mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2025-10-14 23:58:11 +08:00
76 lines
3.0 KiB
Python
76 lines
3.0 KiB
Python
# Inspired by: https://github.com/CarperAI/trlx/blob/main/examples/summarize_rlhf/reward_model/train_reward_model_gptj.py
|
|
|
|
from typing import TYPE_CHECKING, List, Optional
|
|
|
|
from ...data import PairwiseDataCollatorWithPadding, get_dataset, split_dataset
|
|
from ...extras.callbacks import FixValueHeadModelCallback
|
|
from ...extras.misc import fix_valuehead_checkpoint
|
|
from ...extras.ploting import plot_loss
|
|
from ...model import load_model, load_tokenizer
|
|
from ..utils import create_modelcard_and_push
|
|
from .metric import compute_accuracy
|
|
from .trainer import PairwiseTrainer
|
|
|
|
|
|
if TYPE_CHECKING:
|
|
from transformers import Seq2SeqTrainingArguments, TrainerCallback
|
|
|
|
from ...hparams import DataArguments, FinetuningArguments, ModelArguments
|
|
|
|
|
|
def run_rm(
|
|
model_args: "ModelArguments",
|
|
data_args: "DataArguments",
|
|
training_args: "Seq2SeqTrainingArguments",
|
|
finetuning_args: "FinetuningArguments",
|
|
callbacks: Optional[List["TrainerCallback"]] = None,
|
|
):
|
|
tokenizer_module = load_tokenizer(model_args)
|
|
tokenizer = tokenizer_module["tokenizer"]
|
|
dataset = get_dataset(model_args, data_args, training_args, stage="rm", **tokenizer_module)
|
|
model = load_model(tokenizer, model_args, finetuning_args, training_args.do_train, add_valuehead=True)
|
|
data_collator = PairwiseDataCollatorWithPadding(tokenizer, pad_to_multiple_of=8)
|
|
|
|
# Update arguments
|
|
training_args.remove_unused_columns = False # important for pairwise dataset
|
|
|
|
# Initialize our Trainer
|
|
trainer = PairwiseTrainer(
|
|
model=model,
|
|
args=training_args,
|
|
finetuning_args=finetuning_args,
|
|
data_collator=data_collator,
|
|
callbacks=callbacks + [FixValueHeadModelCallback()],
|
|
compute_metrics=compute_accuracy,
|
|
**tokenizer_module,
|
|
**split_dataset(dataset, data_args, training_args),
|
|
)
|
|
|
|
# Training
|
|
if training_args.do_train:
|
|
train_result = trainer.train(resume_from_checkpoint=training_args.resume_from_checkpoint)
|
|
trainer.save_model()
|
|
if training_args.should_save:
|
|
fix_valuehead_checkpoint(model, training_args.output_dir, training_args.save_safetensors)
|
|
trainer.log_metrics("train", train_result.metrics)
|
|
trainer.save_metrics("train", train_result.metrics)
|
|
trainer.save_state()
|
|
if trainer.is_world_process_zero() and finetuning_args.plot_loss:
|
|
plot_loss(training_args.output_dir, keys=["loss", "eval_loss", "eval_accuracy"])
|
|
|
|
# Evaluation
|
|
if training_args.do_eval:
|
|
metrics = trainer.evaluate(metric_key_prefix="eval")
|
|
trainer.log_metrics("eval", metrics)
|
|
trainer.save_metrics("eval", metrics)
|
|
|
|
# Predict
|
|
if training_args.do_predict:
|
|
predict_results = trainer.predict(dataset, metric_key_prefix="predict")
|
|
trainer.log_metrics("predict", predict_results.metrics)
|
|
trainer.save_metrics("predict", predict_results.metrics)
|
|
trainer.save_predictions(predict_results)
|
|
|
|
# Create model card
|
|
create_modelcard_and_push(trainer, model_args, data_args, training_args, finetuning_args)
|