README.md 9.52 KB
Newer Older
zhuwenwen's avatar
zhuwenwen committed
1
2
3
4
<!--
 * @Author: zhuww
 * @email: zhuww@sugon.com
 * @Date: 2024-06-13 14:38:07
zhuwenwen's avatar
zhuwenwen committed
5
 * @LastEditTime: 2024-09-30 09:16:01
zhuwenwen's avatar
zhuwenwen committed
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
-->
## 论文

`GLM: General Language Model Pretraining with Autoregressive Blank Infilling`

- [https://arxiv.org/abs/2103.10360](https://arxiv.org/abs/2103.10360)

## 模型结构

ChatGLM-6B 是清华大学开源的开源的、支持中英双语的对话语言模型,基于 [General Language Model (GLM)](https://github.com/THUDM/GLM) 架构,具有 62 亿参数。ChatGLM-6B 使用了和 ChatGPT 相似的技术,针对中文问答和对话进行了优化。经过约 1T 标识符的中英双语训练,辅以监督微调、反馈自助、人类反馈强化学习等技术的加持,62 亿参数的 ChatGLM-6B 已经能生成相当符合人类偏好的回答。ChatGLM2-6B 是开源中英双语对话模型 ChatGLM-6B 的第二代版本,ChatGLM3 是智谱AI和清华大学 KEG 实验室联合发布的新一代对话预训练模型。ChatGLM3-6B 是 ChatGLM3 系列中的开源模型,在保留了前两代模型对话流畅、部署门槛低等众多优秀特性的基础上,ChatGLM3-6B 具有更强大的基础模型、更完整的功能支持、更全面的开源序列。

<div align="center">
<img src="docs/transformers.jpg" width="300" height="400">
</div>

以下是ChatGLM系列模型的主要网络参数配置:

| 模型名称    | 隐含层维度 | 层数 | 头数 | 词表大小 | 位置编码 | 最大序列长度 |
| ----------- | ---------- | ---- | ---- | -------- | -------- | ------------ |
| ChatGLM2-6B | 4096       | 28   | 32   | 65024    | RoPE     | 8192         |
| ChatGLM3-6B | 4096       | 28   | 32   | 65024    | RoPE     | 8192         |
| glm-4-9b | 4096       | 40   | 32   | 151552    | RoPE     | 131072         |

## 算法原理

ChatGLM系列模型基于GLM架构开发。GLM是一种基于Transformer的语言模型,以自回归空白填充为训练目标, 同时具备自回归和自编码能力。

<div align="center">
<img src="docs/GLM.png" width="550" height="200">
</div>


## 环境配置

### Docker(方法一)
提供[光源](https://www.sourcefind.cn/#/image/dcu/custom)拉取推理的docker镜像:

```
zhuwenwen's avatar
zhuwenwen committed
44
docker pull image.sourcefind.cn:5000/dcu/admin/base/pytorch:2.1.0-ubuntu20.04-dtk24.04.2-py3.10
zhuwenwen's avatar
zhuwenwen committed
45
46
47
48
49
# <Image ID>用上面拉取docker镜像的ID替换
# <Host Path>主机端路径
# <Container Path>容器映射路径
docker run -it --name chatglm_vllm --privileged --shm-size=64G  --device=/dev/kfd --device=/dev/dri/ --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v /opt/hyhal:/opt/hyhal -v <Host Path>:<Container Path> <Image ID> /bin/bash
```
zhuwenwen's avatar
zhuwenwen committed
50
`Tips:若在K100/Z100L上使用,使用定制镜像docker pull image.sourcefind.cn:5000/dcu/admin/base/custom:vllm0.5.0-dtk24.04.1-ubuntu20.04-py310-zk-v1,K100/Z100L不支持awq量化`
zhuwenwen's avatar
zhuwenwen committed
51
52
53
54
55
56

### Dockerfile(方法二)
```
# <Host Path>主机端路径
# <Container Path>容器映射路径
docker build -t chatglm:latest .
zhuwenwen's avatar
zhuwenwen committed
57
docker run -it --name chatglm_vllm --privileged --shm-size=64G  --device=/dev/kfd --device=/dev/dri/ --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v /opt/hyhal:/opt/hyhal:ro -v <Host Path>:<Container Path> llama:latest /bin/bash
zhuwenwen's avatar
zhuwenwen committed
58
59
60
61
62
63
64
```

