coco1990 commited on
Commit
f4bb85a
·
verified ·
1 Parent(s): e7fe4cc

Upload 2 files

Browse files
Files changed (2) hide show
  1. app.py +45 -0
  2. requirements.txt +3 -0
app.py ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+ from llama_cpp import Llama
3
+ from huggingface_hub import hf_hub_download
4
+
5
+ # 1. 下载模型(针对 Qwen 官方 GGUF 仓库的命名规范)
6
+ # 建议使用 Q4_K_M 版本,平衡性能与 16GB 内存限制
7
+ model_path = hf_hub_download(
8
+ repo_id="Qwen/Qwen2.5-Coder-7B-Instruct-GGUF",
9
+ filename="qwen2.5-coder-7b-instruct-q4_k_m.gguf"
10
+ )
11
+
12
+ # 2. 初始化模型
13
+ llm = Llama(
14
+ model_path=model_path,
15
+ n_ctx=4096, # 运维场景通常需要处理较长日志,设置为 4k
16
+ n_threads=2 # 对应免费 CPU 的核数
17
+ )
18
+
19
+ def respond(message, history):
20
+ system_prompt = "你是一位资深大数据运维专家。请为用户提供 Hadoop/Spark/Flink 等组件的代码调优、故障排查建议或 Shell 脚本。"
21
+
22
+ # 构造对话历史
23
+ messages = [{"role": "system", "content": system_prompt}]
24
+ for user_msg, assistant_msg in history:
25
+ messages.append({"role": "user", "content": user_msg})
26
+ messages.append({"role": "assistant", "content": assistant_msg})
27
+ messages.append({"role": "user", "content": message})
28
+
29
+ # 流式生成
30
+ response_text = ""
31
+ for chunk in llm.create_chat_completion(messages=messages, stream=True):
32
+ delta = chunk['choices'][0]['delta']
33
+ if 'content' in delta:
34
+ response_text += delta['content']
35
+ yield response_text
36
+
37
+ # 3. 启动 Gradio 界面
38
+ demo = gr.ChatInterface(
39
+ fn=respond,
40
+ title="BigData Ops Copilot (Qwen2.5-Coder-7B)",
41
+ examples=["如何优化 Spark 任务的内存分配?", "写一个监控 Kafka 消费延迟的 Python 脚本"]
42
+ )
43
+
44
+ if __name__ == "__main__":
45
+ demo.launch()
requirements.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ llama-cpp-python
2
+ gradio
3
+ huggingface_hub