#通过intel发布的bigDL,可以用cpu运行chatglm2
, #操作步骤:
#1.conda create -n inthj python==3.9
#2.conda activate inthj
#3.cd inthj
#4.pip install --pre --upgrade bigdl-llm[all]
#5.进入chatglm2的cli_demo.py文件
#6.把from transformers import AutoTokenizer, AutoModel
#改成from bigdl.llm.transformers import AutoModel
#from transformers import AutoTokenizer
#7.把model = AutoModel.from_pretrained("你的路径",trust_remote_code=True).float()
#改成model = AutoModel.from_pretrained("你的路径",load_in_4bit=True,trust_remote_code=True).float()
#8.然后你就看见你的大模型在CPU下飞快的运转了
源代码
import os
import platform
import signal
#from transformers import AutoTokenizer, AutoModel
#import readline
import torch
import time
import argparse
import numpy as np
from bigdl.llm.transformers import AutoModel
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("D:\\c-glm\\ChatGLM2-6B", trust_remote_code=True)
model = AutoModel.from_pretrained("D:\\c-glm\\ChatGLM2-6B", load_in_4bit=True,trust_remote_code=True)#.float()#.cuda()
# 多显卡支持,使用下面两行代替上面一行,将num_gpus改为你实际的显卡数量
# from utils import load_model_on_gpus
# model = load_model_on_gpus("THUDM/chatglm2-6b", num_gpus=2)
model = model.eval()
os_name = platform.system()
clear_command = 'cls' if os_name == 'Windows' else 'clear'
stop_stream = False
def build_prompt(history):
prompt = "欢迎使用 ChatGLM2-6B 模型,输入内容即可进行对话,clear 清空对话历史,stop 终止程序"
for query, response in history:
prompt += f"\n\n用户:{query}"
prompt += f"\n\nChatGLM2-6B:{response}"
return prompt
def signal_handler(signal, frame):
global stop_stream
stop_stream = True
def main():
past_key_values, history = None, []
global stop_stream
print("欢迎使用 ChatGLM2-6B 模型,输入内容即可进行对话,clear 清空对话历史,stop 终止程序")
while True:
query = input("\n用户:")
if query.strip() == "stop":
break
if query.strip() == "clear":
past_key_values, history = None, []
os.system(clear_command)
print("欢迎使用 ChatGLM2-6B 模型,输入内容即可进行对话,clear 清空对话历史,stop 终止程序")
continue
print("\nChatGLM:", end="")
current_length = 0
for response, history, past_key_values in model.stream_chat(tokenizer, query, history=history,
past_key_values=past_key_values,
return_past_key_values=True):
if stop_stream:
stop_stream = False
break
else:
print(response[current_length:], end="", flush=True)
current_length = len(response)
print("")
if __name__ == "__main__":
main()