monai
medical
katielink commited on
Commit
11a1866
1 Parent(s): 31dc790

fix the wrong GPU index issue of multi-node

Browse files
configs/metadata.json CHANGED
@@ -1,7 +1,8 @@
1
  {
2
  "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
3
- "version": "0.4.9",
4
  "changelog": {
 
5
  "0.4.9": "remove error dollar symbol in readme",
6
  "0.4.8": "add RAM usage with CacheDataset",
7
  "0.4.7": "deterministic retrain benchmark",
@@ -26,7 +27,7 @@
26
  "0.1.0": "complete the model package",
27
  "0.0.1": "initialize the model package structure"
28
  },
29
- "monai_version": "1.2.0rc6",
30
  "pytorch_version": "1.13.1",
31
  "numpy_version": "1.22.2",
32
  "optional_packages_version": {
 
1
  {
2
  "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
3
+ "version": "0.5.0",
4
  "changelog": {
5
+ "0.5.0": "fix the wrong GPU index issue of multi-node",
6
  "0.4.9": "remove error dollar symbol in readme",
7
  "0.4.8": "add RAM usage with CacheDataset",
8
  "0.4.7": "deterministic retrain benchmark",
 
27
  "0.1.0": "complete the model package",
28
  "0.0.1": "initialize the model package structure"
29
  },
30
+ "monai_version": "1.2.0",
31
  "pytorch_version": "1.13.1",
32
  "numpy_version": "1.22.2",
33
  "optional_packages_version": {
configs/multi_gpu_train.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "device": "$torch.device(f'cuda:{dist.get_rank()}')",
3
  "network": {
4
  "_target_": "torch.nn.parallel.DistributedDataParallel",
5
  "module": "$@network_def.to(@device)",
 
1
  {
2
+ "device": "$torch.device('cuda:' + os.environ['LOCAL_RANK'])",
3
  "network": {
4
  "_target_": "torch.nn.parallel.DistributedDataParallel",
5
  "module": "$@network_def.to(@device)",