diff --git a/src/mednet/scripts/upload.py b/src/mednet/scripts/upload.py index 615b81b25b08d9c8c07ae9e1dbaff1a9753f195e..f02d09343d0b5ecc524bfda2d3717d5e4cdc2291 100644 --- a/src/mednet/scripts/upload.py +++ b/src/mednet/scripts/upload.py @@ -125,9 +125,9 @@ def upload( ) # get train files + train_log_file = experiment_folder / "trainlog.pdf" train_folder = experiment_folder / "model" train_meta_file = train_folder / "meta.json" - train_log_file = train_folder / "trainlog.pdf" train_model_file = get_checkpoint_to_run_inference(train_folder) train_files = [train_meta_file, train_model_file, train_log_file] @@ -149,7 +149,6 @@ def upload( f"permitted maximum ({upload_limit_mb:.2f} MB)." ) - # prepare experiment and run names with train_meta_file.open("r") as meta_file: train_data = json.load(meta_file) @@ -157,6 +156,9 @@ def upload( evaluation_data = json.load(meta_file) evaluation_data = evaluation_data["test"] + # get lowest validation epoch + best_epoch = str(train_model_file).split(".")[0].split("=")[1] + experiment_name = ( experiment_name or f"{train_data['model-name']}-{train_data['database-name']}" @@ -177,6 +179,22 @@ def upload( click.echo("Uploading metrics...") + for k in [ + "epochs", + "batch-size", + ]: + click.secho(f" -> `{k}` ({train_data[k]})") + mlflow.log_param(k, train_data[k]) + + click.secho( + f" -> `#accumulations` ({train_data['accumulate-grad-batches']})" + ) + mlflow.log_param( + "#Accumulations", train_data["accumulate-grad-batches"] + ) + click.secho(f" -> `epoch (best)` ({best_epoch})") + mlflow.log_param("Epoch (best)", best_epoch) + for k in [ "threshold", "precision",