Addressing NCCL issue with binary classification for distributed trai…

…ning. dmlc/xgboost#7982 (comment) dmlc/xgboost#8257
NikhilRaverkar · Mar 17, 2023 · 5982cee · 5982cee
1 parent 6dcd442
commit 5982cee
Showing 1 changed file with 3 additions and 0 deletions.
diff --git a/docker/1.7-1/final/Dockerfile.cpu b/docker/1.7-1/final/Dockerfile.cpu
@@ -69,6 +69,9 @@ ENV SM_INPUT /opt/ml/input
 ENV SM_INPUT_TRAINING_CONFIG_FILE $SM_INPUT/config/hyperparameters.json
 ENV SM_INPUT_DATA_CONFIG_FILE $SM_INPUT/config/inputdataconfig.json
 ENV SM_CHECKPOINT_CONFIG_FILE $SM_INPUT/config/checkpointconfig.json
+# See: https://github.com/dmlc/xgboost/issues/7982#issuecomment-1379390906 https://github.com/dmlc/xgboost/pull/8257
+ENV NCCL_SOCKET_IFNAME eth 
+
 
 # Set SageMaker serving environment variables
 ENV SM_MODEL_DIR /opt/ml/model