Notebooks classificador terminados

classifier/notebooks/02_preprocessing.ipynb

@@ -327,250 +327,93 @@
     "## 2.1 Phase 2 augmentation settings\n",
     "<!-- augmentation-table -->\n",
     "\n",
-    "The Phase 2 augmentation configs define a deliberately light policy. The table below is generated from `classifier/configs/phase2/*.json` and `DFFImagePipeline.DEFAULTS`, so it documents the actual values available to training rather than a manually copied list.\n"
+    "The Phase 2 augmentation runs use one shared light policy. Instead of listing every pipeline default, the next cell reads the actual Phase 2 config files and reports only the stochastic operations that are active during training. Values marked `config` are written in the run JSON files; values marked `pipeline default` are active because the config gives partial overrides and the pipeline fills the rest from `DFFImagePipeline.DEFAULTS`.\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "428047fa",
    "metadata": {},
-   "outputs": [
+   "outputs": [],
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th>run</th>\n",
-       "      <th>p2d_resnet18_aug</th>\n",
-       "      <th>p2d_simplecnn_aug</th>\n",
-       "      <th>p2e_resnet18_facecrop_aug</th>\n",
-       "      <th>p2e_simplecnn_facecrop_aug</th>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>setting</th>\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>blur_p</th>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>blur_radius</th>\n",
-       "      <td>[0.1, 1.5]</td>\n",
-       "      <td>[0.1, 1.5]</td>\n",
-       "      <td>[0.1, 1.5]</td>\n",
-       "      <td>[0.1, 1.5]</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>brightness</th>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>center_jitter</th>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>contrast</th>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>crop_scale</th>\n",
-       "      <td>[0.85, 1.0]</td>\n",
-       "      <td>[0.85, 1.0]</td>\n",
-       "      <td>[0.85, 1.0]</td>\n",
-       "      <td>[0.85, 1.0]</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>erase_p</th>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "      <td>0.2</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>erase_scale</th>\n",
-       "      <td>[0.02, 0.15]</td>\n",
-       "      <td>[0.02, 0.15]</td>\n",
-       "      <td>[0.02, 0.15]</td>\n",
-       "      <td>[0.02, 0.15]</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>grayscale_p</th>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>hflip_p</th>\n",
-       "      <td>0.5</td>\n",
-       "      <td>0.5</td>\n",
-       "      <td>0.5</td>\n",
-       "      <td>0.5</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>hue</th>\n",
-       "      <td>0.02</td>\n",
-       "      <td>0.02</td>\n",
-       "      <td>0.02</td>\n",
-       "      <td>0.02</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>jpeg_p</th>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>jpeg_quality</th>\n",
-       "      <td>[65, 95]</td>\n",
-       "      <td>[65, 95]</td>\n",
-       "      <td>[65, 95]</td>\n",
-       "      <td>[65, 95]</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>noise_p</th>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "      <td>0.3</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>noise_std</th>\n",
-       "      <td>0.04</td>\n",
-       "      <td>0.04</td>\n",
-       "      <td>0.04</td>\n",
-       "      <td>0.04</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>rotation_degrees</th>\n",
-       "      <td>10</td>\n",
-       "      <td>10</td>\n",
-       "      <td>10</td>\n",
-       "      <td>10</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>saturation</th>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "      <td>0.1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "run              p2d_resnet18_aug p2d_simplecnn_aug p2e_resnet18_facecrop_aug  \\\n",
-       "setting                                                                         \n",
-       "blur_p                        0.1               0.1                       0.1   \n",
-       "blur_radius            [0.1, 1.5]        [0.1, 1.5]                [0.1, 1.5]   \n",
-       "brightness                    0.2               0.2                       0.2   \n",
-       "center_jitter                 0.1               0.1                       0.1   \n",
-       "contrast                      0.2               0.2                       0.2   \n",
-       "crop_scale            [0.85, 1.0]       [0.85, 1.0]               [0.85, 1.0]   \n",
-       "erase_p                       0.2               0.2                       0.2   \n",
-       "erase_scale          [0.02, 0.15]      [0.02, 0.15]              [0.02, 0.15]   \n",