### Anaconda(方法三)
```
conda create -n chatglm_vllm python=3.10
```
关于本项目DCU显卡所需的特殊深度学习库可从[光合](https://developer.hpccube.com/tool/)开发者社区下载安装。
zhuwenwen's avatar
zhuwenwen committed
65
* DTK驱动:dtk24.04.2
zhuwenwen's avatar
zhuwenwen committed
66
67
* Pytorch: 2.1.0
* triton:2.1.0
zhuwenwen's avatar
zhuwenwen committed
68
* lmslim: 0.1.0
zhuwenwen's avatar
zhuwenwen committed
69
70
* xformers: 0.0.25
* flash_attn: 2.0.4
zhuwenwen's avatar
zhuwenwen committed
71
* vllm: 0.5.0
zhuwenwen's avatar
zhuwenwen committed
72
73
* python: python3.10

zhuwenwen's avatar
zhuwenwen committed
74
`Tips:需先安装相关依赖,最后安装vllm包`
zhuwenwen's avatar
zhuwenwen committed
75
76
77
78
79
80
81
82

## 数据集


## 推理

### 模型下载

zhuwenwen's avatar
zhuwenwen committed
83
84
| 基座模型 | 长文本模型 |
| ------- | ------- |
zhuwenwen's avatar
zhuwenwen committed
85
86
| [chatglm2-6b](http://113.200.138.88:18080/aimodels/chatglm2-6b) | [chatglm2-6b-32k](https://huggingface.co/THUDM/chatglm2-6b-32k) | 
| [chatglm3-6b](http://113.200.138.88:18080/aimodels/chatglm3-6b)  | [chatglm3-6b-32k](http://113.200.138.88:18080/aimodels/chatglm3-6b-32k) | 
zhuwenwen's avatar
zhuwenwen committed
87
88
89
90
| [glm-4-9b-chat](https://huggingface.co/THUDM/glm-4-9b-chat) | 

### 离线批量推理
```bash
zhuwenwen's avatar
zhuwenwen committed
91
python examples/offline_inference.py
zhuwenwen's avatar
zhuwenwen committed
92
93
94
95
96
97
98
99
```
其中,`prompts`为提示词;`temperature`为控制采样随机性的值,值越小模型生成越确定,值变高模型生成更随机,0表示贪婪采样,默认为1;`max_tokens=16`为生成长度,默认为1;
`model`为模型路径;`tensor_parallel_size=1`为使用卡数,默认为1;`dtype="float16"`为推理数据类型,如果模型权重是bfloat16,需要修改为float16推理,`quantization="gptq"`为使用gptq量化进行推理,需下载以上GPTQ模型。


### 离线批量推理性能测试
1、指定输入输出
```bash
zhuwenwen's avatar
zhuwenwen committed
100
python benchmarks/benchmark_throughput.py --num-prompts 1 --input-len 32 --output-len 128 --model THUDM/glm-4-9b-chat -tp 1 --trust-remote-code --enforce-eager --dtype float16
zhuwenwen's avatar
zhuwenwen committed
101
102
103
104
105
106
107
108
109
110
```
其中`--num-prompts`是batch数,`--input-len`是输入seqlen,`--output-len`是输出token长度,`--model`为模型路径,`-tp`为使用卡数,`dtype="float16"`为推理数据类型,如果模型权重是bfloat16,需要修改为float16推理。若指定`--output-len  1`即为首字延迟。`-q gptq`为使用gptq量化模型进行推理。

2、使用数据集
下载数据集:
```bash
wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json
```

```bash
zhuwenwen's avatar
zhuwenwen committed
111
python benchmarks/benchmark_throughput.py --num-prompts 1 --model THUDM/glm-4-9b-chat --dataset ShareGPT_V3_unfiltered_cleaned_split.json -tp 1 --trust-remote-code --enforce-eager --dtype float16
zhuwenwen's avatar
zhuwenwen committed
112
113
114
115
116
117
118
```
其中`--num-prompts`是batch数,`--model`为模型路径,`--dataset`为使用的数据集,`-tp`为使用卡数,`dtype="float16"`为推理数据类型,如果模型权重是bfloat16,需要修改为float16推理。`-q gptq`为使用gptq量化模型进行推理。


### api服务推理性能测试
1、启动服务端:
```bash
zhuwenwen's avatar
zhuwenwen committed
119
python -m vllm.entrypoints.openai.api_server  --model THUDM/glm-4-9b-chat  --dtype float16 --enforce-eager -tp 1 
zhuwenwen's avatar
zhuwenwen committed
120
121
122
123
```

