90 commited on
Commit
9267136
·
1 Parent(s): 13fa91c

Update space

Browse files
Files changed (3) hide show
  1. README.md +43 -1
  2. app.py +41 -58
  3. requirements.txt +4 -1
README.md CHANGED
@@ -9,4 +9,46 @@ app_file: app.py
9
  pinned: false
10
  ---
11
 
12
- An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  pinned: false
10
  ---
11
 
12
+ # MiMo-7B-RL 聊天机器人
13
+
14
+ 这是一个使用小米 MiMo-7B-RL 模型构建的简单聊天机器人应用。
15
+
16
+ ## 环境要求
17
+
18
+ - Python 3.8+
19
+ - CUDA 支持(推荐用于更快的推理)
20
+
21
+ ## 安装
22
+
23
+ 1. 克隆此仓库:
24
+ ```bash
25
+ git clone <repository-url>
26
+ cd <repository-name>
27
+ ```
28
+
29
+ 2. 安装依赖:
30
+ ```bash
31
+ pip install -r requirements.txt
32
+ ```
33
+
34
+ ## 运行应用
35
+
36
+ 执行以下命令启动聊天机器人:
37
+ ```bash
38
+ python app.py
39
+ ```
40
+
41
+ 启动后,应用将在本地运行,并提供一个网页界面(通常是 http://localhost:7860)。
42
+
43
+ ## 功能特点
44
+
45
+ - 基于小米 MiMo-7B-RL 大语言模型
46
+ - 支持多轮对话
47
+ - 简洁的 Gradio 网页界面
48
+ - 内置示例问题
49
+
50
+ ## 注意事项
51
+
52
+ - 首次运行时需要下载模型,可能需要一些时间
53
+ - 建议使用具有足够显存的 GPU 来运行模型(建议至少 16GB 显存)
54
+ - 如果遇到显存不足的问题,可以尝试调整 `app.py` 中的生成参数
app.py CHANGED
@@ -1,64 +1,47 @@
1
  import gradio as gr
2
- from huggingface_hub import InferenceClient
3
-
4
- """
5
- For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
6
- """
7
- client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
8
-
9
-
10
- def respond(
11
- message,
12
- history: list[tuple[str, str]],
13
- system_message,
14
- max_tokens,
15
- temperature,
16
- top_p,
17
- ):
18
- messages = [{"role": "system", "content": system_message}]
19
-
20
- for val in history:
21
- if val[0]:
22
- messages.append({"role": "user", "content": val[0]})
23
- if val[1]:
24
- messages.append({"role": "assistant", "content": val[1]})
25
-
26
- messages.append({"role": "user", "content": message})
27
-
28
- response = ""
29
-
30
- for message in client.chat_completion(
31
- messages,
32
- max_tokens=max_tokens,
33
- stream=True,
34
- temperature=temperature,
35
- top_p=top_p,
36
- ):
37
- token = message.choices[0].delta.content
38
-
39
- response += token
40
- yield response
41
-
42
 
43
- """
44
- For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
45
- """
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46
  demo = gr.ChatInterface(
47
- respond,
48
- additional_inputs=[
49
- gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
50
- gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
51
- gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
52
- gr.Slider(
53
- minimum=0.1,
54
- maximum=1.0,
55
- value=0.95,
56
- step=0.05,
57
- label="Top-p (nucleus sampling)",
58
- ),
59
- ],
60
  )
61
 
62
-
63
  if __name__ == "__main__":
64
- demo.launch()
 
1
  import gradio as gr
2
+ from transformers import AutoTokenizer, AutoModelForCausalLM
3
+ import torch
4
+
5
+ # 加载模型和分词器
6
+ model_name = "XiaomiMiMo/MiMo-7B-RL"
7
+ tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
8
+ model = AutoModelForCausalLM.from_pretrained(
9
+ model_name,
10
+ torch_dtype=torch.float16,
11
+ device_map="auto",
12
+ trust_remote_code=True
13
+ )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
 
15
+ def predict(message, history):
16
+ # 构建输入
17
+ history_text = ""
18
+ for human, assistant in history:
19
+ history_text += f"Human: {human}\nAssistant: {assistant}\n"
20
+ prompt = f"{history_text}Human: {message}\nAssistant:"
21
+
22
+ # 生成回复
23
+ inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
24
+ outputs = model.generate(
25
+ **inputs,
26
+ max_new_tokens=512,
27
+ do_sample=True,
28
+ temperature=0.7,
29
+ top_p=0.9,
30
+ repetition_penalty=1.1,
31
+ pad_token_id=tokenizer.eos_token_id
32
+ )
33
+ response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
34
+
35
+ return response.strip()
36
+
37
+ # 创建Gradio界面
38
  demo = gr.ChatInterface(
39
+ predict,
40
+ title="MiMo-7B-RL 聊天机器人",
41
+ description="这是一个基于小米 MiMo-7B-RL 模型的聊天机器人。",
42
+ examples=["你好!", "请介绍一下你自己", "你能做什么?"],
43
+ theme=gr.themes.Soft()
 
 
 
 
 
 
 
 
44
  )
45
 
 
46
  if __name__ == "__main__":
47
+ demo.launch(share=True)
requirements.txt CHANGED
@@ -1 +1,4 @@
1
- huggingface_hub==0.25.2
 
 
 
 
1
+ gradio==4.19.2
2
+ torch>=2.0.0
3
+ transformers>=4.36.0
4
+ accelerate>=0.27.0