020733833313,0.0052406624890863895,0.8892260119318962,0.9370429441332817,0.8698786884077503,0.8790019311402973
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7692721901848919,0.5759490728378296,0.16087286174297333,0.34940776228904724,0.005482954904437065,0.7001591473817825,0.6703035905957222,0.8808370083943823,0.9194941112869665
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.813525316996492,0.44119101762771606,0.13355450332164764,0.37842822074890137,0.0068312836810946465,0.8787219300866127,0.5564770971735319,0.9343441919860559,0.9958637388128984
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.9533576257526875,0.4510265588760376,0.22441618144512177,0.39836180210113525,0.00955219380557537,0.9094579964876175,0.9350674226880074,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7707626793366871,0.527869462966919,0.17080965638160706,0.3755146265029907,0.0017887844005599618,0.8504079282283783,0.7117069015900295,0.6155950902441472,0.9881963855341861
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7182628997854772,0.24694395065307617,0.2194398045539856,0.6423986554145813,0.004823611117899418,0.27169984579086315,0.9143325189749401,0.8498127614229448,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.789276619090148,0.34370309114456177,0.20826925337314606,0.48268985748291016,0.004386099521070719,0.5740721598267555,0.867788555721442,0.826873617702755,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0015.png,15,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0015.png,0.7051714816375783,0.38240617513656616,0.18909160792827606,0.02609632909297943,0.010261263698339462,0.6950192973017693,0.7878816997011503,1.0,0.06867455024468272
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.7801536326345644,0.45335298776626587,0.19793348014354706,0.01955316960811615,0.013701657764613628,0.9167280867695808,0.8247228339314461,1.0,0.0514557094950425
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.8445020968780705,0.4707901179790497,0.14413464069366455,0.3115190863609314,0.013069191947579384,0.9712191186845303,0.600561002890269,1.0,0.8197870693708721
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.6878062608426095,0.3866174817085266,0.19492429494857788,0.10360478609800339,0.0033624463248997927,0.7081796303391457,0.8121845622857412,0.7632015078553052,0.27264417394211415
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.6389193837510914,0.4596977233886719,0.12902195751667023,0.012631891295313835,0.003412984311580658,0.9365553855895996,0.5375914896527927,0.7667561931945783,0.0332418191981943
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.8216232769191266,0.3050840497016907,0.2284855842590332,0.3820759356021881,0.015298811718821526,0.45338765531778347,0.9520232677459717,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.7096776374076542,0.5780296921730042,0.19735117256641388,0.012391820549964905,0.012805609032511711,0.693657211959362,0.8222965523600578,1.0,0.03261005407885501
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.8407918077372092,0.620857834815979,0.24256296455860138,0.35614013671875,0.006684855557978153,0.5598192662000656,1.0,0.929057579847574,0.9372108861019737
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.8991474531989166,0.4658392667770386,0.24270987510681152,0.3388429880142212,0.00273983390070498,0.9557477086782455,1.0,0.7146773181487912,0.8916920737216347
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.5958681341978838,0.6486636400222778,0.14591249823570251,0.15401607751846313,0.004693643189966679,0.4729261249303818,0.6079687426487606,0.843215415404254,0.40530546715385035
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.8129371797175784,0.5317171812057495,0.2578273415565491,0.028935827314853668,0.00998673401772976,0.8383838087320328,1.0,1.0,0.07614691398645702
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.5651608628931603,0.6833639740943909,0.16268715262413025,0.0062209744937717915,0.012641256675124168,0.36448758095502853,0.677863135933876,1.0,0.016370985509925766
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.877409378066659,0.39324894547462463,0.2069907933473587,0.41296687722206116,0.013196568936109543,0.728902954608202,0.862461638947328,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.7032346452533137,0.39415043592453003,0.18804651498794556,0.032015345990657806,0.007109512109309435,0.7317201122641563,0.7835271457831066,0.9440913250555009,0.08425091050173107
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.7165740866450822,0.26834478974342346,0.18711236119270325,0.40012964606285095,0.006559509783983231,0.33857746794819843,0.7796348383029302,0.924441579078974,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.5664347354941224,0.2962338328361511,0.09687282145023346,0.4931526184082031,0.0022690289188176394,0.42573072761297237,0.40363675604263943,0.6704979615897552,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0016.png,16,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0016.png,0.7396954286610681,0.5592689514160156,0.17220357060432434,0.1704862117767334,0.006597068160772324,0.7522845268249512,0.7175148775180181,0.9258336739957619,0.4486479257282458
