vishwesh21 commited on
Commit
0c0adaa
·
verified ·
1 Parent(s): 0fa8a6f

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +6 -29
app.py CHANGED
@@ -1,32 +1,15 @@
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,
@@ -35,16 +18,12 @@ def respond(
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"),
@@ -54,11 +33,9 @@ demo = gr.ChatInterface(
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 huggingface_hub import InferenceClient
3
 
 
 
 
4
  client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
5
 
6
+ def respond(message, history, system_message, max_tokens, temperature, top_p):
7
+ # history is now a list of {"role": ..., "content": ...}
 
 
 
 
 
 
 
8
  messages = [{"role": "system", "content": system_message}]
9
+ messages += history # history is already in the correct format
 
 
 
 
 
 
10
  messages.append({"role": "user", "content": message})
11
 
12
  response = ""
 
13
  for message in client.chat_completion(
14
  messages,
15
  max_tokens=max_tokens,
 
18
  top_p=top_p,
19
  ):
20
  token = message.choices[0].delta.content
 
21
  response += token
22
  yield response
23
 
 
 
 
 
24
  demo = gr.ChatInterface(
25
+ fn=respond,
26
+ chatbot=gr.Chatbot(type="messages"), # ✅ specify new message format
27
  additional_inputs=[
28
  gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
29
  gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
 
33
  maximum=1.0,
34
  value=0.95,
35
  step=0.05,
36
+ label="Top-p (nucleus sampling)"),
 
37
  ],
38
  )
39
 
 
40
  if __name__ == "__main__":
41
  demo.launch()