Update dataset names from array to dictionary (#9000)

* Migrate dataset names to dictionary

* fix check

* backwards compat

* predict fix

* val fix

* Keep dataset stats behavior identical

Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>

Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>

utils/dataloaders.py

@@ -1004,7 +1004,7 @@ class HUBDatasetStats():
         self.hub_dir = Path(data['path'] + '-hub')
         self.im_dir = self.hub_dir / 'images'
         self.im_dir.mkdir(parents=True, exist_ok=True)  # makes /images
-        self.stats = {'nc': data['nc'], 'names': data['names']}  # statistics dictionary
+        self.stats = {'nc': data['nc'], 'names': list(data['names'].values())}  # statistics dictionary
         self.data = data
 
     @staticmethod

utils/general.py

@@ -481,11 +481,11 @@ def check_dataset(data, autodownload=True):
             data = yaml.safe_load(f)  # dictionary
 
     # Checks
-    for k in 'train', 'val', 'nc':
+    for k in 'train', 'val', 'names':
         assert k in data, f"data.yaml '{k}:' field missing ❌"
-    if 'names' not in data:
-        LOGGER.warning("data.yaml 'names:' field missing ⚠️, assigning default names 'class0', 'class1', etc.")
-        data['names'] = [f'class{i}' for i in range(data['nc'])]  # default names
+    if isinstance(data['names'], (list, tuple)):  # old array format
+        data['names'] = dict(enumerate(data['names']))  # convert to dict
+    data['nc'] = len(data['names'])
 
     # Resolve paths
     path = Path(extract_dir or data.get('path') or '')  # optional 'path' default to '.'