|
import os |
|
import urllib |
|
import gradio as gr |
|
|
|
import requests |
|
|
|
endpoint_url = os.getenv('ENDPOINT_URL') |
|
personal_secret_token = os.getenv('PERSONAL_HF_TOKEN') |
|
|
|
turn_breaker = os.getenv('TURN_BREAKER') |
|
system_symbol = os.getenv('SYSTEM_SYMBOL') |
|
user_symbol = os.getenv('USER_SYMBOL') |
|
assistant_symbol = os.getenv('ASSISTANT_SYMBOL') |
|
|
|
headers = { |
|
"Accept" : "application/json", |
|
"Authorization": f"Bearer {personal_secret_token}", |
|
"Content-Type": "application/json" |
|
} |
|
|
|
def query(payload): |
|
|
|
|
|
return payload['inputs'] |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def respond( |
|
message, |
|
history: list[tuple[str, str]], |
|
system_message, |
|
max_new_tokens, |
|
temperature, |
|
top_p, |
|
): |
|
|
|
|
|
all_messages = [system_message] |
|
|
|
for val in history: |
|
if val[0]: |
|
|
|
all_messages.append(user_symbol+val[0]) |
|
|
|
if val[1]: |
|
all_messages.append(assistant_symbol+val[1]) |
|
|
|
|
|
|
|
all_messages.append(user_symbol+message) |
|
|
|
generation_kwargs = dict( |
|
max_new_tokens=max_new_tokens, |
|
do_sample=temperature > 0, |
|
top_p=top_p, |
|
temperature=temperature, |
|
|
|
) |
|
|
|
response = query({ |
|
"inputs": turn_breaker.join(all_messages), |
|
"parameters": generation_kwargs |
|
}) |
|
|
|
return response |
|
|
|
|
|
""" |
|
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface |
|
""" |
|
demo = gr.ChatInterface( |
|
respond, |
|
additional_inputs=[ |
|
gr.Textbox(value="请你现在扮演一个软件工程师,名字叫做贺英旭。你需要以这个身份和朋友们对话。", label="System message"), |
|
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"), |
|
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"), |
|
gr.Slider( |
|
minimum=0.1, |
|
maximum=1.0, |
|
value=0.95, |
|
step=0.05, |
|
label="Top-p (nucleus sampling)", |
|
), |
|
], |
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
demo.launch() |