feat: add cli config for size
Browse files
train.py
CHANGED
|
@@ -81,7 +81,7 @@ parser.add_argument(
|
|
| 81 |
"-n",
|
| 82 |
"--model-name",
|
| 83 |
type=str,
|
| 84 |
-
default=
|
| 85 |
help="Model name (default: current tag)",
|
| 86 |
)
|
| 87 |
parser.add_argument(
|
|
@@ -90,6 +90,13 @@ parser.add_argument(
|
|
| 90 |
action="store_true",
|
| 91 |
help="Font classification only (default: False)",
|
| 92 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
|
| 94 |
args = parser.parse_args()
|
| 95 |
|
|
@@ -99,6 +106,8 @@ single_batch_size = args.single_batch_size
|
|
| 99 |
total_num_workers = os.cpu_count()
|
| 100 |
single_device_num_workers = total_num_workers // len(devices)
|
| 101 |
|
|
|
|
|
|
|
| 102 |
if os.name == "nt":
|
| 103 |
single_device_num_workers = 0
|
| 104 |
|
|
@@ -137,7 +146,7 @@ data_module = FontDataModule(
|
|
| 137 |
num_iters = data_module.get_train_num_iter(num_device) * num_epochs
|
| 138 |
num_warmup_iter = data_module.get_train_num_iter(num_device) * num_warmup_epochs
|
| 139 |
|
| 140 |
-
model_name = args.model_name
|
| 141 |
|
| 142 |
logger_unconditioned = TensorBoardLogger(
|
| 143 |
save_dir=os.getcwd(), name="tensorboard", version=model_name
|
|
|
|
| 81 |
"-n",
|
| 82 |
"--model-name",
|
| 83 |
type=str,
|
| 84 |
+
default=None,
|
| 85 |
help="Model name (default: current tag)",
|
| 86 |
)
|
| 87 |
parser.add_argument(
|
|
|
|
| 90 |
action="store_true",
|
| 91 |
help="Font classification only (default: False)",
|
| 92 |
)
|
| 93 |
+
parser.add_argument(
|
| 94 |
+
"-z",
|
| 95 |
+
"--size",
|
| 96 |
+
type=int,
|
| 97 |
+
default=512,
|
| 98 |
+
help="Model feature image input size (default: 512)",
|
| 99 |
+
)
|
| 100 |
|
| 101 |
args = parser.parse_args()
|
| 102 |
|
|
|
|
| 106 |
total_num_workers = os.cpu_count()
|
| 107 |
single_device_num_workers = total_num_workers // len(devices)
|
| 108 |
|
| 109 |
+
config.INPUT_SIZE = args.size
|
| 110 |
+
|
| 111 |
if os.name == "nt":
|
| 112 |
single_device_num_workers = 0
|
| 113 |
|
|
|
|
| 146 |
num_iters = data_module.get_train_num_iter(num_device) * num_epochs
|
| 147 |
num_warmup_iter = data_module.get_train_num_iter(num_device) * num_warmup_epochs
|
| 148 |
|
| 149 |
+
model_name = get_current_tag() if args.model_name is None else args.model_name
|
| 150 |
|
| 151 |
logger_unconditioned = TensorBoardLogger(
|
| 152 |
save_dir=os.getcwd(), name="tensorboard", version=model_name
|