Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -1,11 +1,8 @@
|
|
1 |
import gradio as gr
|
2 |
from huggingface_hub import InferenceClient
|
3 |
|
4 |
-
|
5 |
-
|
6 |
-
"""
|
7 |
-
client = InferenceClient("TheBloke/claude2-alpaca-13B-GGUF")
|
8 |
-
|
9 |
|
10 |
def respond(
|
11 |
message,
|
@@ -15,6 +12,18 @@ def respond(
|
|
15 |
temperature,
|
16 |
top_p,
|
17 |
):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
messages = [{"role": "system", "content": system_message}]
|
19 |
|
20 |
for val in history:
|
@@ -27,6 +36,7 @@ def respond(
|
|
27 |
|
28 |
response = ""
|
29 |
|
|
|
30 |
for message in client.chat_completion(
|
31 |
messages,
|
32 |
max_tokens=max_tokens,
|
@@ -35,30 +45,42 @@ def respond(
|
|
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(
|
50 |
-
|
51 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
|
4 |
+
# Initialize the inference client with the model ID
|
5 |
+
client = InferenceClient(model="umd-zhou-lab/claude2_alpaca", token=None) # Add your HF token if needed
|
|
|
|
|
|
|
6 |
|
7 |
def respond(
|
8 |
message,
|
|
|
12 |
temperature,
|
13 |
top_p,
|
14 |
):
|
15 |
+
"""
|
16 |
+
Generate responses for the chatbot using the Claude2 Alpaca model.
|
17 |
+
|
18 |
+
Args:
|
19 |
+
message (str): The current user input message
|
20 |
+
history (list): List of previous conversation turns
|
21 |
+
system_message (str): System prompt to guide the model's behavior
|
22 |
+
max_tokens (int): Maximum number of tokens to generate
|
23 |
+
temperature (float): Controls randomness in generation
|
24 |
+
top_p (float): Controls nucleus sampling
|
25 |
+
"""
|
26 |
+
# Format the conversation history into messages
|
27 |
messages = [{"role": "system", "content": system_message}]
|
28 |
|
29 |
for val in history:
|
|
|
36 |
|
37 |
response = ""
|
38 |
|
39 |
+
# Stream the response tokens
|
40 |
for message in client.chat_completion(
|
41 |
messages,
|
42 |
max_tokens=max_tokens,
|
|
|
45 |
top_p=top_p,
|
46 |
):
|
47 |
token = message.choices[0].delta.content
|
|
|
48 |
response += token
|
49 |
yield response
|
50 |
|
51 |
+
# Create the Gradio interface
|
|
|
|
|
|
|
52 |
demo = gr.ChatInterface(
|
53 |
respond,
|
54 |
additional_inputs=[
|
55 |
+
gr.Textbox(
|
56 |
+
value="You are a friendly Chatbot trained on the Claude2 Alpaca dataset. Provide helpful and informative responses.",
|
57 |
+
label="System message"
|
58 |
+
),
|
59 |
+
gr.Slider(
|
60 |
+
minimum=1,
|
61 |
+
maximum=2048,
|
62 |
+
value=512,
|
63 |
+
step=1,
|
64 |
+
label="Max new tokens"
|
65 |
+
),
|
66 |
+
gr.Slider(
|
67 |
+
minimum=0.1,
|
68 |
+
maximum=4.0,
|
69 |
+
value=0.7,
|
70 |
+
step=0.1,
|
71 |
+
label="Temperature"
|
72 |
+
),
|
73 |
gr.Slider(
|
74 |
minimum=0.1,
|
75 |
maximum=1.0,
|
76 |
value=0.95,
|
77 |
step=0.05,
|
78 |
+
label="Top-p (nucleus sampling)"
|
79 |
),
|
80 |
],
|
81 |
+
title="Claude2 Alpaca Chatbot",
|
82 |
+
description="A conversational AI powered by the Claude2 Alpaca model",
|
83 |
)
|
84 |
|
|
|
85 |
if __name__ == "__main__":
|
86 |
+
demo.launch()
|