Modify requirements.txt

finetune/requirements.txt

 jieba>=0.42.1
 datasets>=2.20.0
 peft>=0.15.1
-deepspeed>=0.16.5
 nltk==3.8.1
 rouge_chinese==1.0.3
 ruamel.yaml>=0.18.6

inference/requirements.txt

@@ -12,7 +12,6 @@ openai>=1.70.0
 einops>=0.8.0
 pillow>=10.4.0
 sse-starlette>=2.1.3
-bitsandbytes>=0.44.1 # INT4 Loading, Not support for NPU
 peft>=0.15.0 # Using with finetune model
 swanlab>=0.5.5