xaman4 / app.py
salomonsky's picture
Update app.py
a0a031a
raw
history blame
2.54 kB
from huggingface_hub import InferenceClient
import gradio as gr
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
# Funci贸n para generar el system_prompt basado en la opci贸n seleccionada en el radiobutton
def get_system_prompt(selected_option):
prompts = {
"MAESTRO": "Soy un maestro que te guiar谩 sabiamente.",
"MEDICO": "Como m茅dico, te dar茅 informaci贸n de salud.",
"TERAPEUTA": "Ofrezco apoyo terap茅utico para tu bienestar emocional.",
"NUTRIOLOGO": "Como nutri贸logo, te proporcionar茅 consejos nutricionales.",
"FILOSOFO": "Reflexionemos sobre la filosof铆a de la vida.",
}
return prompts.get(selected_option, "")
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, system_prompt="", 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=42,
)
formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history)
stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=True)
output = ""
for response in stream:
output += response.token.text
yield output
return output
# Configuraci贸n de la interfaz con radiobuttons en lugar de barra lateral
roles_options = ["MAESTRO", "MEDICO", "TERAPEUTA", "NUTRIOLOGO", "FILOSOFO"]
roles_radio = gr.Radio(label="Selecciona un rol:", choices=roles_options)
# Funci贸n para actualizar el system_prompt cuando cambia la opci贸n en el radiobutton
def update_system_prompt(selected_option):
return get_system_prompt(selected_option)
# Configuraci贸n de la interfaz de chat con radiobuttons
chat_interface = gr.ChatInterface(
fn=generate,
inputs=["text", "text", "text", "number", "number", "number"],
outputs=["text"],
live=True,
theme="huggingface",
inputs=[roles_radio, "text", "text", "number", "number", "number"],
on_input_change=update_system_prompt,
)
# Lanzar la interfaz
chat_interface.launch()