BUG: Fix CSV parsing of tiles dataset columns with commas (#544)

microsoft · Jul 29, 2022 · 87237db · 87237db
1 parent 090368c
commit 87237db
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 1 deletion.
diff --git a/hi-ml-cpath/src/health_cpath/datasets/default_paths.py b/hi-ml-cpath/src/health_cpath/datasets/default_paths.py
@@ -10,7 +10,7 @@
 TCGA_CRCK_DATASET_ID = "TCGA-CRCk"
 
 TCGA_PRAD_DATASET_ID = "TCGA-PRAD_20220712"
-TCGA_PRAD_10X_TILES_DATASET_ID = "TCGA-PRAD_10X_tiles_20220722_103508_level1_224"
+TCGA_PRAD_10X_TILES_DATASET_ID = "TCGA-PRAD_10X_tiles_20220728_123811_level1_224"
 
 DEFAULT_DATASET_LOCATION = "/tmp/datasets/"
 PANDA_5X_TILES_DATASET_DIR = DEFAULT_DATASET_LOCATION + PANDA_5X_TILES_DATASET_ID

diff --git a/hi-ml-cpath/src/health_cpath/preprocessing/create_tiles_dataset.py b/hi-ml-cpath/src/health_cpath/preprocessing/create_tiles_dataset.py
@@ -125,8 +125,11 @@ def format_csv_row(tile_info: Dict[TileKey, Any], keys_to_save: Iterable[TileKey
     :return: The formatted CSV row.
     """
     tile_slide_metadata = tile_info.pop(TileKey.SLIDE_METADATA)
+
     fields = [str(tile_info[key]) for key in keys_to_save]
     fields.extend(str(tile_slide_metadata[key]) for key in metadata_keys)
+    fields = ['"' + value + '"' if ',' in value else value for value in fields]  # if field contains a , add extra " "
+
     dataset_row = ','.join(fields)
     return dataset_row