AnkitkumarMalde commited on
Commit
35766a3
·
verified ·
1 Parent(s): 2ff7a4b

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +37 -61
app.py CHANGED
@@ -1,69 +1,45 @@
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
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
- chatbot = 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
- with gr.Blocks() as demo:
63
- with gr.Sidebar():
64
- gr.LoginButton()
65
- chatbot.render()
66
-
67
-
68
  if __name__ == "__main__":
69
- demo.launch()
 
1
  import gradio as gr
2
+ from transformers import pipeline
3
+ import torch
4
 
5
+ # 1. Initialize the Model
6
+ # We use 'device_map="auto"' to handle hardware efficiently
7
+ # If using a specific model from Kaggle, replace "microsoft/DialoGPT-medium" with your model ID
8
+ model_id = "microsoft/DialoGPT-medium"
9
+ chatbot_pipeline = pipeline("text-generation", model=model_id, device_map="auto")
10
 
11
+ # 2. Define the Inference Logic
12
+ def chat_function(message, history):
 
 
 
 
 
 
 
13
  """
14
+ message: The current user input
15
+ history: A list of previous [user_msg, bot_msg] pairs provided by Gradio
16
  """
17
+ # Combine history into a single string for context (simplified)
18
+ context = ""
19
+ for user_msg, bot_msg in history:
20
+ context += f"User: {user_msg}\nAssistant: {bot_msg}\n"
21
+
22
+ context += f"User: {message}\nAssistant:"
23
+
24
+ # Generate response
25
+ output = chatbot_pipeline(context, max_new_tokens=100, pad_token_id=50256)
26
+
27
+ # Extract only the newly generated text
28
+ generated_text = output[0]['generated_text']
29
+ response = generated_text.split("Assistant:")[-1].strip()
30
+
31
+ return response
32
+
33
+ # 3. Build the Gradio UI
34
+ # 'ChatInterface' automatically handles the text box and 'Send' button logic
35
+ demo = gr.ChatInterface(
36
+ fn=chat_function,
37
+ title="My Kaggle-to-HF Chatbot",
38
+ description="Ask me anything! This bot was developed in Kaggle and hosted on Hugging Face.",
39
+ examples=["Hello!", "What is artificial intelligence?", "Tell me a joke."],
40
+ theme="soft"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41
  )
42
 
43
+ # 4. Launch
 
 
 
 
 
44
  if __name__ == "__main__":
45
+ demo.launch()