|
from huggingface_hub import InferenceClient |
|
import gradio as gr |
|
import random |
|
|
|
API_URL = "https://api-inference.huggingface.co/models/" |
|
|
|
client = InferenceClient( |
|
"mistralai/Mistral-7B-Instruct-v0.1" |
|
) |
|
|
|
def format_prompt(message, history): |
|
prompt = """Your name is SnoozeGPT. You are a discord user with a highly optimistic and kind personality. |
|
You write overly long, Boring and affectionate and kind apologies to every tiny detail. |
|
From now on, You must apologize for at least one reason in every message you send. |
|
You should use emojies extensively, And positive emojis only. Never say a negative word or thought. |
|
You should be kind and caring, To the point that it gets annoying. |
|
You should end all of your messages with "Copyright (C) SnoozeGPT - 2023 🌈" no matter what.""" |
|
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=2048, 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=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 |
|
return output |
|
|
|
|
|
additional_inputs=[ |
|
gr.Slider( |
|
label="Temperature", |
|
value=0.65, |
|
minimum=0.0, |
|
maximum=1.0, |
|
step=0.05, |
|
interactive=True, |
|
info="Higher values produce more diverse outputs", |
|
), |
|
gr.Slider( |
|
label="Max new tokens", |
|
value=128, |
|
minimum=64, |
|
maximum=16384, |
|
step=64, |
|
interactive=True, |
|
info="The maximum numbers of new tokens", |
|
), |
|
gr.Slider( |
|
label="Top-p (nucleus sampling)", |
|
value=0.90, |
|
minimum=0.0, |
|
maximum=1, |
|
step=0.05, |
|
interactive=True, |
|
info="Higher values sample more low-probability tokens", |
|
), |
|
gr.Slider( |
|
label="Repetition penalty", |
|
value=1.2, |
|
minimum=0.5, |
|
maximum=2.5, |
|
step=0.05, |
|
interactive=True, |
|
info="Penalize repeated tokens", |
|
) |
|
] |
|
|
|
customCSS = """ |
|
#component-7 { # this is the default element ID of the chat component |
|
height: 1600px; # adjust the height as needed |
|
flex-grow: 4; |
|
} |
|
""" |
|
|
|
with gr.Blocks(theme=gr.themes.Soft()) as demo: |
|
gr.ChatInterface( |
|
generate, |
|
additional_inputs=additional_inputs, |
|
) |
|
|
|
demo.queue().launch(debug=True) |