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.8442183920524334,0.48688799142837524,0.15965047478675842,0.5003736019134521,0.003995539154857397,0.9784750267863274,0.6652103116114935,0.8044511621323487,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.6410519987720096,0.4011753797531128,0.1851675808429718,0.020025700330734253,0.0025999138597398996,0.7536730617284775,0.7715315868457159,0.7023428899610052,0.05269921139666909
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.86377886967814,0.5184291005134583,0.19683726131916046,0.5097706913948059,0.004175588022917509,0.879909060895443,0.8201552554965019,0.8150382990422259,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.758978031212746,0.3221725821495056,0.17948639392852783,0.3638976812362671,0.007457105442881584,0.506789319217205,0.7478599747021993,0.9557576859851721,0.957625476937545
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.9038423381941882,0.409789115190506,0.23749284446239471,0.31109076738357544,0.014844270423054695,0.7805909849703312,0.9895535185933113,1.0,0.8186599141673038
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.9634627433234912,0.47291696071624756,0.23459888994693756,0.3239399790763855,0.0087862154468894,0.9778655022382736,0.9774953747789066,0.9959337433843193,0.8524736291483829
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.9587230589240789,0.47054538130760193,0.2140694111585617,0.4209824800491333,0.013771463185548782,0.970454316586256,0.8919558798273405,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.47707103936026096,0.37434113025665283,0.0729745477437973,0.24355140328407288,0.0005230667302384973,0.6698160320520401,0.3040606155991554,0.35507733296896815,0.6409247454844023
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.6473111374152174,0.27569398283958435,0.1519845426082611,0.2707657217979431,0.007837953045964241,0.3615436963737012,0.6332689275344213,0.9679445770796333,0.7125413731524819
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.8116120765595167,0.5236519575119019,0.14505447447299957,0.47546645998954773,0.005574851296842098,0.8635876327753067,0.6043936436374983,0.8848707745427008,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.8127743720471664,0.444951593875885,0.16335873305797577,0.45281827449798584,0.0033983970060944557,0.8904737308621407,0.6806613877415657,0.7657353458642178,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.6712224901360864,0.3072311580181122,0.13959455490112305,0.4694788157939911,0.004532766528427601,0.4600973688066007,0.5816439787546794,0.8348003434708092,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.8597688288354238,0.39560994505882263,0.20252938568592072,0.5092368125915527,0.007074663415551186,0.7362810783088207,0.843872440358003,0.9428910929415066,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.9041264336093214,0.4977225661277771,0.18159882724285126,0.46509695053100586,0.008066127076745033,0.9446169808506966,0.7566617801785469,0.9749712212021933,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.7938350441206685,0.3075231909751892,0.20913930237293243,0.4482620358467102,0.008114363066852093,0.4610099717974664,0.8714137598872185,0.9764316984610523,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0017.png,17,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0017.png,0.7612275901320376,0.4324954152107239,0.20078690350055695,0.1060132086277008,0.004862003494054079,0.8515481725335121,0.8366120979189873,0.8517287592045708,0.2789821279676337
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.6608368756922062,0.22072871029376984,0.1651599407196045,0.40084564685821533,0.008569758385419846,0.18977721966803085,0.6881664196650188,0.989815135569165,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8468255383795813,0.33724701404571533,0.22844411432743073,0.3675900101661682,0.014804944396018982,0.5538969188928604,0.9518504763642948,1.0,0.9673421320162321
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.665890697917889,0.6605539321899414,0.16458719968795776,0.281404972076416,0.005316935013979673,0.4357689619064331,0.6857799986998241,0.8733803988233908,0.7405394002010948
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8004285773722115,0.42226481437683105,0.13724349439144135,0.6330792307853699,0.006766077131032944,0.819577544927597,0.5718478932976723,0.9320037836185229,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.6232161501415286,0.6281447410583496,0.17022386193275452,0.027113670483231544,0.0074181510135531425,0.5370476841926575,0.7092660913864772,0.9544770112134189,0.0713517644295567
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8030673289741919,0.41276875138282776,0.16799092292785645,0.28648263216018677,0.007971818558871746,0.7899023480713367,0.6999621788660686,0.9720908854242182,0.7539016635794389
