zxl2025 commited on
Commit
7081534
·
verified ·
1 Parent(s): b0f5fa9

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +21 -0
app.py ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+ from transformers import AutoTokenizer, AutoModel
3
+ import torch
4
+
5
+ # 加载模型
6
+ model_path = "THUDM/chatglm2-6b"
7
+ tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
8
+ model = AutoModel.from_pretrained(model_path, trust_remote_code=True)
9
+ model = model.half() # 使用 CPU 而不是 GPU
10
+ model.eval()
11
+
12
+ def generate_code(text):
13
+ inputs = tokenizer(text, return_tensors="pt") # 删去 `.to('cuda')`
14
+ with torch.no_grad():
15
+ response = model.generate(**inputs, max_length=200)
16
+ output_text = tokenizer.decode(response[0], skip_special_tokens=True)
17
+ return output_text
18
+
19
+ # 创建 Gradio 接口
20
+ demo = gr.Interface(fn=generate_code, inputs="text", outputs="text")
21
+ demo.launch()