|
from huggingface_hub import InferenceAPI |
|
from urllib.parse import urlparse, parse_qs |
|
import gradio as gr |
|
|
|
api = InferenceAPI("mistralai/Mistral-7B-Instruct-v0.1") |
|
|
|
def format_prompt(message, history): |
|
prompt = "<s>" |
|
for user_prompt, bot_response in history: |
|
prompt += f"[INST] {user_prompt} [/INST]" |
|
prompt += f" {bot_response}</s> " |
|
prompt += f"[INST] {message} [/INST]" |
|
return prompt |
|
|
|
def generate(prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0): |
|
temperature = float(temperature) |
|
if temperature < 1e-2: |
|
temperature = 1e-2 |
|
top_p = float(top_p) |
|
|
|
generate_kwargs = dict( |
|
temperature=temperature, |
|
max_new_tokens=max_new_tokens, |
|
top_p=top_p, |
|
repetition_penalty=repetition_penalty, |
|
do_sample=True, |
|
seed=42, |
|
) |
|
|
|
formatted_prompt = format_prompt(prompt, history) |
|
response = api.generate_text(formatted_prompt, **generate_kwargs) |
|
return response |
|
|
|
def chatbot(message): |
|
history = [] |
|
response = generate(message, history) |
|
history.append(("User", message)) |
|
history.append(("ChatBot", response)) |
|
return response |
|
|
|
def get_message_from_url(): |
|
url = urlparse(gradio.Interface.get_share_url()) |
|
query_params = parse_qs(url.query) |
|
if "message" in query_params: |
|
return query_params["message"][0] |
|
return "" |
|
|
|
message = get_message_from_url() |
|
inputs = gr.inputs.Textbox(lines=2, placeholder="Type your message here...", initial_message=message) |
|
outputs = gr.outputs.Textbox() |
|
|
|
title = "Mistral 7B Chatbot" |
|
description = "Chat with Mistral 7B, a powerful language model!" |
|
|
|
gr.Interface(fn=chatbot, inputs=inputs, outputs=outputs, title=title, description=description).launch() |