Spaces:
Sleeping
Sleeping
import json | |
import gradio as gr | |
import random | |
from huggingface_hub import InferenceClient | |
API_URL = "https://api-inference.huggingface.co/models/" | |
client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.1") | |
def format_prompt(message, history): | |
prompt = "You're a helpful assistant." | |
for user_prompt, bot_response in history: | |
prompt += f" [INST] {user_prompt} [/INST] {bot_response}</s> " | |
prompt += f" [INST] {message} [/INST]" | |
return prompt | |
def generate(prompt, history, temperature=0.9, max_new_tokens=2048, top_p=0.95, repetition_penalty=1.0): | |
temperature = float(temperature) if temperature > 0 else 0.01 | |
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=random.randint(0, 10**7), | |
) | |
formatted_prompt = format_prompt(prompt, history) | |
stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False) | |
output = "" | |
for response in stream: | |
output += response.token.text | |
yield output | |
def load_database(): | |
try: | |
with open("database.json", "r", encoding="utf-8") as f: | |
return json.load(f) | |
except (FileNotFoundError, json.JSONDecodeError): | |
print("Error loading database: File not found or invalid format. Creating an empty database.") | |
return [] | |
def save_database(data): | |
try: | |
with open("database.json", "w", encoding="utf-8") as f: | |
json.dump(data, f, indent=4) | |
except (IOError, json.JSONEncodeError): | |
print("Error saving database: Encountered an issue while saving.") | |
def chat_interface(message): | |
database = load_database() | |
if (message, None) not in database: | |
response = next(generate(message, history=[])) | |
database.append((message, response)) | |
save_database(database) | |
else: | |
_, stored_response = next(item for item in database if item[0] == message) | |
response = stored_response | |
return response | |
with gr.Interface(fn=chat_interface, inputs="textbox", outputs="textbox", title="Chat Interface") as iface: | |
iface.launch() | |