2、启动客户端:
```bash
zhuwenwen's avatar
zhuwenwen committed
124
python benchmarks/benchmark_serving.py --model THUDM/glm-4-9b-chat --dataset ShareGPT_V3_unfiltered_cleaned_split.json  --num-prompts 1 --trust-remote-code
zhuwenwen's avatar
zhuwenwen committed
125
```
zhuwenwen's avatar
zhuwenwen committed
126
参数同使用数据集,离线批量推理性能测试,具体参考[benchmarks/benchmark_serving.py](benchmarks/benchmark_serving.py)
zhuwenwen's avatar
zhuwenwen committed
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151


### OpenAI兼容服务
启动服务:
```bash
python -m vllm.entrypoints.openai.api_server --model THUDM/glm-4-9b-chat --enforce-eager --dtype float16 --trust-remote-code
```
这里`--model`为加载模型路径,`--dtype`为数据类型:float16,默认情况使用tokenizer中的预定义聊天模板,`--chat-template`可以添加新模板覆盖默认模板,`-q gptq`为使用gptq量化模型进行推理。

列出模型型号:
```bash
curl http://localhost:8000/v1/models
```

### OpenAI Completions API和vllm结合使用
```bash
curl http://localhost:8000/v1/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "THUDM/glm-4-9b-chat",
        "prompt": "晚上睡不着怎么办",
        "max_tokens": 7,
        "temperature": 0
    }'
```
zhuwenwen's avatar
zhuwenwen committed
152
或者使用[examples/openai_completion_client.py](examples/openai_completion_client.py)
zhuwenwen's avatar
zhuwenwen committed
153
154
155
156
157
158
159
160
161
162
163
164
165
166


### OpenAI Chat API和vllm结合使用
```bash
curl http://localhost:8000/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "THUDM/glm-4-9b-chat",
        "messages": [
            {"role": "system", "content": "晚上睡不着怎么办"},
            {"role": "user", "content": "晚上睡不着怎么办"}
        ]
    }'
```
zhuwenwen's avatar
zhuwenwen committed
167
或者使用[examples/openai_chatcompletion_client.py](examples/openai_chatcompletion_client.py)
laibao's avatar
laibao committed
168
### **gradio和vllm结合使用**
zhuwenwen's avatar
zhuwenwen committed
169

laibao's avatar
laibao committed
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
1.安装gradio

```
pip install gradio
```

2.安装必要文件

    2.1 启动gradio服务,根据提示操作

```
python  gradio_openai_chatbot_webserver.py --model "THUDM/glm-4-9b-chat" --model-url http://localhost:8000/v1 --temp 0.8 --stop-token-ids ""
```

    2.2 更改文件权限

打开提示下载文件目录,输入以下命令给予权限

```
chmod +x frpc_linux_amd64_v0.*
```

3.启动OpenAI兼容服务

```
python -m vllm.entrypoints.openai.api_server --model THUDM/glm-4-9b-chatt --enforce-eager --dtype float16 --trust-remote-code --port 8000
```

4.启动gradio服务

```
python  gradio_openai_chatbot_webserver.py --model "THUDM/glm-4-9b-chat" --model-url http://localhost:8000/v1 --temp 0.8 --stop-token-ids ""
```

5.使用对话服务

在浏览器中输入本地 URL,可以使用 Gradio 提供的对话服务。
zhuwenwen's avatar
zhuwenwen committed
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229

## result
使用的加速卡:1张 DCU-K100_AI-64G
```
Prompt: '晚上睡不着怎么办', Generated text: '?\n晚上睡不着可以尝试以下方法来改善睡眠质量:\n\n1. **调整作息时间**:尽量每天同一时间上床睡觉和起床,建立规律的生物钟。\n\n2. **放松身心**:睡前进行深呼吸、冥想或瑜伽等放松活动,有助于减轻压力和焦虑。\n\n3. **避免咖啡因和酒精**:晚上避免摄入咖啡因和酒精,因为它们可能会干扰睡眠。\n\n'
```

### 精度


## 应用场景

### 算法类别
对话问答

### 热点应用行业
医疗,金融,科研,教育

## 源码仓库及问题反馈
* [https://developer.hpccube.com/codes/modelzoo/llama_vllm](https://developer.hpccube.com/codes/modelzoo/chatglm_vllm)

## 参考资料
* [https://github.com/vllm-project/vllm](https://github.com/vllm-project/vllm)
laibao's avatar
laibao committed
230
* [https://github.com/THUDM/ChatGLM3](https://github.com/THUDM/ChatGLM3)