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.9916329786181449,0.471075177192688,0.24396774172782898,0.457231342792511,0.012744851410388947,0.97210992872715,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8295583172373873,0.39070284366607666,0.17669259011745453,0.5697683691978455,0.007892250083386898,0.7209463864564896,0.7362191254893939,0.9696346546144888,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.814958595368014,0.40864747762680054,0.1892300248146057,0.28821277618408203,0.006606536917388439,0.7770233675837517,0.7884584367275238,0.9261834117973434,0.7584546741686369
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.7675276328846306,0.40161049365997314,0.1523541361093521,0.46607843041419983,0.003959886729717255,0.7550327926874161,0.6348089004556339,0.8023004997668624,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.5200068490660519,0.256272554397583,0.07324738055467606,0.3423658013343811,0.004126410931348801,0.300851732492447,0.30519741897781694,0.812190833445605,0.9009626350904766
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.7481858869803213,0.5221759080886841,0.11329855769872665,0.4596588611602783,0.0036749073769897223,0.8682002872228622,0.4720773237446944,0.7844104147602172,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8341397816919324,0.46161893010139465,0.21556951105594635,0.23363739252090454,0.003300016513094306,0.9425591565668583,0.8982062960664432,0.7587394375221289,0.6148352434760646
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.674450934023086,0.593754768371582,0.16519565880298615,0.15953929722309113,0.004757953807711601,0.6445163488388062,0.6883152450124423,0.8465016699607018,0.4198402558502398
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8466502315333896,0.36832594871520996,0.2587817907333374,0.47484612464904785,0.0040110088884830475,0.6510185897350311,1.0,0.805378618451521,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.8780793850537953,0.5441845655441284,0.1914602518081665,0.3823481798171997,0.008778149262070656,0.7994232326745987,0.7977510492006938,0.9957084019648305,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.7743748755831468,0.5691772699356079,0.17154815793037415,0.2369765341281891,0.016300462186336517,0.7213210314512253,0.714783991376559,1.0,0.6236224582320765
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8619959031812633,0.4507926404476166,0.16111496090888977,0.40773117542266846,0.007341461256146431,0.9087270013988018,0.6713123371203741,0.9519364065020419,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.7066250439198589,0.5716242790222168,0.18737395107746124,0.05050738900899887,0.0073877498507499695,0.7136741280555725,0.7807247961560886,0.953472957663865,0.1329141816026286
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.5056347468934457,0.22614547610282898,0.09118492156267166,0.549953281879425,0.0027854307554662228,0.20670461282134067,0.3799371731777986,0.7185688443748154,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.6227453587840699,0.4331698417663574,0.09616827219724655,0.3962051272392273,0.0006131107220426202,0.8536557555198669,0.400701134155194,0.38575316752620664,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8396480940282345,0.3269829750061035,0.22648124396800995,0.4687620997428894,0.01470126025378704,0.5218217968940735,0.9436718498667082,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.7099210181013608,0.31649407744407654,0.16886259615421295,0.4748417139053345,0.004063806030899286,0.4890439920127393,0.703594150642554,0.8085183012190911,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8099728170084275,0.35479500889778137,0.19246485829353333,0.38528430461883545,0.007197186350822449,0.6087344028055668,0.8019369095563889,0.9470856931993631,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8545190396680131,0.47948700189590454,0.13741663098335266,0.4840865731239319,0.006791440770030022,0.9983968809247017,0.5725692957639694,0.9329167466456473,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8397022318094969,0.34533509612083435,0.21276046335697174,0.5063046216964722,0.009771408513188362,0.5791721753776073,0.8865019306540489,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.790564654718459,0.5082401037216187,0.21723245084285736,0.013210451230406761,0.007622961420565844,0.9117496758699417,0.905135211845239,0.9611381464097138,0.034764345343175684
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.6134575094655182,0.31865814328193665,0.09847656637430191,0.4015432596206665,0.0034090199042111635,0.49580669775605213,0.4103190265595913,0.7664791686833007,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.6539911699843112,0.5120658278465271,0.14365725219249725,0.023638959974050522,0.003616808447986841,0.8997942879796028,0.5985718841354053,0.7806005997559976,0.06220778940539611
