HydroFlyer53 commited on
Commit
9fa5fdb
·
verified ·
1 Parent(s): 94052bf

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +18 -37
app.py CHANGED
@@ -1,37 +1,32 @@
1
  import gradio as gr
2
  from huggingface_hub import InferenceClient
3
 
4
-
5
- def respond(
6
- message,
7
- history: list[dict[str, str]],
8
- system_message,
9
- max_tokens,
10
- temperature,
11
- top_p,
12
- hf_token: gr.OAuthToken,
13
- ):
14
  """
15
- 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
16
  """
17
- client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
18
-
19
- messages = [{"role": "system", "content": system_message}]
20
 
 
21
  messages.extend(history)
22
-
23
  messages.append({"role": "user", "content": message})
24
 
25
  response = ""
26
-
27
- for message in client.chat_completion(
28
  messages,
29
- max_tokens=max_tokens,
30
  stream=True,
31
- temperature=temperature,
32
- top_p=top_p,
33
  ):
34
- choices = message.choices
35
  token = ""
36
  if len(choices) and choices[0].delta.content:
37
  token = choices[0].delta.content
@@ -40,24 +35,11 @@ def respond(
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
  chatbot = gr.ChatInterface(
47
  respond,
48
  type="messages",
49
- additional_inputs=[
50
- gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
51
- gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
52
- gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
53
- gr.Slider(
54
- minimum=0.1,
55
- maximum=1.0,
56
- value=0.95,
57
- step=0.05,
58
- label="Top-p (nucleus sampling)",
59
- ),
60
- ],
61
  )
62
 
63
  with gr.Blocks() as demo:
@@ -65,6 +47,5 @@ with gr.Blocks() as demo:
65
  gr.LoginButton()
66
  chatbot.render()
67
 
68
-
69
  if __name__ == "__main__":
70
  demo.launch()
 
1
  import gradio as gr
2
  from huggingface_hub import InferenceClient
3
 
4
+ # === SETTINGS YOU CAN CHANGE IN THE CODE ===
5
+ SYSTEM_MESSAGE = "You are a chatbot that has to use gen-z slang as much as you can."
6
+ MAX_TOKENS = 512
7
+ TEMPERATURE = 0.7
8
+ TOP_P = 0.95
9
+ HF_MODEL = "openai/gpt-oss-20b"
10
+
11
+ def respond(message, history: list[dict[str, str]], hf_token: gr.OAuthToken):
 
 
12
  """
13
+ Chat function using Hugging Face Inference API with preset parameters.
14
  """
15
+ client = InferenceClient(token=hf_token.token, model=HF_MODEL)
 
 
16
 
17
+ messages = [{"role": "system", "content": SYSTEM_MESSAGE}]
18
  messages.extend(history)
 
19
  messages.append({"role": "user", "content": message})
20
 
21
  response = ""
22
+ for message_chunk in client.chat_completion(
 
23
  messages,
24
+ max_tokens=MAX_TOKENS,
25
  stream=True,
26
+ temperature=TEMPERATURE,
27
+ top_p=TOP_P,
28
  ):
29
+ choices = message_chunk.choices
30
  token = ""
31
  if len(choices) and choices[0].delta.content:
32
  token = choices[0].delta.content
 
35
  yield response
36
 
37
 
38
+ # === GRADIO INTERFACE ===
 
 
39
  chatbot = gr.ChatInterface(
40
  respond,
41
  type="messages",
42
+ additional_inputs=[], # no sliders or textboxes
 
 
 
 
 
 
 
 
 
 
 
43
  )
44
 
45
  with gr.Blocks() as demo:
 
47
  gr.LoginButton()
48
  chatbot.render()
49
 
 
50
  if __name__ == "__main__":
51
  demo.launch()