ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update rl.py

Browse source
This commit is contained in:
Daniel Han 2025-02-05 07:13:40 -08:00
commit 1022dcfa6e
1 changed files with 7 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

10
unsloth/models/rl.py
View file

@ -33,16 +33,16 @@ from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, Union
def PatchRL(FastLanguageModel):
from trl.models.utils import unwrap_model_for_generation
from trl.models import unwrap_model_for_generation
from contextlib import contextmanager
@contextmanager
def unsloth_unwrap_model_for_generation(model, accelerator):
# Must use for_inference to allow inference in Unsloth
FastLanguageModel.for_inference(model)
with unwrap_model_for_generation(model, accelerator) as unwrapped_model:
FastLanguageModel.for_inference(unwrapped_model)
try:
yield unwrapped_model.eval()
yield unwrapped_model
finally:
# Finally return back training
FastLanguageModel.for_training(model)
@ -50,6 +50,10 @@ def PatchRL(FastLanguageModel):
pass
pass
import trl.models
trl.models.utils.unwrap_model_for_generation = unwrap_model_for_generation
trl.models.unwrap_model_for_generation = unwrap_model_for_generation
import trl.trainer
trainers = dir(trl.trainer)
trainers = [x for x in trainers if x.endswith("_trainer")]