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.8337498741309814,0.38103026151657104,0.20580703020095825,0.35122478008270264,0.006508420221507549,0.6907195672392845,0.8575292925039928,0.9225354225961467,0.9242757370597438
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.888482208297846,0.53099524974823,0.19411543011665344,0.4020345211029053,0.008053380995988846,0.8406398445367813,0.8088142921527227,0.974583869163979,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0019.png,19,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0019.png,0.6482220436817949,0.37397441267967224,0.13799653947353363,0.28632599115371704,0.0020630434155464172,0.6686700396239758,0.5749855811397235,0.6484077595680293,0.7534894504045185
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,1,0,0,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8964344077792606,0.4967292845249176,0.2178792655467987,0.22741487622261047,0.010392201133072376,0.9477209858596325,0.9078302731116613,1.0,0.5984602005858171
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,2,0,1,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.7477494503518469,0.3490465581417084,0.1692298799753189,0.2760850787162781,0.010202908888459206,0.5907704941928387,0.7051244998971622,1.0,0.7265396808323107
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,3,0,2,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8461171741282683,0.45089292526245117,0.1624472588300705,0.3305509686470032,0.00757088977843523,0.9090403914451599,0.6768635784586271,0.9594613505543131,0.8698709701236925
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,4,0,3,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8527149935240242,0.43802428245544434,0.2156882882118225,0.26951342821121216,0.005120911635458469,0.8688258826732635,0.8987012008825939,0.8642799556008389,0.7092458637137162
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,5,1,0,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.782702446741304,0.34150466322898865,0.20672714710235596,0.6069340705871582,0.004201602190732956,0.5672020725905895,0.8613631129264832,0.8165315643447286,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,6,1,1,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.6019026508833238,0.4371374249458313,0.12076646089553833,0.006374691613018513,0.003241011407226324,0.8660544529557228,0.503193587064743,0.754447652961865,0.01677550424478556
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.9943782195448875,0.4740034341812134,0.24226152896881104,0.5552682280540466,0.011902406811714172,0.9812607318162918,1.0,1.0,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,8,1,3,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8351591751228237,0.4067757725715637,0.22853095829486847,0.24988508224487305,0.005419593304395676,0.7711742892861366,0.9522123262286186,0.8780173688553681,0.6575923216970343
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.7196681389088123,0.3361770510673523,0.1668766885995865,0.4398764967918396,0.00366286002099514,0.5505532845854759,0.6953195358316104,0.7836251711347457,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,10,2,1,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8628471709974193,0.4437311291694641,0.21354557573795319,0.3641795516014099,0.0031100288033485413,0.8866597786545753,0.8897732322414716,0.7446487262806156,0.9583672410563419
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,11,2,2,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.5654179924064291,0.41548284888267517,0.06326419860124588,0.41734370589256287,0.0006179199554026127,0.7983839027583599,0.2636008275051912,0.38729029330945514,1.0
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,12,2,3,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.5551017851929729,0.2972780466079712,0.1138349249958992,0.20973512530326843,0.004009346477687359,0.4289938956499101,0.47431218748291337,0.8052791168494484,0.551934540271759
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,13,3,0,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.7427720903309117,0.49463269114494324,0.18332000076770782,0.007203459739685059,0.005884123034775257,0.9542728401720524,0.7638333365321159,0.8979870654791423,0.018956472999171206
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,14,3,1,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.9437301296936838,0.428253710269928,0.2368381768465042,0.37035953998565674,0.0142231285572052,0.8382928445935249,0.9868257368604343,1.0,0.9746303683833072
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,15,3,2,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.6189949802472814,0.6584123373031616,0.11095558851957321,0.36505600810050964,0.004154894035309553,0.44246144592761993,0.4623149521648884,0.8138440199615212,0.960673705527657
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,16,3,3,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.8894035668581899,0.45796364545822144,0.17688898742198944,0.5253802537918091,0.007458568550646305,0.931136392056942,0.7370374475916227,0.9558056598544821,1.0
|
||||||
|
Binary file not shown.
