feat: add device_target config and set default to Ascend. (#796)

Author: IASZHT

config.py

@@ -43,6 +43,8 @@ def create_parser():
                        help='Interval for print training log. Unit: step (default=100)')
     group.add_argument('--seed', type=int, default=42,
                        help='Seed value for determining randomness in numpy, random, and mindspore (default=42)')
+    group.add_argument('--device_target', type=str, default='Ascend',
+                       help='Device target for computing, which can be Ascend, GPU or CPU. (default=Ascend)')
 
     # Dataset parameters
     group = parser.add_argument_group('Dataset parameters')
@@ -94,7 +96,7 @@ def create_parser():
                             'Example: "randaug-m10-n2-w0-mstd0.5-mmax10-inc0", "autoaug-mstd0.5" or autoaugr-mstd0.5.')
     group.add_argument('--aug_splits', type=int, default=0,
                        help='Number of augmentation splits (default: 0, valid: 3 (currently, only support 3 splits))'
-                       'it should be set with one auto_augment')
+                            'it should be set with one auto_augment')
     group.add_argument('--re_prob', type=float, default=0.0,
                        help='Probability of performing erasing (default=0.0)')
     group.add_argument('--re_scale', type=tuple, default=(0.02, 0.33),
@@ -267,6 +269,8 @@ def create_parser():
                        help='Whether to shuffle the evaluation data (default=False)')
 
     return parser_config, parser
+
+
 # fmt: on
 
 

tests/tasks/test_train_val_imagenet_subset.py

@@ -30,6 +30,7 @@ def test_train(mode, val_while_train, model="resnet18"):
         DownLoad().download_and_extract_archive(dataset_url, root_dir)
 
     # ---------------- test running train.py using the toy data ---------
+    device_target = "CPU"
     dataset = "imagenet"
     num_classes = 2
     ckpt_dir = "./tests/ckpt_tmp"
@@ -48,7 +49,8 @@ def test_train(mode, val_while_train, model="resnet18"):
         f"python {train_file} --dataset={dataset} --num_classes={num_classes} --model={model} "
         f"--epoch_size={num_epochs} --ckpt_save_interval=2 --lr=0.0001 --num_samples={num_samples} --loss=CE "
         f"--weight_decay=1e-6 --ckpt_save_dir={ckpt_dir} {download_str} --train_split=train --batch_size={batch_size} "
-        f"--pretrained --num_parallel_workers=2 --val_while_train={val_while_train} --val_split=val --val_interval=1"
+        f"--pretrained --num_parallel_workers=2 --val_while_train={val_while_train} --val_split=val --val_interval=1 "
+        f"--device_target={device_target}"
     )
 
     print(f"Running command: \n{cmd}")
@@ -57,10 +59,11 @@ def test_train(mode, val_while_train, model="resnet18"):
 
     # --------- Test running validate.py using the trained model ------------- #
     # begin_ckpt = os.path.join(ckpt_dir, f'{model}-1_1.ckpt')
-    end_ckpt = os.path.join(ckpt_dir, f"{model}-{num_epochs}_{num_samples//batch_size}.ckpt")
+    end_ckpt = os.path.join(ckpt_dir, f"{model}-{num_epochs}_{num_samples // batch_size}.ckpt")
     cmd = (
         f"python validate.py --model={model} --dataset={dataset} --val_split=val --data_dir={data_dir} "
-        f"--num_classes={num_classes} --ckpt_path={end_ckpt} --batch_size=40 --num_parallel_workers=2"
+        f"--num_classes={num_classes} --ckpt_path={end_ckpt} --batch_size=40 --num_parallel_workers=2 "
+        f"--device_target={device_target}"
     )
     # ret = subprocess.call(cmd.split(), stdout=sys.stdout, stderr=sys.stderr)
     print(f"Running command: \n{cmd}")

validate.py

@@ -26,6 +26,7 @@ def check_batch_size(num_samples, ori_batch_size=32, refine=True):
 
 
 def validate(args):
+    ms.set_context(device_target=args.device_target)
     ms.set_context(mode=args.mode)
     if args.mode == ms.GRAPH_MODE:
         ms.set_context(jit_config={"jit_level": "O2"})