大模型部署_书生浦语大模型

大模型部署_书生浦语大模型 _作业2

本节课可以让同学们实践 4 个主要内容，分别是：

1、部署 `InternLM2-Chat-1.8B` 模型进行智能对话

1.1安装依赖库：

pip install huggingface-hub==0.17.3
pip install transformers==4.34 
pip install psutil==5.9.8
pip install accelerate==0.24.1
pip install streamlit==1.32.2 
pip install matplotlib==3.8.3 
pip install modelscope==1.9.5
pip install sentencepiece==0.1.99

1.2下载 `InternLM2-Chat-1.8B` 模型

import os
from modelscope.hub.snapshot_download import snapshot_download# 创建保存模型目录
os.system("mkdir /root/models")# save_dir是模型保存到本地的目录
save_dir="/root/models"snapshot_download("Shanghai_AI_Laboratory/internlm2-chat-1_8b", cache_dir=save_dir, revision='v1.1.0')

1.3运行 cli_demo

import torch
from transformers import AutoTokenizer, AutoModelForCausalLMmodel_name_or_path = "/root/models/Shanghai_AI_Laboratory/internlm2-chat-1_8b"tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True, device_map='cuda:0')
model = AutoModelForCausalLM.from_pretrained(model_name_or_path, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map='cuda:0')
model = model.eval()system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文.
"""messages = [(system_prompt, '')]print("=============Welcome to InternLM chatbot, type 'exit' to exit.=============")while True:input_text = input("\nUser  >>> ")input_text = input_text.replace(' ', '')if input_text == "exit":breaklength = 0for response, _ in model.stream_chat(tokenizer, input_text, messages):if response is not None:print(response[length:], flush=True, end="")length = len(response)

2、部署实战营优秀作品 `八戒-Chat-1.8B` 模型

八戒-Chat-1.8B：魔搭社区
聊天-嬛嬛-1.8B：OpenXLab浦源 - 模型中心
Mini-Horo-巧耳：OpenXLab浦源 - 模型中心

git clone https://gitee.com/InternLM/Tutorial -b camp2

执行下载模型：

python /root/Tutorial/helloworld/bajie_download.py

待程序下载完成后，输入运行命令：

streamlit run /root/Tutorial/helloworld/bajie_chat.py --server.address 127.0.0.1 --server.port 6006

3、通过 `InternLM2-Chat-7B` 运行 `Lagent` 智能体 `Demo`

Lagent 的特性总结如下：

流式输出：提供 stream_chat 接口作流式输出，本地就能演示酷炫的流式 Demo。
接口统一，设计全面升级，提升拓展性，包括：
- 型号：不论是 OpenAI API， Transformers 还是推理加速框架 LMDeploy 一网打尽，模型切换可以游刃有余;
- Action：简单的继承和装饰，即可打造自己个人的工具集，不论 InternLM 还是 GPT 均可适配;
- Agent：与 Model 的输入接口保持一致，模型到智能体的蜕变只需一步，便捷各种 agent 的探索实现;
文档全面升级，API 文档全覆盖。

下载模型:

git clone https://gitee.com/internlm/lagent.git
# git clone https://github.com/internlm/lagent.git
cd /root/demo/lagent
git checkout 581d9fb8987a5d9b72bb9ebd37a95efd47d479ac
pip install -e . # 源码安装

在 terminal 中输入指令，构造软链接快捷访问方式：

ln -s /root/share/new_models/Shanghai_AI_Laboratory/internlm2-chat-7b /root/models/internlm2-chat-7b

打开路径下文件，并修改对应位置 (71行左右) 代码 :internlm2_agent_web_demo_hf.py

修改模型地址：

运行前端代码：

streamlit run /root/demo/lagent/examples/internlm2_agent_web_demo_hf.py --server.address 127.0.0.1 --server.port 6006

4、实践部署 `浦语·灵笔2` 模型

补充环境包，选用 进行开发：50% A100

pip install timm==0.4.12 sentencepiece==0.1.99 markdown2==2.4.10 xlsxwriter==3.1.2 gradio==4.13.0 modelscope==1.9.5

下载 InternLM-XComposer 仓库 相关的代码资源：

cd /root/demo
git clone https://gitee.com/internlm/InternLM-XComposer.git
# git clone https://github.com/internlm/InternLM-XComposer.git
cd /root/demo/InternLM-XComposer
git checkout f31220eddca2cf6246ee2ddf8e375a40457ff626

在中输入指令，构造软链接快捷访问方式：terminal

ln -s /root/share/new_models/Shanghai_AI_Laboratory/internlm-xcomposer2-7b /group_share/01/models/internlm-xcomposer2-7b
ln -s /root/share/new_models/Shanghai_AI_Laboratory/internlm-xcomposer2-vl-7b /group_share/01/models/internlm-xcomposer2-vl-7b

图文写作实战（开启 50% A100 权限后才可开启此章节）

cd /root/demo/InternLM-XComposer
python /root/demo/InternLM-XComposer/examples/gradio_demo_composition.py  \
--code_path /root/models/internlm-xcomposer2-7b \
--private \
--num_gpus 1 \
--port 6006

图片理解实战（开启 50% A100 权限后才可开启此章节）

根据附录 6.4 的方法，关闭并重新启动一个新的，继续输入指令，启动：terminalInternLM-XComposer2-vl

conda activate democd /root/demo/InternLM-XComposer
python gradio_demo_chat.py  \
--code_path /group_share/01/models/internlm-xcomposer2-vl-7b \
--private \
--num_gpus 1 \
--port 6006