Merge pull request #5 from nttcslab/v2.x.x/v2.0.1

fixed decision threshold
nttcslab · Jan 13, 2024 · 2f861ba · 2f861ba
2 parents 2605ba1 + 17dd8da
commit 2f861ba
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -306,6 +306,22 @@ We developed and tested the source code on Ubuntu 18.04.6 LTS.
 - seaborn == 0.12.2
 - fasteners == 0.18
 
+## Change Log
+
+### [2.0.1](https://github.com/nttcslab/dcase2023_task2_baseline_ae/releases/tag/v2.0.1)
+
+#### Fixed
+
+- Fix anomaly score distribution.
+- Decision threshold has changed, but AUC, pAUC, etc. have not changed.
+
+### [2.0.0](https://github.com/nttcslab/dcase2023_task2_baseline_ae/releases/tag/v2.0.0)
+
+#### Added
+
+- provides support for the legacy datasets used in DCASE2020, 2021, 2022, and 2023.
+
+
 ## Citation
 
 If you use this system, please cite all the following four papers:

diff --git a/networks/dcase2023t2_ae/dcase2023t2_ae.py b/networks/dcase2023t2_ae/dcase2023t2_ae.py
@@ -199,7 +199,7 @@ def train(self, epoch):
                 val_loss += float(loss)
 
                 # calculate y_pred for fitting anomaly score distribution
-                y_pred.append(self.loss.item())
+                y_pred.append(loss.item())
 
         if not is_calc_cov:
             # plot train log