enable deterministic training

README.md

@@ -15,7 +15,7 @@ The [PyTorch model](https://drive.google.com/file/d/14CS-s1uv2q6WedYQGeFbZeEWIko
 ## Data
 The datasets used in this work were provided by [Activ Surgical](https://www.activsurgical.com/).
 
-We've provided a [link](https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/inbody_outbody_samples.zip) of 20 samples (10 in-body and 10 out-body) to show what this dataset looks like.
+Since datasets are private, we provide a [link](https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/inbody_outbody_samples.zip) of 20 samples (10 in-body and 10 out-body) to show what they look like.
 
 ### Preprocessing
 After downloading this dataset, python script in `scripts` folder named `data_process` can be used to generate label json files by running the command below and modifying `datapath` to path of unziped downloaded data. Generated label json files will be stored in `label` folder under the bundle path.
@@ -112,11 +112,7 @@ The classification result of every images in `test.json` will be printed to the
 #### Export checkpoint to TorchScript file:
 
 ```
-python -m monai.bundle ckpt_export network_def \
-    --filepath models/model.ts \
-    --ckpt_file models/model.pt \
-    --meta_file configs/metadata.json \
-    --config_file configs/inference.json
+python -m monai.bundle ckpt_export network_def --filepath models/model.ts --ckpt_file models/model.pt --meta_file configs/metadata.json --config_file configs/inference.json
 ```
 
 # References

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": "0.3.7",
+    "version": "0.3.8",
     "changelog": {
+        "0.3.8": "enable deterministic training",
         "0.3.7": "adapt to BundleWorkflow interface",
         "0.3.6": "add name tag",
         "0.3.5": "fix a comment issue in the data_process script",
@@ -16,7 +17,7 @@
         "0.1.0": "complete the first version model package",
         "0.0.1": "initialize the model package structure"
     },
-    "monai_version": "1.2.0rc3",
+    "monai_version": "1.2.0rc4",
     "pytorch_version": "1.13.1",
     "numpy_version": "1.22.2",
     "optional_packages_version": {

configs/multi_gpu_train.json

@@ -28,8 +28,7 @@
         "$import torch.distributed as dist",
         "$dist.is_initialized() or dist.init_process_group(backend='nccl')",
         "$torch.cuda.set_device(@device)",
-        "$monai.utils.set_determinism(seed=123)",
-        "$setattr(torch.backends.cudnn, 'benchmark', True)"
+        "$monai.utils.set_determinism(seed=123)"
     ],
     "run": [
         "$@train#trainer.run()"

configs/train.json

@@ -252,8 +252,7 @@
         }
     },
     "initialize": [
-        "$monai.utils.set_determinism(seed=123)",
-        "$setattr(torch.backends.cudnn, 'benchmark', True)"
+        "$monai.utils.set_determinism(seed=123)"
     ],
     "run": [
         "$@train#trainer.run()"

docs/README.md

@@ -8,7 +8,7 @@ The [PyTorch model](https://drive.google.com/file/d/14CS-s1uv2q6WedYQGeFbZeEWIko
 ## Data
 The datasets used in this work were provided by [Activ Surgical](https://www.activsurgical.com/).
 
-We've provided a [link](https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/inbody_outbody_samples.zip) of 20 samples (10 in-body and 10 out-body) to show what this dataset looks like.
+Since datasets are private, we provide a [link](https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/inbody_outbody_samples.zip) of 20 samples (10 in-body and 10 out-body) to show what they look like.
 
 ### Preprocessing
 After downloading this dataset, python script in `scripts` folder named `data_process` can be used to generate label json files by running the command below and modifying `datapath` to path of unziped downloaded data. Generated label json files will be stored in `label` folder under the bundle path.
@@ -105,11 +105,7 @@ The classification result of every images in `test.json` will be printed to the
 #### Export checkpoint to TorchScript file:
 
 ```
-python -m monai.bundle ckpt_export network_def \
-    --filepath models/model.ts \
-    --ckpt_file models/model.pt \
-    --meta_file configs/metadata.json \
-    --config_file configs/inference.json
+python -m monai.bundle ckpt_export network_def --filepath models/model.ts --ckpt_file models/model.pt --meta_file configs/metadata.json --config_file configs/inference.json
 ```
 
 # References