Leave TcgaPradDataset as an example

microsoft · Dec 16, 2021 · f599e61 · f599e61
1 parent fa5ec34
commit f599e61
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 37 deletions.
diff --git a/InnerEye/ML/Histopathology/datasets/tcga_prad_dataset.py b/InnerEye/ML/Histopathology/datasets/tcga_prad_dataset.py
@@ -36,6 +36,7 @@ def __init__(self, root: Union[str, Path],
  from the dataset CSV file, e.g. after some filtering. If given, overrides `dataset_csv`.
  """
  super().__init__(root, dataset_csv, dataset_df, validate_columns=False)
- self.dataset_df[self.LABEL_COLUMN] = (self.dataset_df['label1_mutation']
- | self.dataset_df['label2_mutation']).astype(int)
+ # Example of how to define a custom label column from existing columns:
+ self.dataset_df[self.LABEL_COLUMN] = (self.dataset_df['label1']
+ | self.dataset_df['label2']).astype(int)
  self.validate_columns()
diff --git a/InnerEye/ML/Histopathology/preprocessing/create_tiles_dataset.py b/InnerEye/ML/Histopathology/preprocessing/create_tiles_dataset.py
@@ -293,6 +293,7 @@ def main(slides_dataset: SlidesDataset, root_output_dir: Union[str, Path],
 if __name__ == '__main__':
  from InnerEye.ML.Histopathology.datasets.tcga_prad_dataset import TcgaPradDataset
 
+ # Example set up for an existing slides dataset:
  main(slides_dataset=TcgaPradDataset("/tmp/datasets/TCGA-PRAD"),
  root_output_dir="/datadrive/TCGA-PRAD_tiles",
  n_slides=5,

diff --git a/Tests/ML/histopathology/datasets/test_tcga_prad_dataset.py b/Tests/ML/histopathology/datasets/test_tcga_prad_dataset.py