Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,90 +1,26 @@
|
|
| 1 |
|
| 2 |
|
| 3 |
import gradio as gr
|
| 4 |
-
|
| 5 |
-
|
| 6 |
import random as rd
|
| 7 |
-
|
| 8 |
-
|
| 9 |
from huggingface_hub import InferenceClient
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
client = InferenceClient("Qwen/Qwen2.5-72B-Instruct")
|
| 18 |
-
|
| 19 |
-
|
| 20 |
def respond (message, history):
|
| 21 |
-
|
| 22 |
-
|
| 23 |
messages = [{"role": "system", "content": "You are a chatbot who is sassy and doesnt explain an anser unless someone asks twice."}]
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
if history:
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
messages.extend(history)
|
| 32 |
-
|
| 33 |
-
|
| 34 |
messages.append({"role": "user", "content": message})
|
| 35 |
-
|
| 36 |
-
|
| 37 |
response = ""
|
| 38 |
-
|
| 39 |
-
|
| 40 |
for message in client.chat_completion(
|
| 41 |
-
|
| 42 |
-
|
| 43 |
messages,
|
| 44 |
-
|
| 45 |
-
|
| 46 |
max_tokens=500,
|
| 47 |
-
|
| 48 |
-
|
| 49 |
temperature=0.2,
|
| 50 |
-
|
| 51 |
-
|
| 52 |
top_p=0.9,
|
| 53 |
-
|
| 54 |
-
|
| 55 |
stream=True
|
| 56 |
-
|
| 57 |
-
|
| 58 |
):
|
| 59 |
-
|
| 60 |
-
|
| 61 |
token = message.choices[0].delta.content
|
| 62 |
-
|
| 63 |
-
|
| 64 |
response += token
|
| 65 |
-
|
| 66 |
-
|
| 67 |
yield response
|
| 68 |
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
chatbot = gr.ChatInterface(respond, type = "messages")
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
chatbot.launch(debug=True)
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
# Define the Gradio interface
|
| 89 |
css = """
|
| 90 |
.gradio-container {
|
|
@@ -92,6 +28,10 @@ css = """
|
|
| 92 |
font-family: 'Times New Roman', Times, serif;
|
| 93 |
}
|
| 94 |
"""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
|
| 96 |
|
| 97 |
|
|
|
|
| 1 |
|
| 2 |
|
| 3 |
import gradio as gr
|
|
|
|
|
|
|
| 4 |
import random as rd
|
|
|
|
|
|
|
| 5 |
from huggingface_hub import InferenceClient
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
client = InferenceClient("Qwen/Qwen2.5-72B-Instruct")
|
|
|
|
|
|
|
| 7 |
def respond (message, history):
|
|
|
|
|
|
|
| 8 |
messages = [{"role": "system", "content": "You are a chatbot who is sassy and doesnt explain an anser unless someone asks twice."}]
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
if history:
|
| 10 |
+
messages.extend(history)
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
messages.append({"role": "user", "content": message})
|
|
|
|
|
|
|
| 12 |
response = ""
|
|
|
|
|
|
|
| 13 |
for message in client.chat_completion(
|
|
|
|
|
|
|
| 14 |
messages,
|
|
|
|
|
|
|
| 15 |
max_tokens=500,
|
|
|
|
|
|
|
| 16 |
temperature=0.2,
|
|
|
|
|
|
|
| 17 |
top_p=0.9,
|
|
|
|
|
|
|
| 18 |
stream=True
|
|
|
|
|
|
|
| 19 |
):
|
|
|
|
|
|
|
| 20 |
token = message.choices[0].delta.content
|
|
|
|
|
|
|
| 21 |
response += token
|
|
|
|
|
|
|
| 22 |
yield response
|
| 23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
# Define the Gradio interface
|
| 25 |
css = """
|
| 26 |
.gradio-container {
|
|
|
|
| 28 |
font-family: 'Times New Roman', Times, serif;
|
| 29 |
}
|
| 30 |
"""
|
| 31 |
+
chatbot = gr.ChatInterface(respond, type = "messages")
|
| 32 |
+
chatbot.launch(debug=True)
|
| 33 |
+
|
| 34 |
+
|
| 35 |
|
| 36 |
|
| 37 |
|