mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2025-08-22 22:02:51 +08:00
74 lines
2.8 KiB
Python
74 lines
2.8 KiB
Python
# Copyright 2024 the LlamaFactory team.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import os
|
|
from types import MethodType
|
|
from typing import TYPE_CHECKING, Dict, Optional
|
|
|
|
from transformers import Trainer
|
|
|
|
from ...extras.logging import get_logger
|
|
from ..trainer_utils import convert_pissa_adapter, create_custom_optimzer, create_custom_scheduler
|
|
|
|
|
|
if TYPE_CHECKING:
|
|
import torch
|
|
from transformers import ProcessorMixin
|
|
|
|
from ...hparams import FinetuningArguments
|
|
|
|
|
|
logger = get_logger(__name__)
|
|
|
|
|
|
class CustomTrainer(Trainer):
|
|
r"""
|
|
Inherits Trainer for custom optimizer.
|
|
"""
|
|
|
|
def __init__(
|
|
self, finetuning_args: "FinetuningArguments", processor: Optional["ProcessorMixin"], **kwargs
|
|
) -> None:
|
|
super().__init__(**kwargs)
|
|
self.finetuning_args = finetuning_args
|
|
self.processor = processor
|
|
|
|
if finetuning_args.pissa_convert:
|
|
self.save_model(os.path.join(self.args.output_dir, "pissa_init"))
|
|
|
|
if finetuning_args.use_badam:
|
|
from badam import clip_grad_norm_old_version, BAdamCallback
|
|
self.accelerator.clip_grad_norm_ = MethodType(clip_grad_norm_old_version, self.accelerator)
|
|
self.callback_handler.add_callback(BAdamCallback)
|
|
|
|
def create_optimizer(self) -> "torch.optim.Optimizer":
|
|
if self.optimizer is None:
|
|
self.optimizer = create_custom_optimzer(self.model, self.args, self.finetuning_args)
|
|
return super().create_optimizer()
|
|
|
|
def create_scheduler(
|
|
self, num_training_steps: int, optimizer: Optional["torch.optim.Optimizer"] = None
|
|
) -> "torch.optim.lr_scheduler.LRScheduler":
|
|
create_custom_scheduler(self.args, num_training_steps, optimizer)
|
|
return super().create_scheduler(num_training_steps, optimizer)
|
|
|
|
def _save(self, output_dir: Optional[str] = None, state_dict: Optional[Dict[str, "torch.Tensor"]] = None) -> None:
|
|
super()._save(output_dir, state_dict)
|
|
output_dir = output_dir if output_dir is not None else self.args.output_dir
|
|
if self.finetuning_args.pissa_convert:
|
|
convert_pissa_adapter(output_dir, state_dict, self.accelerator, self.model, self.args)
|
|
|
|
if self.processor is not None:
|
|
getattr(self.processor, "image_processor").save_pretrained(output_dir)
|