Open laurencejackson opened 3 years ago
currently, multi GPU training on the DGX will create an mlflow run for each GPU process, making it difficult to review results.
https://github.com/mlflow/mlflow/issues/1879
currently, multi GPU training on the DGX will create an mlflow run for each GPU process, making it difficult to review results.