-       "grayscale_p                   0.1               0.1                       0.1   \n",
-       "hflip_p                       0.5               0.5                       0.5   \n",
-       "hue                          0.02              0.02                      0.02   \n",
-       "jpeg_p                        0.3               0.3                       0.3   \n",
-       "jpeg_quality             [65, 95]          [65, 95]                  [65, 95]   \n",
-       "noise_p                       0.3               0.3                       0.3   \n",
-       "noise_std                    0.04              0.04                      0.04   \n",
-       "rotation_degrees               10                10                        10   \n",
-       "saturation                    0.1               0.1                       0.1   \n",
-       "\n",
-       "run              p2e_simplecnn_facecrop_aug  \n",
-       "setting                                      \n",
-       "blur_p                                  0.1  \n",
-       "blur_radius                      [0.1, 1.5]  \n",
-       "brightness                              0.2  \n",
-       "center_jitter                           0.1  \n",
-       "contrast                                0.2  \n",
-       "crop_scale                      [0.85, 1.0]  \n",
-       "erase_p                                 0.2  \n",
-       "erase_scale                    [0.02, 0.15]  \n",
-       "grayscale_p                             0.1  \n",
-       "hflip_p                                 0.5  \n",
-       "hue                                    0.02  \n",
-       "jpeg_p                                  0.3  \n",
-       "jpeg_quality                       [65, 95]  \n",
-       "noise_p                                 0.3  \n",
-       "noise_std                              0.04  \n",
-       "rotation_degrees                         10  \n",
-       "saturation                              0.1  "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
    "source": [
     "# <!-- augmentation-table -->\n",
-    "phase2_aug_runs = [\n",
+    "aug_run_paths = sorted((CONFIGS_DIR / \"phase2\").glob(\"*.json\"))\n",
-    "    \"p2d_resnet18_aug\",\n",
+    "aug_runs = []\n",
-    "    \"p2d_simplecnn_aug\",\n",
+    "for cfg_path in aug_run_paths:\n",
-    "    \"p2e_resnet18_facecrop_aug\",\n",
-    "    \"p2e_simplecnn_facecrop_aug\",\n",
-    "]\n",
-    "\n",
-    "rows = []\n",
-    "for run_name in phase2_aug_runs:\n",
-    "    cfg_path = CONFIGS_DIR / \"phase2\" / f\"{run_name}.json\"\n",
     "    cfg = load_config(str(cfg_path))\n",
-    "    aug_cfg = cfg.get(\"augment\") or {}\n",
+    "    if cfg.get(\"augment\"):\n",
-    "    merged = {**DFFImagePipeline.DEFAULTS, **aug_cfg}\n",
+    "        aug_runs.append((cfg_path.stem, cfg))\n",
-    "    for key in [\n",
-    "        \"crop_scale\", \"center_jitter\", \"hflip_p\", \"rotation_degrees\",\n",
-    "        \"brightness\", \"contrast\", \"saturation\", \"hue\", \"grayscale_p\",\n",
-    "        \"blur_p\", \"blur_radius\", \"jpeg_p\", \"jpeg_quality\", \"erase_p\",\n",
-    "        \"erase_scale\", \"noise_p\", \"noise_std\",\n",
-    "    ]:\n",
-    "        rows.append({\n",
-    "            \"run\": run_name,\n",
-    "            \"data_dir\": cfg.get(\"data_dir\"),\n",
-    "            \"setting\": key,\n",
-    "            \"value\": merged.get(key),\n",
-    "            \"explicit_in_config\": key in aug_cfg,\n",
-    "        })\n",
     "\n",
-    "augmentation_table = pd.DataFrame(rows)\n",
+    "active_keys = []\n",
-    "display(augmentation_table.pivot(index=\"setting\", columns=\"run\", values=\"value\"))\n"
+    "for key in DFFImagePipeline.DEFAULTS:\n",
+    "    values = []\n",
+    "    for _, cfg in aug_runs:\n",
+    "        merged = {**DFFImagePipeline.DEFAULTS, **cfg.get(\"augment\", {})}\n",
+    "        value = merged.get(key)\n",
+    "        values.append(value)\n",
+    "    if any(value not in (None, 0, 0.0, [1.0, 1.0], [1, 1]) for value in values):\n",
+    "        active_keys.append(key)\n",
+    "\n",
+    "setting_groups = {\n",
+    "    \"crop_scale\": \"crop jitter\",\n",
+    "    \"center_jitter\": \"crop jitter\",\n",
+    "    \"hflip_p\": \"geometry\",\n",
+    "    \"rotation_degrees\": \"geometry\",\n",
+    "    \"brightness\": \"color\",\n",
+    "    \"contrast\": \"color\",\n",
+    "    \"saturation\": \"color\",\n",
+    "    \"hue\": \"color\",\n",
+    "    \"grayscale_p\": \"color\",\n",
+    "    \"blur_p\": \"quality degradation\",\n",
+    "    \"blur_radius\": \"quality degradation\",\n",
+    "    \"jpeg_p\": \"quality degradation\",\n",
+    "    \"jpeg_quality\": \"quality degradation\",\n",
