add non-deterministic note

README.md

@@ -59,6 +59,9 @@ In Evaluation Mode: A list of dictionaries of predicted box, classification labe
 ## Performance
 Coco metric is used for evaluating the performance of the model. The pre-trained model was trained and validated on data fold 0. This model achieves a mAP=0.853, mAR=0.994, AP(IoU=0.1)=0.862, AR(IoU=0.1)=1.0.
 
+Please note that this bundle is non-deterministic because of the max pooling layer used in the network. Therefore, reproducing the training process may not get exactly the same performance.
+Please refer to https://pytorch.org/docs/stable/notes/randomness.html#reproducibility for more details about reproducibility.
+
 #### Training Loss
 ![A graph showing the detection train loss](https://developer.download.nvidia.com/assets/Clara/Images/monai_retinanet_detection_train_loss.png)
 
@@ -73,16 +76,19 @@ In addition to the Pythonic APIs, a few command line interfaces (CLI) are provid
 For more details usage instructions, visit the [MONAI Bundle Configuration Page](https://docs.monai.io/en/latest/config_syntax.html).
 
 #### Execute training:
+
 ```
 python -m monai.bundle run --config_file configs/train.json
 ```
 
 #### Override the `train` config to execute evaluation with the trained model:
+
 ```
 python -m monai.bundle run --config_file "['configs/train.json','configs/evaluate.json']"
 ```
 
 #### Execute inference on resampled LUNA16 images by setting `"whether_raw_luna16": false` in `inference.json`:
+
 ```
 python -m monai.bundle run --config_file configs/inference.json
 ```

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": "0.5.3",
+    "version": "0.5.4",
     "changelog": {
+        "0.5.4": "add non-deterministic note",
         "0.5.3": "adapt to BundleWorkflow interface",
         "0.5.2": "black autofix format and add name tag",
         "0.5.1": "modify dataset key name",
@@ -17,7 +18,7 @@
         "0.1.1": "add reference for LIDC dataset",
         "0.1.0": "complete the model package"
     },
-    "monai_version": "1.2.0rc3",
+    "monai_version": "1.2.0rc4",
     "pytorch_version": "1.13.1",
     "numpy_version": "1.22.2",
     "optional_packages_version": {

docs/README.md

@@ -52,6 +52,9 @@ In Evaluation Mode: A list of dictionaries of predicted box, classification labe
 ## Performance
 Coco metric is used for evaluating the performance of the model. The pre-trained model was trained and validated on data fold 0. This model achieves a mAP=0.853, mAR=0.994, AP(IoU=0.1)=0.862, AR(IoU=0.1)=1.0.
 
+Please note that this bundle is non-deterministic because of the max pooling layer used in the network. Therefore, reproducing the training process may not get exactly the same performance.
+Please refer to https://pytorch.org/docs/stable/notes/randomness.html#reproducibility for more details about reproducibility.
+
 #### Training Loss
 ![A graph showing the detection train loss](https://developer.download.nvidia.com/assets/Clara/Images/monai_retinanet_detection_train_loss.png)
 
@@ -66,16 +69,19 @@ In addition to the Pythonic APIs, a few command line interfaces (CLI) are provid
 For more details usage instructions, visit the [MONAI Bundle Configuration Page](https://docs.monai.io/en/latest/config_syntax.html).
 
 #### Execute training:
+
 ```
 python -m monai.bundle run --config_file configs/train.json
 ```
 
 #### Override the `train` config to execute evaluation with the trained model:
+
 ```
 python -m monai.bundle run --config_file "['configs/train.json','configs/evaluate.json']"
 ```
 
 #### Execute inference on resampled LUNA16 images by setting `"whether_raw_luna16": false` in `inference.json`:
+
 ```
 python -m monai.bundle run --config_file configs/inference.json
 ```