|
After Width: | Height: | Size: 204 KiB |
@@ -0,0 +1,10 @@
|
|||||||
|
run,architecture,grid,grid_index,tile_index,row,col,source_path,score,mean,std,saturation,sharpness,exposure_score,contrast_score,detail_score,color_score,rank,tile_path
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0020.png,20,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0020.png,0.9943782195448875,0.4740034341812134,0.24226152896881104,0.5552682280540466,0.011902406811714172,0.9812607318162918,1.0,1.0,1.0,1,outputs\samples\final_showcase\top_tiles\p5_ddpm\rank_01_grid_0020_tile_07.png
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0018.png,18,7,1,2,outputs\samples\final_comparison\p5_ddpm\grid_0018.png,0.9916329786181449,0.471075177192688,0.24396774172782898,0.457231342792511,0.012744851410388947,0.97210992872715,1.0,1.0,1.0,2,outputs\samples\final_showcase\top_tiles\p5_ddpm\rank_02_grid_0018_tile_07.png
|
||||||
|
p5_ddpm,DDPM - cosine v-pred wider,grid_0008.png,8,9,2,0,outputs\samples\final_comparison\p5_ddpm\grid_0008.png,0.9835371665656567,0.483590304851532,0.2652520537376404,0.346821129322052,0.009857337921857834,0.9887802973389626,1.0,1.0,0.9126871824264526,3,outputs\samples\final_showcase\top_tiles\p5_ddpm\rank_03_grid_0008_tile_09.png
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0012.png,12,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0012.png,0.9931557374587933,0.4784943461418152,0.26595890522003174,0.5266944169998169,0.008175451308488846,0.9952948316931725,1.0,0.9782691518033664,1.0,1,outputs\samples\final_showcase\top_tiles\p5_gan\rank_01_grid_0012_tile_06.png
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0013.png,13,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0013.png,0.9891186870634555,0.4683932662010193,0.25315365195274353,0.42203789949417114,0.012284314259886742,0.9637289568781853,1.0,1.0,1.0,2,outputs\samples\final_showcase\top_tiles\p5_gan\rank_02_grid_0013_tile_06.png
|
||||||
|
p5_gan,GAN - WGAN-GP + SN + Attn,grid_0020.png,20,6,1,1,outputs\samples\final_comparison\p5_gan\grid_0020.png,0.9824905775487424,0.49867671728134155,0.2486007660627365,0.3911525011062622,0.02533857524394989,0.9416352584958076,1.0,1.0,1.0,3,outputs\samples\final_showcase\top_tiles\p5_gan\rank_03_grid_0020_tile_06.png
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0010.png,10,6,1,1,outputs\samples\final_comparison\p5_vae\grid_0010.png,0.9116364613990301,0.4356265366077423,0.24881285429000854,0.2984734773635864,0.007039450109004974,0.8613329268991947,1.0,0.9416724216903715,0.785456519377859,1,outputs\samples\final_showcase\top_tiles\p5_vae\rank_01_grid_0010_tile_06.png
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0020.png,20,16,3,3,outputs\samples\final_comparison\p5_vae\grid_0020.png,0.9072376066109756,0.4530244469642639,0.20702789723873138,0.5726377367973328,0.0058115217834711075,0.9157013967633247,0.8626162384947141,0.8949692641342557,1.0,2,outputs\samples\final_showcase\top_tiles\p5_vae\rank_02_grid_0020_tile_16.png
|
||||||
|
p5_vae,VAE - perceptual + PatchGAN,grid_0013.png,13,1,0,0,outputs\samples\final_comparison\p5_vae\grid_0013.png,0.901039089333058,0.4923376142978668,0.19744431972503662,0.40606826543807983,0.005098136607557535,0.9614449553191662,0.822684665520986,0.8632008123240492,1.0,3,outputs\samples\final_showcase\top_tiles\p5_vae\rank_03_grid_0013_tile_01.png
|
||||||
|
@@ -0,0 +1,191 @@
|
|||||||
|
[
|
||||||
|
{
|
||||||
|
"run":"p5_ddpm",
|
||||||
|
"architecture":"DDPM - cosine v-pred wider",
|
||||||
|
"grid":"grid_0020.png",
|
||||||
|
"grid_index":20,
|
||||||
|
"tile_index":7,
|
||||||
|
"row":1,
|
||||||
|
"col":2,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_ddpm\\grid_0020.png",
|
||||||
|
"score":0.9943782195,
|
||||||
|
"mean":0.4740034342,
|
||||||
|
"std":0.242261529,
|
||||||
|
"saturation":0.5552682281,
|
||||||
|
"sharpness":0.0119024068,
|
||||||
|
"exposure_score":0.9812607318,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":1.0,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":1,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_ddpm\\rank_01_grid_0020_tile_07.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_ddpm",
|
||||||
