423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-17 20:30:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

release v0.1.0

Browse Source
This commit is contained in:
hiyouga
2023-07-18 00:18:25 +08:00
parent 85c2210452
commit f8193e8009
30 changed files with 1513 additions and 309 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/llmtuner/tuner/rm/workflow.py
View File

@@ -2,7 +2,8 @@
# https://github.com/lvwerra/trl/blob/main/examples/summarization/scripts/reward_summarization.py
# https://github.com/CarperAI/trlx/blob/main/examples/summarize_rlhf/reward_model/train_reward_model_gptj.py
from transformers import Seq2SeqTrainingArguments
from typing import Optional, List
from transformers import Seq2SeqTrainingArguments, TrainerCallback
from llmtuner.dsets import get_dataset, preprocess_dataset
from llmtuner.extras.callbacks import LogCallback
@@ -18,7 +19,8 @@ def run_rm(
model_args: ModelArguments,
data_args: DataArguments,
training_args: Seq2SeqTrainingArguments,
finetuning_args: FinetuningArguments
finetuning_args: FinetuningArguments,
callbacks: Optional[List[TrainerCallback]] = [LogCallback()]
):
dataset = get_dataset(model_args, data_args)
model, tokenizer = load_model_and_tokenizer(model_args, finetuning_args, training_args.do_train, stage="rm")
@@ -44,7 +46,7 @@ def run_rm(
args=training_args,
tokenizer=tokenizer,
data_collator=data_collator,
callbacks=[LogCallback()],
callbacks=callbacks,
compute_metrics=compute_accuracy,
**trainer_kwargs
)