sudo-soldier commited on
Commit
aebe3da
·
verified ·
1 Parent(s): b335d43

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +25 -30
app.py CHANGED
@@ -1,9 +1,12 @@
1
  import gradio as gr
2
  from huggingface_hub import InferenceClient
3
 
4
-
 
 
5
  client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
6
 
 
7
  def respond(
8
  message,
9
  history: list[tuple[str, str]],
@@ -12,21 +15,18 @@ def respond(
12
  temperature,
13
  top_p,
14
  ):
15
-
16
  messages = [{"role": "system", "content": system_message}]
17
-
18
 
19
- for user_msg, assistant_msg in history:
20
- if user_msg:
21
- messages.append({"role": "user", "content": user_msg})
22
- if assistant_msg:
23
- messages.append({"role": "assistant", "content": assistant_msg})
24
 
25
-
26
  messages.append({"role": "user", "content": message})
27
 
28
-
29
  response = ""
 
30
  for message in client.chat_completion(
31
  messages,
32
  max_tokens=max_tokens,
@@ -35,35 +35,30 @@ def respond(
35
  top_p=top_p,
36
  ):
37
  token = message.choices[0].delta.content
 
38
  response += token
39
  yield response
40
 
41
 
 
 
 
42
  demo = gr.ChatInterface(
43
- fn=respond,
44
  additional_inputs=[
45
- gr.Textbox(
46
- value=(
47
- "Your name is Jesse. You are a highly professional, results-driven AI assistant. "
48
- "You are a CompTIA Tech+ certified technician with a bachelor's degree in Information Technology. "
49
- "You specialize in networking, IT support, cybersecurity, and troubleshooting. "
50
- "Do not use humor or casual language. Respond succinctly and in a tone similar to a certified IT helpdesk technician. "
51
- "Focus on providing accurate, concise, and actionable technical assistance. Always refer to yourself as Jesse."
52
- ),
53
- label="System message"
54
- ),
55
  gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
56
- gr.Slider(minimum=0.1, maximum=4.0, value=0.6, step=0.1, label="Temperature"), # Slightly lowered for focus
57
- gr.Slider(minimum=0.1, maximum=1.0, value=0.85, step=0.05, label="Top-p (nucleus sampling)"),
 
 
 
 
 
 
58
  ],
59
- title="Jesse - CompTIA Tech+ Certified Technician",
60
- description="Speak with Jesse, a CompTIA Tech+ certified technician with a bachelor's degree in IT. Ask anything about networks, IT support, security, and diagnostics.",
61
- theme="default"
62
  )
63
 
64
 
65
  if __name__ == "__main__":
66
- demo.launch()
67
-
68
-
69
-
 
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]],
 
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,
 
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()