explicitly log number of images per contrast per split to datalist json

monai/create_msd_data.py

@@ -313,6 +313,20 @@ def main():
     logger.info(f"Number of validation images (not subjects): {params['numValidationImagesTotal']}")
     logger.info(f"Number of testing images (not subjects): {params['numTestImagesTotal']}")
 
+    # update the dataframe to remove subjects whose labels don't exist
+    df = df[~df['subjectID'].isin(subjects_to_remove)]
+
+    # log the number of images per contrasts
+    params["numImagesPerContrast"] = {
+        "train": {},
+        "validation": {},
+        "test": {},
+    }
+    for contrast in params["contrasts"]:
+        params["numImagesPerContrast"]["train"][contrast] = len(df[(df['subjectID'].isin(train_subs_all)) & (df['contrastID'] == contrast)])
+        params["numImagesPerContrast"]["validation"][contrast] = len(df[(df['subjectID'].isin(val_subs_all)) & (df['contrastID'] == contrast)])
+        params["numImagesPerContrast"]["test"][contrast] = len(df[(df['contrastID'] == contrast) & (df['subjectID'].isin(test_subs_all))])
+
     # dump train/val/test splits into a yaml file
     with open(f"datasplits/datasplit_{dataset_name}_seed{args.seed}.yaml", 'w') as file:
         yaml.dump({'train': sorted(train_subs_all), 'val': sorted(val_subs_all), 'test': sorted(test_subs_all)}, file, indent=2, sort_keys=True)