+    "    \"erase_p\": \"occlusion/noise\",\n",
+    "    \"erase_scale\": \"occlusion/noise\",\n",
+    "    \"noise_p\": \"occlusion/noise\",\n",
+    "    \"noise_std\": \"occlusion/noise\",\n",
+    "}\n",
+    "\n",
+    "def compact_value(value):\n",
+    "    if isinstance(value, float):\n",
+    "        return f\"{value:g}\"\n",
+    "    if isinstance(value, list):\n",
+    "        return \"[\" + \", \".join(compact_value(v) for v in value) + \"]\"\n",
+    "    return str(value)\n",
+    "\n",
+    "summary_rows = []\n",
+    "for key in active_keys:\n",
+    "    merged_values = []\n",
+    "    explicit = []\n",
+    "    for run_name, cfg in aug_runs:\n",
+    "        aug_cfg = cfg.get(\"augment\", {})\n",
+    "        merged = {**DFFImagePipeline.DEFAULTS, **aug_cfg}\n",
+    "        merged_values.append(merged.get(key))\n",
+    "        explicit.append(key in aug_cfg)\n",
+    "    same_value = all(value == merged_values[0] for value in merged_values)\n",
+    "    source = \"config\" if any(explicit) else \"pipeline default\"\n",
+    "    summary_rows.append({\n",
+    "        \"group\": setting_groups.get(key, \"other\"),\n",
+    "        \"setting\": key,\n",
+    "        \"phase2_value\": compact_value(merged_values[0]) if same_value else \"varies by run\",\n",
+    "        \"value_source\": source,\n",
+    "    })\n",
+    "\n",
+    "augmentation_summary = pd.DataFrame(summary_rows)\n",
+    "run_summary = pd.DataFrame([\n",
+    "    {\n",
+    "        \"run\": run_name,\n",
+    "        \"model\": cfg.get(\"backbone\"),\n",
+    "        \"data_dir\": cfg.get(\"data_dir\"),\n",
+    "        \"explicit_aug_keys\": len(cfg.get(\"augment\", {})),\n",
+    "    }\n",
+    "    for run_name, cfg in aug_runs\n",
+    "])\n",
+    "\n",
+    "print(\"Augmentation runs read from configs:\")\n",
+    "display(run_summary)\n",
+    "display(augmentation_summary)\n"
    ]
   },
   {
@@ -579,7 +422,9 @@
    "metadata": {},
    "source": [
     "<!-- augmentation-table -->\n",
-    "**Readout:** Phase 2 explicitly reduces several defaults: rotation is `10` degrees instead of the pipeline default `15`, color jitter is milder, blur probability is `0.1`, erase probability is `0.2`, and Gaussian noise uses `p=0.3`, `std=0.04`. JPEG recompression is inherited from the pipeline defaults when augmentation is enabled. Validation/test still bypass all stochastic operations because their pipeline is called with `train=False`.\n"
+    "These settings are intentionally modest. Crop jitter, flips, and rotation test whether the detector is robust to small pose and framing changes. Color jitter, grayscale, blur, JPEG recompression, and Gaussian noise test whether the model survives normal image-quality variation. Random erase is the strongest perturbation because it can hide local evidence.\n",
+    "\n",
+    "That last point is also the risk: deepfake cues can be subtle, so augmentation can remove useful evidence as well as nuisance variation. This is why Phase 2 treats augmentation as an ablation rather than assuming it will help. Validation and test transforms still use `train=False`, so they keep only deterministic square crop, resize, and normalization.\n"
    ]
   },
   {
@@ -737,7 +582,7 @@
    "metadata": {},
    "source": [
     "<!-- facecrop-augmentation-examples -->\n",
-    "**Readout:** evaluation stays stable, while training draws can flip, rotate, jitter color, blur, erase, add noise, or recompress the cropped face. This is exactly the question later tested in Phase 2: whether extra variation helps the facecrop model generalize or over-regularizes it.\n"
+    "For facecropped data, evaluation remains deterministic but training draws can still flip, rotate, jitter color, blur, erase, add noise, or recompress the cropped face. This is the exact Phase 2 question: after the input has already been focused on the face, does extra variation improve robustness or does it hide the evidence the model needs?\n"
    ]
   },
   {
@@ -857,16 +702,6 @@
     "For the report, this table supports the normalization ablation in Phase 2. The actual decision is made in `04_phase2_analysis.ipynb`, where `real_norm` is compared against ImageNet/default using the saved logs.\n"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "id": "fb95d062",
-   "metadata": {},
-   "source": [
-    "## 5. Reproducibility note\n",
-    "\n",
-    "These checks are not preprocessing operations. They simply confirm that the experiment setup is safe: identity groups do not leak across splits, validation/test transforms are deterministic, source-specific metrics handle edge cases, and config inheritance works as expected. The full test suite lives in `classifier/tests/`.\n"
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "b02fd790",

