Gyu-min Nigel Lee commited on
Commit
ee17265
1 Parent(s): 55ad9bf

updated the app

Browse files
Files changed (1) hide show
  1. app.py +21 -24
app.py CHANGED
@@ -1,43 +1,38 @@
 
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,
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
  For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
@@ -45,9 +40,11 @@ For information on how to customize the ChatInterface, peruse the gradio docs: h
45
  demo = gr.ChatInterface(
46
  respond,
47
  additional_inputs=[
48
- gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
 
 
49
  gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
50
- gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
51
  gr.Slider(
52
  minimum=0.1,
53
  maximum=1.0,
@@ -60,4 +57,4 @@ demo = gr.ChatInterface(
60
 
61
 
62
  if __name__ == "__main__":
63
- demo.launch()
 
1
+ import torch
2
  import gradio as gr
3
+ from transformers import pipeline
4
 
5
  """
6
  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
7
  """
8
+ # client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
9
+ pipe = pipeline(
10
+ model="gyulukeyi/llama3-nallm-qa-8B_240708",
11
+ torch_dtype=torch.float16,
12
+ )
13
 
14
 
15
  def respond(
16
+ municipality,
17
+ title,
18
+ question,
19
  max_tokens,
20
  temperature,
21
  top_p,
22
  ):
23
+ messages = [{"role": "municipality", "content": municipality}]
24
+ messages.append({"role": "title", "content": title})
25
+ messages.append({"role": "question", "content": question})
 
 
 
 
 
 
 
 
26
 
27
+ output = pipe(
28
  messages,
29
  max_tokens=max_tokens,
30
  stream=True,
31
  temperature=temperature,
32
  top_p=top_p,
33
+ )
34
+ return output[0]["generated_text"]
35
 
 
 
36
 
37
  """
38
  For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 
40
  demo = gr.ChatInterface(
41
  respond,
42
  additional_inputs=[
43
+ gr.Textbox(value="서울특별시", label="municipality"),
44
+ gr.Textbox(value="", label="title"),
45
+ gr.Textbox(value="", label="question"),
46
  gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
47
+ gr.Slider(minimum=0.1, maximum=1.0, value=0.5, step=0.05, label="Temperature"),
48
  gr.Slider(
49
  minimum=0.1,
50
  maximum=1.0,
 
57
 
58
 
59
  if __name__ == "__main__":
60
+ demo.launch()