Add alternative label support to WDS for imagenet22k/12k split, add 21k/22k/12k indices filters to results/

timm/data/parsers/parser_wds.py

@@ -45,6 +45,7 @@ class SplitInfo:
     num_samples: int
     filenames: Tuple[str]
     shard_lengths: Tuple[int] = ()
+    alt_label: str = ''
     name: str = ''
 
 
@@ -54,6 +55,7 @@ def _parse_split_info(split: str, info: Dict):
             num_samples=dict_info['num_samples'],
             filenames=tuple(dict_info['filenames']),
             shard_lengths=tuple(dict_info['shard_lengths']),
+            alt_label=dict_info.get('alt_label', ''),
             name=dict_info['name'],
         )
 
@@ -98,7 +100,7 @@ def _parse_split_info(split: str, info: Dict):
     return split_info
 
 
-def _decode(sample, image_key='jpg', image_format='RGB', target_key='cls'):
+def _decode(sample, image_key='jpg', image_format='RGB', target_key='cls', alt_label=''):
     """ Custom sample decode
     * decode and convert PIL Image
     * cls byte string label to int
@@ -109,6 +111,12 @@ def _decode(sample, image_key='jpg', image_format='RGB', target_key='cls'):
         img.load()
     if image_format:
         img = img.convert(image_format)
+    if alt_label:
+        # alternative labels are encoded in json metadata
+        assert 'json' in sample
+        meta = json.loads(sample['json'])
+        return dict(jpg=img, cls=int(meta[alt_label]), json=meta)
+    else:
         return dict(jpg=img, cls=int(sample[target_key]), json=sample.get('json', None))
 
 
@@ -209,7 +217,11 @@ class ParserWebdataset(Parser):
                 wds.tarfile_to_samples(),
             ])
         pipeline.extend([
-            wds.map(partial(_decode, image_key=self.image_key, image_format=self.image_format))
+            wds.map(partial(
+                _decode,
+                image_key=self.image_key,
+                image_format=self.image_format,
+                alt_label=self.split_info.alt_label))
         ])
         self.ds = wds.DataPipeline(*pipeline)
         self.init_count += 1