markdown source builds

Auto-generated via `{sandpaper}` Source : b38b709 Branch : main Author : Chris Endemann <[email protected]> Time : 2024-11-26 15:56:17 +0000 Message : Update 7b-OOD-detection-softmax.md
carpentries-incubator · Nov 26, 2024 · 7a3228c · 7a3228c
1 parent d067f6c
commit 7a3228c
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 4 deletions.
diff --git a/7b-OOD-detection-softmax.md b/7b-OOD-detection-softmax.md
@@ -105,7 +105,7 @@ def plot_data_sample(train_data, ood_data):
 ```python
 train_data, test_data, ood_data, train_labels, test_labels, ood_labels = prep_ID_OOD_datasests([0,1], [5])
 fig = plot_data_sample(train_data, ood_data)
-fig.savefig('../images/OOD-detection_image-data-preview.png', dpi=300, bbox_inches='tight')
+#fig.savefig('../images/OOD-detection_image-data-preview.png', dpi=300, bbox_inches='tight')
 plt.show()
 
 
@@ -424,7 +424,7 @@ def plot_metrics_vs_thresholds(thresholds, f1_scores, precisions, recalls, OOD_s
 ```
 ```python
 fig, best_f1_threshold, best_precision_threshold, best_recall_threshold = plot_metrics_vs_thresholds(thresholds, f1_scores, precisions, recalls, 'Softmax')
-fig.savefig('../images/OOD-detection_metrics_vs_softmax-thresholds.png', dpi=300, bbox_inches='tight')
+#fig.savefig('../images/OOD-detection_metrics_vs_softmax-thresholds.png', dpi=300, bbox_inches='tight')
 
 ```
 ![OOD-detection_metrics_vs_softmax-thresholds](https://raw.githubusercontent.com/carpentries-incubator/fair-explainable-ml/main/images/OOD-detection_metrics_vs_softmax-thresholds.png)
@@ -464,4 +464,4 @@ plt.show()
 - While simple and widely used, softmax-based methods have limitations, including sensitivity to threshold choices and reduced reliability in high-dimensional settings.
 - Understanding softmax-based OOD detection lays the groundwork for exploring more advanced techniques like energy-based detection.
 
-::::::::::::::::::::::::::::::::::::::::::::::::::
+::::::::::::::::::::::::::::::::::::::::::::::::::
diff --git a/md5sum.txt b/md5sum.txt
@@ -15,7 +15,7 @@
 "episodes/5d-gradcam.md" "852756b061ec23f8a6726d8f0eee315e" "site/built/5d-gradcam.md" "2024-11-26"
 "episodes/6-confidence-intervals.md" "d228c34adc703f989b318acdb52c8334" "site/built/6-confidence-intervals.md" "2024-11-15"
 "episodes/7a-OOD-detection-overview.md" "144b96f0cf33c23c5d72781836f3498a" "site/built/7a-OOD-detection-overview.md" "2024-11-15"
-"episodes/7b-OOD-detection-softmax.md" "af474663c54f97492ad91ca8e9b75187" "site/built/7b-OOD-detection-softmax.md" "2024-11-15"
+"episodes/7b-OOD-detection-softmax.md" "dbfb86a18556126be122e2e8ccb36459" "site/built/7b-OOD-detection-softmax.md" "2024-11-26"
 "episodes/7c-OOD-detection-energy.md" "9290897ecaf6854a229bff3d0045d7a0" "site/built/7c-OOD-detection-energy.md" "2024-11-15"
 "episodes/7d-OOD-detection-distance-based.md" "84ad8f1a14dc547a03df47b65aed2483" "site/built/7d-OOD-detection-distance-based.md" "2024-11-15"
 "episodes/7e-OOD-detection-algo-design.md" "84b8a1f7d8afacdb1e9f7fce0420b8da" "site/built/7e-OOD-detection-algo-design.md" "2024-11-15"