def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config = load_config(args.config_path)
run_train_model(config)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config = load_config(args.config_path)
run_grid_search(config)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config = load_config(args.config_path)
build_inference_data(config)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config: ConfigBase = load_config(args.config_path)
config.model.predict()
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config: ConfigBase = load_config(args.config_path)
config.data_builder.build_training_data()
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
logger.info("load config")
config: ConfigBase = load_config(args.config_path)
logger.info("start beautify")
beautifier = config.data_beautifier
beautifier.beautify()
def main():
parser = argparse.ArgumentParser()
parser.add_argument("config_path")
args = parser.parse_args()
config = load_config(args.config_path)
train_data_loader = DataLoader(config.train_dataset,
batch_size=config.batch_size,
shuffle=True,
pin_memory=True,
num_workers=config.num_workers)
val_data_loader = DataLoader(config.val_dataset,
batch_size=config.batch_size,
shuffle=False,
num_workers=config.num_workers)
model = config.model
model_save_path = config.model_save_path
os.makedirs(model_save_path, exist_ok=True)
logger_path = os.path.join(model_save_path, "log.txt")
setup_logger(out_file=logger_path)
trainer = config.trainer_cls(model=model,
train_data_loader=train_data_loader,
val_data_loader=val_data_loader,
epoch_count=config.epoch_count,
optimizer=config.optimizer,
scheduler=config.scheduler,
loss_calculator=config.loss_calculator,
metric_calculator=config.metric_calculator,
print_frequency=config.print_frequency)
trainer.run()