|
"architecture":"DDPM - cosine v-pred wider",
|
||||||
|
"grid":"grid_0018.png",
|
||||||
|
"grid_index":18,
|
||||||
|
"tile_index":7,
|
||||||
|
"row":1,
|
||||||
|
"col":2,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_ddpm\\grid_0018.png",
|
||||||
|
"score":0.9916329786,
|
||||||
|
"mean":0.4710751772,
|
||||||
|
"std":0.2439677417,
|
||||||
|
"saturation":0.4572313428,
|
||||||
|
"sharpness":0.0127448514,
|
||||||
|
"exposure_score":0.9721099287,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":1.0,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":2,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_ddpm\\rank_02_grid_0018_tile_07.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_ddpm",
|
||||||
|
"architecture":"DDPM - cosine v-pred wider",
|
||||||
|
"grid":"grid_0008.png",
|
||||||
|
"grid_index":8,
|
||||||
|
"tile_index":9,
|
||||||
|
"row":2,
|
||||||
|
"col":0,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_ddpm\\grid_0008.png",
|
||||||
|
"score":0.9835371666,
|
||||||
|
"mean":0.4835903049,
|
||||||
|
"std":0.2652520537,
|
||||||
|
"saturation":0.3468211293,
|
||||||
|
"sharpness":0.0098573379,
|
||||||
|
"exposure_score":0.9887802973,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":1.0,
|
||||||
|
"color_score":0.9126871824,
|
||||||
|
"rank":3,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_ddpm\\rank_03_grid_0008_tile_09.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_gan",
|
||||||
|
"architecture":"GAN - WGAN-GP + SN + Attn",
|
||||||
|
"grid":"grid_0012.png",
|
||||||
|
"grid_index":12,
|
||||||
|
"tile_index":6,
|
||||||
|
"row":1,
|
||||||
|
"col":1,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_gan\\grid_0012.png",
|
||||||
|
"score":0.9931557375,
|
||||||
|
"mean":0.4784943461,
|
||||||
|
"std":0.2659589052,
|
||||||
|
"saturation":0.526694417,
|
||||||
|
"sharpness":0.0081754513,
|
||||||
|
"exposure_score":0.9952948317,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":0.9782691518,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":1,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_gan\\rank_01_grid_0012_tile_06.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_gan",
|
||||||
|
"architecture":"GAN - WGAN-GP + SN + Attn",
|
||||||
|
"grid":"grid_0013.png",
|
||||||
|
"grid_index":13,
|
||||||
|
"tile_index":6,
|
||||||
|
"row":1,
|
||||||
|
"col":1,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_gan\\grid_0013.png",
|
||||||
|
"score":0.9891186871,
|
||||||
|
"mean":0.4683932662,
|
||||||
|
"std":0.253153652,
|
||||||
|
"saturation":0.4220378995,
|
||||||
|
"sharpness":0.0122843143,
|
||||||
|
"exposure_score":0.9637289569,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":1.0,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":2,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_gan\\rank_02_grid_0013_tile_06.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_gan",
|
||||||
|
"architecture":"GAN - WGAN-GP + SN + Attn",
|
||||||
|
"grid":"grid_0020.png",
|
||||||
|
"grid_index":20,
|
||||||
|
"tile_index":6,
|
||||||
|
"row":1,
|
||||||
|
"col":1,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_gan\\grid_0020.png",
|
||||||
|
"score":0.9824905775,
|
||||||
|
"mean":0.4986767173,
|
||||||
|
"std":0.2486007661,
|
||||||
|
"saturation":0.3911525011,
|
||||||
|
"sharpness":0.0253385752,
|
||||||
|
"exposure_score":0.9416352585,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":1.0,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":3,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_gan\\rank_03_grid_0020_tile_06.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_vae",
|
||||||
|
"architecture":"VAE - perceptual + PatchGAN",
|
||||||
|
"grid":"grid_0010.png",
|
||||||
|
"grid_index":10,
|
||||||
|
"tile_index":6,
|
||||||
|
"row":1,
|
||||||
|
"col":1,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_vae\\grid_0010.png",
|
||||||
|
"score":0.9116364614,
|
||||||
|
"mean":0.4356265366,
|
||||||
|
"std":0.2488128543,
|
||||||
|
"saturation":0.2984734774,
|
||||||
|
"sharpness":0.0070394501,
|
||||||
|
"exposure_score":0.8613329269,
|
||||||
|
"contrast_score":1.0,
|
||||||
|
