support full-parameter PPO

This commit is contained in:
hiyouga
2023-11-16 02:08:04 +08:00
parent 8350bcf85d
commit ce78303600
20 changed files with 288 additions and 145 deletions

View File

@@ -1,3 +1,3 @@
from llmtuner.model.loader import load_model_and_tokenizer
from llmtuner.model.parser import get_train_args, get_infer_args, get_eval_args
from llmtuner.model.utils import dispatch_model, generate_model_card
from llmtuner.model.utils import create_ref_model, create_reward_model, dispatch_model, generate_model_card