fix #4209

DeepSpeed ZeRO3 has inflight param error when calling model.eval()

src/llamafactory/train/dpo/trainer.py

@@ -1,3 +1,4 @@
+import warnings
 from collections import defaultdict
 from contextlib import nullcontext
 from types import MethodType
@@ -10,7 +11,7 @@ from trl import DPOTrainer
 from trl.trainer import disable_dropout_in_model
 
 from ...extras.constants import IGNORE_INDEX
-from ..trainer_utils import create_custom_optimzer, create_custom_scheduler, get_batch_logps, get_ref_context
+from ..trainer_utils import create_custom_optimzer, create_custom_scheduler, get_batch_logps
 
 
 if TYPE_CHECKING:
@@ -61,6 +62,8 @@ class CustomDPOTrainer(DPOTrainer):
         if not hasattr(self, "accelerator"):
             raise AttributeError("Please update `transformers`.")
 
+        warnings.simplefilter("ignore")  # remove gc warnings on ref model
+
         if ref_model is not None:
             if self.is_deepspeed_enabled:
                 if not (
@@ -176,7 +179,7 @@ class CustomDPOTrainer(DPOTrainer):
 
         if self.ref_model is None:
             ref_model = model
-            ref_context = get_ref_context(self.accelerator, model)
+            ref_context = self.accelerator.unwrap_model(model).disable_adapter()
         else:
             ref_model = self.ref_model
             ref_context = nullcontext()

src/llamafactory/train/kto/trainer.py

@@ -1,3 +1,4 @@
+import warnings
 from collections import defaultdict
 from contextlib import nullcontext
 from types import MethodType
@@ -9,7 +10,7 @@ from trl import KTOTrainer
 from trl.trainer import disable_dropout_in_model
 
 from ...extras.constants import IGNORE_INDEX
-from ..trainer_utils import create_custom_optimzer, create_custom_scheduler, get_batch_logps, get_ref_context
+from ..trainer_utils import create_custom_optimzer, create_custom_scheduler, get_batch_logps
 
 
 if TYPE_CHECKING:
@@ -60,6 +61,8 @@ class CustomKTOTrainer(KTOTrainer):
         if not hasattr(self, "accelerator"):
             raise AttributeError("Please update `transformers`.")
 
+        warnings.simplefilter("ignore")  # remove gc warnings on ref model
+
         if ref_model is not None:
             if self.is_deepspeed_enabled:
                 if not (
@@ -143,7 +146,7 @@ class CustomKTOTrainer(KTOTrainer):
         """
         if self.ref_model is None:
             ref_model = model
-            ref_context = get_ref_context(self.accelerator, model)
+            ref_context = self.accelerator.unwrap_model(model).disable_adapter()
         else:
             ref_model = self.ref_model
             ref_context = nullcontext()

src/llamafactory/train/ppo/trainer.py

@@ -1,6 +1,7 @@
 import math
 import os
 import sys
+import warnings
 from types import MethodType
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
 
@@ -136,6 +137,7 @@ class CustomPPOTrainer(PPOTrainer, Trainer):
 
         device_type = unwrapped_model.pretrained_model.device.type
         self.amp_context = torch.autocast(device_type, dtype=model_args.compute_dtype)
+        warnings.simplefilter("ignore")  # remove gc warnings on ref model
 
         if finetuning_args.reward_model_type == "full":
             if self.is_deepspeed_enabled:

src/llamafactory/train/trainer_utils.py

@@ -1,4 +1,3 @@
-from contextlib import contextmanager
 from typing import TYPE_CHECKING, Callable, Dict, List, Optional, Tuple, Union
 
 import torch
@@ -19,7 +18,6 @@ if is_galore_available():
 
 
 if TYPE_CHECKING:
-    from accelerate import Accelerator
     from transformers import PreTrainedModel, Seq2SeqTrainingArguments
     from trl import AutoModelForCausalLMWithValueHead
 
@@ -154,17 +152,6 @@ def create_reward_model(
         return reward_model
 
 
-@contextmanager
-def get_ref_context(accelerator: "Accelerator", model: "PreTrainedModel"):
-    r"""
-    Gets adapter context for the reference model.
-    """
-    with accelerator.unwrap_model(model).disable_adapter():
-        model.eval()
-        yield
-        model.train()
-
-
 def _get_decay_parameter_names(model: "PreTrainedModel") -> List[str]:
     r"""
     Returns a list of names of parameters with weight decay. (weights in non-layernorm layers)