diff --git a/training/scripts/train.py b/training/scripts/train.py index fed7e20..627d90c 100755 --- a/training/scripts/train.py +++ b/training/scripts/train.py @@ -86,7 +86,7 @@ def train(config_path): learning_rate=config["train_params"]["learning_rate"], lr_scheduler_type=config["train_params"]["lr_scheduler_type"], weight_decay=config["train_params"]["weight_decay"], - warmup_ratio=config["train_params"]["warmup_ratio"], + warmup_steps=int(config["train_params"]["warmup_ratio"] * config["train_params"]["num_train_epochs"] * len(dataset["train"]) // config["train_params"]["per_device_train_batch_size"]), logging_steps=config["train_params"]["logging_steps"], save_steps=config["train_params"]["save_steps"], save_total_limit=config["train_params"]["save_total_limit"],