deterministic retrain benchmark

README.md

@@ -59,14 +59,14 @@ The training as performed with the following:
 - 13: Left adrenal gland
 
 ## Performance
-Dice score was used for evaluating the performance of the model. This model achieves a mean dice score of 0.8120
+Dice score was used for evaluating the performance of the model. This model achieves a mean dice score of 0.82
 
 #### Training Loss
-![The figure shows the training loss curve for 10K iterations.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_trainloss_v1.png)
+![The figure shows the training loss curve for 10K iterations.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_train_loss_v2.png)
 
 #### Validation Dice
 
-![A graph showing the validation mean Dice for 5000 epochs.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_validation_meandice_v1.png)
+![A graph showing the validation mean Dice for 5000 epochs.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_val_dice_v2.png)
 
 ## MONAI Bundle Commands
 In addition to the Pythonic APIs, a few command line interfaces (CLI) are provided to interact with the bundle. The CLI supports flexible use cases, such as overriding configs at runtime and predefining arguments in a file.

configs/evaluate.json

@@ -70,9 +70,6 @@
             "summary_ops": "*"
         }
     ],
-    "initialize": [
-        "$setattr(torch.backends.cudnn, 'benchmark', True)"
-    ],
     "run": [
         "$@validate#evaluator.run()"
     ]

configs/inference.json

@@ -136,7 +136,7 @@
         "amp": true
     },
     "initialize": [
-        "$setattr(torch.backends.cudnn, 'benchmark', True)"
+        "$monai.utils.set_determinism(seed=123)"
     ],
     "run": [
         "$@evaluator.run()"

configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
-    "version": "0.4.6",
+    "version": "0.4.7",
     "changelog": {
+        "0.4.7": "deterministic retrain benchmark",
         "0.4.6": "fix mgpu finalize issue",
         "0.4.5": "enable deterministic training",
         "0.4.4": "update numbers",
@@ -42,7 +43,7 @@
     "label_classes": "multi-channel data,0:background,1:spleen, 2:Right Kidney, 3:Left Kideny, 4:Gallbladder, 5:Esophagus, 6:Liver, 7:Stomach, 8:Aorta, 9:IVC, 10:Portal and Splenic Veins, 11:Pancreas, 12:Right adrenal gland, 13:Left adrenal gland",
     "pred_classes": "14 channels OneHot data, 0:background,1:spleen, 2:Right Kidney, 3:Left Kideny, 4:Gallbladder, 5:Esophagus, 6:Liver, 7:Stomach, 8:Aorta, 9:IVC, 10:Portal and Splenic Veins, 11:Pancreas, 12:Right adrenal gland, 13:Left adrenal gland",
     "eval_metrics": {
-        "mean_dice": 0.812
+        "mean_dice": 0.82
     },
     "intended_use": "This is an example, not to be used for diagnostic purposes",
     "references": [

docs/README.md

@@ -52,14 +52,14 @@ The training as performed with the following:
 - 13: Left adrenal gland
 
 ## Performance
-Dice score was used for evaluating the performance of the model. This model achieves a mean dice score of 0.8120
+Dice score was used for evaluating the performance of the model. This model achieves a mean dice score of 0.82
 
 #### Training Loss
-![The figure shows the training loss curve for 10K iterations.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_trainloss_v1.png)
+![The figure shows the training loss curve for 10K iterations.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_train_loss_v2.png)
 
 #### Validation Dice
 
-![A graph showing the validation mean Dice for 5000 epochs.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_validation_meandice_v1.png)
+![A graph showing the validation mean Dice for 5000 epochs.](https://developer.download.nvidia.com/assets/Clara/Images/monai_swin_unetr_btcv_segmentation_val_dice_v2.png)
 
 ## MONAI Bundle Commands
 In addition to the Pythonic APIs, a few command line interfaces (CLI) are provided to interact with the bundle. The CLI supports flexible use cases, such as overriding configs at runtime and predefining arguments in a file.

models/model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5486702e73e4ca3eef492e3b53cf91304302805ae49a9f5159038637da818bda
-size 256345027
+oid sha256:52e7c3114444e41bb14f644e0dd2b7d42d70ad4b4dec0c1bfa4a552a4b92a096
+size 256336065