classifier/notebooks/03_phase1_analysis.ipynb

@@ -380,7 +380,7 @@
    "metadata": {},
    "source": [
     "<!-- phase1-protocol -->\n",
-    "**Readout:** both Phase 1 runs share seed `42`, `5` folds, batch size `32`, learning rate `1e-4`, weight decay `1e-4`, cosine `T_max=15`, and early-stopping patience `5`. The only intended comparison is model capacity/pretraining: SimpleCNN from scratch versus pretrained ResNet18.\n"
+    "Both Phase 1 runs use the same protocol: seed `42`, `5` folds, batch size `32`, learning rate `1e-4`, weight decay `1e-4`, cosine `T_max=15`, and early-stopping patience `5`. The intended comparison is therefore model capacity and pretraining: SimpleCNN from scratch versus pretrained ResNet18.\n"
    ]
   },
   {
@@ -757,7 +757,7 @@
    "id": "97cfc057",
    "metadata": {},
    "source": [
-    "**Readout:** SimpleCNN improves slowly and its train/validation AUC curves stay close together, which suggests limited capacity rather than severe overfitting. ResNet18 learns much faster and reaches very high training AUC, but validation AUC plateaus around the low 0.93 range after the first few epochs. The gap between train and validation AUC means the pretrained model has enough capacity to fit the training folds more strongly than it generalizes to validation. This is not a failure, because ResNet18 still has much better validation and test performance than SimpleCNN, but it is a warning that later improvements must be checked on held-out folds and source-wise metrics rather than training curves alone.\n"
+    "SimpleCNN improves slowly and its train/validation AUC curves stay close together, which points more to limited capacity than severe overfitting. ResNet18 learns much faster and reaches very high training AUC, while validation AUC plateaus around the low `0.93` range after the first few epochs. That gap means the pretrained model can fit the training folds more strongly than it generalizes, so later improvements need to be checked on held-out folds and source-wise metrics, not training curves alone.\n"
    ]
   },
   {
@@ -815,17 +815,7 @@
    "id": "2ddecd94",
    "metadata": {},
    "source": [
-    "**Readout:** The confusion matrices show the same pattern as the AUC results, but in error-count form. SimpleCNN correctly classifies about `71%` of real images and `70%` of fake images, so it misses many examples in both directions. ResNet18 improves both sides: about `81%` of real images are kept real, and about `88%` of fake images are detected as fake. The biggest practical gain is fewer fake images predicted as real (`30%` -> `12%`), which matters because false negatives are the dangerous failure mode for a deepfake detector. ResNet18 still has some false alarms on real images (`19%`), so it is stronger but not perfect.\n"
+    "The confusion matrices show the AUC story in error-count form. SimpleCNN correctly classifies about `71%` of real images and `70%` of fake images, so it misses many examples in both directions. ResNet18 improves both sides: about `81%` of real images are kept real, and about `88%` of fake images are detected as fake. The most important practical gain is fewer fake images predicted as real (`30%` -> `12%`), although the model still produces some false alarms on real images (`19%`).\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "id": "0591bcea",
-   "metadata": {},
-   "source": [
-    "## 6. Reproducibility note\n",
-    "\n",
-    "These checks are not extra results. They simply support the credibility of the Phase 1 comparison: the same config-loading rules, grouped splits, deterministic evaluation transforms, and safe metric handling are covered by tests in `classifier/tests/`. This lets the baseline comparison focus on the intended difference: SimpleCNN versus pretrained ResNet18.\n"
    ]
   },
   {
@@ -833,7 +823,7 @@
    "id": "624839d8",
    "metadata": {},
    "source": [
-    "## Report-ready conclusion\n",
+    "## Conclusion\n",
     "\n",
     "Under identical Phase 1 conditions, ResNet18 is the stronger baseline. SimpleCNN reaches AUC `0.7786`, accuracy `0.7039`, and F1 `0.7801`. ResNet18 reaches AUC `0.9366`, accuracy `0.8650`, and F1 `0.9073`. The mean fold-wise AUC improvement is about `+0.1580`.\n",
     "\n",

