thuanan commited on
Commit
ff3e6d7
·
1 Parent(s): f3e9dde

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +76 -38
app.py CHANGED
@@ -1,64 +1,102 @@
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
- """
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()
 
 
1
  import gradio as gr
2
  from huggingface_hub import InferenceClient
3
+ import os
4
+ import base64
5
 
 
 
 
6
  client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
7
 
8
+ # Base64 encode image
9
+ def image_to_base64(image_path):
10
+ try:
11
+ with open(image_path, "rb") as image_file:
12
+ return base64.b64encode(image_file.read()).decode("utf-8")
13
+ except:
14
+ return ""
15
 
16
+ # Header HTML
17
+ def get_header_html():
18
+ if os.path.exists("static/aivn_logo.png"):
19
+ logo_base64 = image_to_base64("static/aivn_logo.png")
20
+ logo_html = f"""
21
+ <img src="data:image/png;base64,{logo_base64}"
22
+ alt="Logo"
23
+ style="height: 120px; width: auto; margin-right: 20px; margin-bottom: 20px;">
24
+ """
25
+ else:
26
+ logo_html = "<h2>AI VIETNAM</h2>"
27
+
28
+ return f"""
29
+ <div style="display: flex; align-items: center; padding: 20px;">
30
+ <div>{logo_html}</div>
31
+ <div style="margin-left: 20px;">
32
+ <h1>🧠 Zephyr Chatbot</h1>
33
+ <p>Conversational Assistant powered by HuggingFace Zephyr-7B</p>
34
+ <p style="color: #555;">🚀 Customize system prompt, temperature, top-p, etc.</p>
35
+ </div>
36
+ </div>
37
+ """
38
 
39
+ # Footer HTML
40
+ def get_footer_html():
41
+ return """
42
+ <div style="
43
+ position: fixed;
44
+ bottom: 0;
45
+ left: 0;
46
+ width: 100%;
47
+ background: white;
48
+ padding: 10px;
49
+ text-align: center;
50
+ box-shadow: 0 -2px 10px rgba(0,0,0,0.1);
51
+ z-index: 1000;
52
+ font-size: 14px;
53
+ ">
54
+ Created by <a href="https://vlai.work" target="_blank" style="color: #007BFF;">VLAI</a> • AI VIETNAM
55
+ </div>
56
+ """
57
+
58
+ # Chat function
59
+ def respond(message, history, system_message, max_tokens, temperature, top_p):
60
+ messages = [{"role": "system", "content": system_message}]
61
+ for msg in history:
62
+ messages.append(msg)
63
 
64
  messages.append({"role": "user", "content": message})
65
 
66
  response = ""
67
+ for chunk in client.chat_completion(
 
68
  messages,
69
  max_tokens=max_tokens,
70
  stream=True,
71
  temperature=temperature,
72
  top_p=top_p,
73
  ):
74
+ token = chunk.choices[0].delta.content
 
75
  response += token
76
+ yield {"role": "assistant", "content": response}
77
 
78
 
79
+ # UI
80
+ chat = gr.ChatInterface(
81
+ fn=respond,
 
 
82
  additional_inputs=[
83
+ gr.Textbox(value="You are a friendly chatbot.", label="System message"),
84
+ gr.Slider(1, 2048, value=512, step=1, label="Max new tokens"),
85
+ gr.Slider(0.1, 4.0, value=0.7, step=0.1, label="Temperature"),
86
+ gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")
 
 
 
 
 
 
87
  ],
88
+ title="Zephyr Chatbot",
89
+ description="A customizable chat assistant powered by HuggingFace H4 Zephyr-7B.",
90
+ type="messages",
91
  )
92
 
93
 
94
+ # Combine with layout elements
95
+ with gr.Blocks(css=".gradio-container {min-height: 100vh;}") as demo:
96
+ gr.HTML(get_header_html())
97
+ chat.render()
98
+ gr.HTML(get_footer_html())
99
+
100
  if __name__ == "__main__":
101
+ os.makedirs("static", exist_ok=True)
102
+ demo.launch(server_name="0.0.0.0", server_port=7860)