Futuresony commited on
Commit
fb97bf4
·
verified ·
1 Parent(s): ed6f3c7

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +41 -31
app.py CHANGED
@@ -1,54 +1,64 @@
1
  import gradio as gr
2
  from huggingface_hub import InferenceClient
3
- from collections import defaultdict
4
 
5
- # Initialize model client
 
 
6
  client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")
7
 
8
- # Memory storage
9
- session_histories = defaultdict(list) # Stores full chat history per session
10
 
11
- def format_chat_history(history):
12
- """Formats history in a structured way for LLaMA models."""
13
- chat_str = ""
14
- for user_msg, bot_response in history:
15
- chat_str += f"User: {user_msg}\nAI: {bot_response}\n"
16
- return chat_str.strip() # Remove unnecessary spaces
 
 
 
17
 
18
- def respond(message, history, system_message, max_tokens, temperature, top_p):
19
- session_id = id(history) # Unique session ID
20
- session_history = session_histories[session_id] # Retrieve stored history
 
 
21
 
22
- # Add user message to history
23
- formatted_history = format_chat_history(session_history)
24
- full_input = f"{system_message}\n\n{formatted_history}\nUser: {message}\nAI:"
25
 
26
- # Generate response
27
- response = client.text_generation(
28
- full_input,
29
- max_new_tokens=max_tokens,
 
 
30
  temperature=temperature,
31
  top_p=top_p,
32
- )
33
-
34
- # Extract & clean response
35
- cleaned_response = response.strip()
36
 
37
- # Update chat history
38
- session_history.append((message, cleaned_response))
39
 
40
- return cleaned_response
41
 
42
- # Gradio Chat Interface
 
 
43
  demo = gr.ChatInterface(
44
  respond,
45
  additional_inputs=[
46
- gr.Textbox(value="You are an AI assistant that remembers previous conversations.", label="System message"),
47
- gr.Slider(minimum=1, maximum=2048, value=250, step=1, label="Max new tokens"),
48
  gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
49
- gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
 
 
 
 
 
 
50
  ],
51
  )
52
 
 
53
  if __name__ == "__main__":
54
  demo.launch()
 
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("Futuresony/future_ai_12_10_2024.gguf")
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,
33
+ stream=True,
34
  temperature=temperature,
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"),
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
+
63
  if __name__ == "__main__":
64
  demo.launch()