-
Notifications
You must be signed in to change notification settings - Fork 5
/
run.py
42 lines (34 loc) · 1.54 KB
/
run.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
import os
import shutil
import argparse
from src.engine import Engine
from src.utils import create_logger
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--config_path', type=str, required=True, help="Path to the YAML config file")
parser.add_argument('--save_dir', default='./logs/run-temp', help='Directory to save config and model checkpoint')
group = parser.add_mutually_exclusive_group()
group.add_argument('--test', action='store_true', default=False,
help='Only run testing - ensure the checkpoint path is provided in the config file')
group.add_argument('--pretrain', action='store_true', default=False,
help='Run pretraining - ensure the correct config is provided through the config_path argument')
args = parser.parse_args()
# Extract arguments
save_dir = args.save_dir
config_path = args.config_path
test = args.test
pretrain = args.pretrain
# Create the save directory
os.makedirs(save_dir, exist_ok=True)
# Copy the provided config file into save_dir
shutil.copyfile(config_path, os.path.join(save_dir, "config.yaml"))
# Create the logger
logger = create_logger(name=save_dir)
# Create the engine (this will create an Engine object that contains models, optimizers, loss functions, etc.)
engine = Engine(config_path=args.config_path, logger=logger, save_dir=args.save_dir)
if test:
engine.evaluate()
elif pretrain:
engine.pretrain()
else:
engine.train()