"detail_score":0.9416724217,
|
||||||
|
"color_score":0.7854565194,
|
||||||
|
"rank":1,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_vae\\rank_01_grid_0010_tile_06.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_vae",
|
||||||
|
"architecture":"VAE - perceptual + PatchGAN",
|
||||||
|
"grid":"grid_0020.png",
|
||||||
|
"grid_index":20,
|
||||||
|
"tile_index":16,
|
||||||
|
"row":3,
|
||||||
|
"col":3,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_vae\\grid_0020.png",
|
||||||
|
"score":0.9072376066,
|
||||||
|
"mean":0.453024447,
|
||||||
|
"std":0.2070278972,
|
||||||
|
"saturation":0.5726377368,
|
||||||
|
"sharpness":0.0058115218,
|
||||||
|
"exposure_score":0.9157013968,
|
||||||
|
"contrast_score":0.8626162385,
|
||||||
|
"detail_score":0.8949692641,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":2,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_vae\\rank_02_grid_0020_tile_16.png"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"run":"p5_vae",
|
||||||
|
"architecture":"VAE - perceptual + PatchGAN",
|
||||||
|
"grid":"grid_0013.png",
|
||||||
|
"grid_index":13,
|
||||||
|
"tile_index":1,
|
||||||
|
"row":0,
|
||||||
|
"col":0,
|
||||||
|
"source_path":"outputs\\samples\\final_comparison\\p5_vae\\grid_0013.png",
|
||||||
|
"score":0.9010390893,
|
||||||
|
"mean":0.4923376143,
|
||||||
|
"std":0.1974443197,
|
||||||
|
"saturation":0.4060682654,
|
||||||
|
"sharpness":0.0050981366,
|
||||||
|
"exposure_score":0.9614449553,
|
||||||
|
"contrast_score":0.8226846655,
|
||||||
|
"detail_score":0.8632008123,
|
||||||
|
"color_score":1.0,
|
||||||
|
"rank":3,
|
||||||
|
"tile_path":"outputs\\samples\\final_showcase\\top_tiles\\p5_vae\\rank_03_grid_0013_tile_01.png"
|
||||||
|
}
|
||||||
|
]
|
||||||
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.5 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.6 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.4 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.8 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 8.0 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.4 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.1 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.4 KiB |
BIN
Binary file not shown.
|
After Width: | Height: | Size: 7.3 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 6.4 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 4.4 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 4.8 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 5.6 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 122 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 117 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 119 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 120 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 21 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 16 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 10 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 16 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 119 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 116 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 116 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 115 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 35 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 35 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 34 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 35 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 125 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 125 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 126 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 129 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 29 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 30 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 28 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 31 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 122 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 124 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 124 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 124 KiB |
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user