GXKIM

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior Defaulting to user installation because normal site-packages is not writeable Processing...

pip error Linux centos 7

1

chat_env) [root@192 LMFlow]# pip install -r requirements.txt Collecting peft@ git+https://github.com/huggingface/peft.git@deff03f2c251534fffd2511fc2d440e84cc54b1b Cloning https://github.com/huggingface/peft.git (to revision deff03f2c251534fffd2511fc2d440e84cc54b1b) to /tmp/pip-install-qp3u6wif/peft_bb110dd6776941069294566614e126c0 Running command git clone --quiet https://github.com/huggingface/peft.git /tmp/pip-install-qp3u6wif/peft_bb110dd6776941069294566614e126c0 fatal: unable to access 'https://github.com/huggingface/peft.git/': TCP...

运行finetune报错

2

NCCL_DEBUG=info NCCL_IB_DISABLE=0 NCCL_NET_GDR_LEVEL=2 deepspeed --master_port 16666 --hostfile hostfile_single finetune_visualglm.py --experiment-name finetune-visualglm-6b --model-parallel-size 1 --mode finetune --train-iters 300 --resume-dataloader --max_source_length 64 --max_target_length 256 --lora_rank 10 --pre_seq_len 4 --train-data ./fewshot-data/dataset.json --valid-data ./fewshot-data/dataset.json...

torch run的问题

4

问题：torchrun找不到 是因为torch cuda版本问题嘛？大佬指教

Qwen1.5 合并

1

如题 代码： ```python from peft import AutoPeftModelForCausalLM from transformers import AutoTokenizer import argparse parser=argparse.ArgumentParser() parser.add_argument("--path_to_adapter",help="path to adapter") parser.add_argument("--merge_path",help="path to merge") args=parser.parse_args() #new_model_directory='merge_qwen' #path_to_adapter="output_qwen" model = AutoPeftModelForCausalLM.from_pretrained( args.path_to_adapter, device_map="auto", trust_remote_code=True ).eval()...

关于模型训练时间和显存资源

6

如题 请问为什么qwen1.5 无论是训练时间还是显存资源比起qwen1 更消耗资源呢，我知道模型结构上发生了一些变化 显卡:A800 model: qwen1.5 -14B-chat 单卡lora 62G 显存 训练时间: qwen1-14B-chat的三倍 虚心请教

加载自定义模型 qwen1.5 0.5B chat

4

### 如上述文字 自定义模型0.5B 是json文件这么写不对吗？请指教，json无法解析  

显卡配置v100 32g单卡

FEAT: support deepseek v2 chat and coder

3

如题