[Fix] Fix bug in non-distributed multi-gpu training/testing (open-mml…

…ab#7019)

* Fix bug in non-distributed training/testing

* Add deprecated message

tools/test.py

@@ -36,7 +36,13 @@ def parse_args():
         '--gpu-ids',
         type=int,
         nargs='+',
-        help='ids of gpus to use '
+        help='(Deprecated, please use --gpu-id) ids of gpus to use '
+        '(only applicable to non-distributed training)')
+    parser.add_argument(
+        '--gpu-id',
+        type=int,
+        default=0,
+        help='id of gpu to use '
         '(only applicable to non-distributed testing)')
     parser.add_argument(
         '--format-only',
@@ -157,19 +163,17 @@ def main():
                 [ds_cfg.pop('samples_per_gpu', 1) for ds_cfg in cfg.data.test])
 
     if args.gpu_ids is not None:
-        cfg.gpu_ids = args.gpu_ids
+        cfg.gpu_ids = args.gpu_ids[0:1]
+        warnings.warn('`--gpu-ids` is deprecated, please use `--gpu-id`. '
+                      'Because we only support single GPU mode in '
+                      'non-distributed testing. Use the first GPU '
+                      'in `gpu_ids` now.')
     else:
-        cfg.gpu_ids = range(1)
+        cfg.gpu_ids = [args.gpu_id]
 
     # init distributed env first, since logger depends on the dist info.
     if args.launcher == 'none':
         distributed = False
-        if len(cfg.gpu_ids) > 1:
-            warnings.warn(
-                f'We treat {cfg.gpu_ids} as gpu-ids, and reset to '
-                f'{cfg.gpu_ids[0:1]} as gpu-ids to avoid potential error in '
-                'non-distribute testing time.')
-            cfg.gpu_ids = cfg.gpu_ids[0:1]
     else:
         distributed = True
         init_dist(args.launcher, **cfg.dist_params)

tools/train.py

@@ -37,13 +37,19 @@ def parse_args():
     group_gpus.add_argument(
         '--gpus',
         type=int,
-        help='number of gpus to use '
+        help='(Deprecated, please use --gpu-id) number of gpus to use '
         '(only applicable to non-distributed training)')
     group_gpus.add_argument(
         '--gpu-ids',
         type=int,
         nargs='+',
-        help='ids of gpus to use '
+        help='(Deprecated, please use --gpu-id) ids of gpus to use '
+        '(only applicable to non-distributed training)')
+    group_gpus.add_argument(
+        '--gpu-id',
+        type=int,
+        default=0,
+        help='id of gpu to use '
         '(only applicable to non-distributed training)')
     parser.add_argument('--seed', type=int, default=None, help='random seed')
     parser.add_argument(
@@ -113,20 +119,23 @@ def main():
     if args.resume_from is not None:
         cfg.resume_from = args.resume_from
     cfg.auto_resume = args.auto_resume
+    if args.gpus is not None:
+        cfg.gpu_ids = range(1)
+        warnings.warn('`--gpus` is deprecated because we only support '
+                      'single GPU mode in non-distributed training. '
+                      'Use `gpus=1` now.')
     if args.gpu_ids is not None:
-        cfg.gpu_ids = args.gpu_ids
-    else:
-        cfg.gpu_ids = range(1) if args.gpus is None else range(args.gpus)
+        cfg.gpu_ids = args.gpu_ids[0:1]
+        warnings.warn('`--gpu-ids` is deprecated, please use `--gpu-id`. '
+                      'Because we only support single GPU mode in '
+                      'non-distributed training. Use the first GPU '
+                      'in `gpu_ids` now.')
+    if args.gpus is None and args.gpu_ids is None:
+        cfg.gpu_ids = [args.gpu_id]
 
     # init distributed env first, since logger depends on the dist info.
     if args.launcher == 'none':
         distributed = False
-        if len(cfg.gpu_ids) > 1:
-            warnings.warn(
-                f'We treat {cfg.gpu_ids} as gpu-ids, and reset to '
-                f'{cfg.gpu_ids[0:1]} as gpu-ids to avoid potential error in '
-                'non-distribute training time.')
-            cfg.gpu_ids = cfg.gpu_ids[0:1]
     else:
         distributed = True
         init_dist(args.launcher, **cfg.dist_params)