classifier/outputs/analysis/gradcam/gradcam_artifact_summary.csv

@@ -1,16 +1,16 @@
-run,n_candidates,n_images,heldout_source,candidate_auc,candidate_acc,panel_path,expanded_panel_path
+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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p1_simplecnn_baseline\panel.png,
+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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p1_resnet18_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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2a_t1_original\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2a_t2_real_norm\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2a_t3_holdout_text2img\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2a_t3_holdout_inpainting\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2a_t3_holdout_insight\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2b_simplecnn_224\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2b_resnet18_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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2c_simplecnn_facecrop\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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2c_resnet18_facecrop\panel.png,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2c_resnet18_facecrop\panel_expanded.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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2d_simplecnn_aug\panel.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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2d_resnet18_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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2e_simplecnn_facecrop_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,c:\Users\diogo\Documents\MIA_UP\2_Semestre\DRL\DRL_2\DRL_PROJ\classifier\outputs\gradcam\p2e_resnet18_facecrop_aug\panel.png,
+p2e_resnet18_facecrop_aug,192,10,,0.9910300925925927,0.9322916666666666,classifier\outputs\gradcam\p2e_resnet18_facecrop_aug\panel.png,,

classifier/outputs/analysis/gradcam/p1_resnet18_baseline_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p1_simplecnn_baseline_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2a_t1_original_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2a_t2_real_norm_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2a_t3_holdout_inpainting_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2a_t3_holdout_insight_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2a_t3_holdout_text2img_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2b_resnet18_224_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2b_simplecnn_224_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2c_resnet18_facecrop_summary.json

@@ -199,5 +199,6 @@
     "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"
+  "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"
 }

classifier/outputs/analysis/gradcam/p2c_simplecnn_facecrop_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2d_resnet18_aug_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2d_simplecnn_aug_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2e_resnet18_facecrop_aug_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/gradcam/p2e_simplecnn_facecrop_aug_summary.json

@@ -133,5 +133,6 @@
     "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
+  "expanded_panel_path": null,
+  "fine_panel_path": null
 }

classifier/outputs/analysis/phase2_story_summary.json

@@ -6,33 +6,32 @@
     {
       "choice": "input size",
       "decision": "224x224",
-      "evidence": "ResNet18 improves from 0.9366 to 0.9660 AUC.",
+      "evidence": "ResNet18 improves from AUC 0.9366 to 0.9660.",
       "confidence": "high"
     },
     {
       "choice": "face crop",
       "decision": "enable",
-      "evidence": "Best run is p2c_resnet18_facecrop with AUC 0.9755.",
+      "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 is 0.9737, below facecrop-only 0.9755; SimpleCNN drops sharply.",
+      "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 and is less aligned with pretrained weights.",
+      "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": "Holdout text2img and insight pairwise AUC drop to 0.7595 and 0.8421.",
+      "evidence": "Holding out text2img and insight drops pairwise AUC to 0.7595 and 0.8421.",
       "confidence": "high"
     }
-  ],
+  ]
-  "note": "Generated by 04_phase2_analysis.ipynb when this cell is executed."
 }

classifier/outputs/analysis/phase4_story_summary.json

@@ -0,0 +1,104 @@
+{
+  "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
+    }
+  ]
+}