423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-08-04 12:42:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix #2081

Browse Source 
Former-commit-id: 33f2c0d4f89cf76671c0fdfbcee79d732b6a020e
This commit is contained in:
hiyouga 2024-01-04 23:19:08 +08:00
parent dabd40750c
commit dc8714a003
2 changed files with 6 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/llmtuner/model/adapter.py
View File

@ -1,5 +1,6 @@
import torch import torch
from typing import TYPE_CHECKING from typing import TYPE_CHECKING
from transformers.integrations import is_deepspeed_zero3_enabled
from peft import PeftModel, TaskType, LoraConfig, get_peft_model from peft import PeftModel, TaskType, LoraConfig, get_peft_model
from llmtuner.extras.logging import get_logger from llmtuner.extras.logging import get_logger
@ -71,6 +72,10 @@ def init_adapter(
assert len(model_args.adapter_name_or_path) == 1, "Quantized model only accepts a single adapter." assert len(model_args.adapter_name_or_path) == 1, "Quantized model only accepts a single adapter."
is_mergeable = False is_mergeable = False
if is_deepspeed_zero3_enabled():
assert len(model_args.adapter_name_or_path) == 1, "Cannot use multiple adapters in DeepSpeed ZeRO-3."
is_mergeable = False
if (is_trainable and not finetuning_args.create_new_adapter) or (not is_mergeable): if (is_trainable and not finetuning_args.create_new_adapter) or (not is_mergeable):
adapter_to_merge = model_args.adapter_name_or_path[:-1] adapter_to_merge = model_args.adapter_name_or_path[:-1]
adapter_to_resume = model_args.adapter_name_or_path[-1] adapter_to_resume = model_args.adapter_name_or_path[-1]

2
src/llmtuner/model/patcher.py
View File

@ -3,7 +3,7 @@ import math
import torch import torch
import random import random
from types import MethodType from types import MethodType
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Set, Tuple from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
from datasets import load_dataset from datasets import load_dataset
from transformers import BitsAndBytesConfig, GPTQConfig, PreTrainedModel, PreTrainedTokenizerBase from transformers import BitsAndBytesConfig, GPTQConfig, PreTrainedModel, PreTrainedTokenizerBase