app.py

import os
import pandas as pd
import torch
from PIL import Image
from ultralytics import YOLO
import gradio as gr

class YOLODetect():
    def __init__(self, modelo):
        self.modelo = modelo

    def predecir(self, source, imgsz=640, conf=0.7, iou=0.50):
        self.results = self.modelo.predict(source=source, save=True, imgsz=imgsz, conf=conf, iou=iou)
        return self.results

    def render(self):
        result = self.results[0]
        file_name = os.path.join(result.save_dir, result.path)
        render = Image.open(file_name)
        return render

# Inicializa el modelo YOLOv8
path_best_model = 'yolov8n-pose.pt'
modelo_yolo = YOLO(path_best_model)

def detect_objects(im, size, iou, conf):
    '''Wrapper para Gradio'''
    g = (int(size) / max(im.size))  # gain
    im = im.resize(tuple([int(x * g) for x in im.size]), Image.LANCZOS) # resize with antialiasing

    model = YOLODetect(modelo_yolo)
    results = model.predecir(source=im, imgsz=int(size), conf=conf, iou=iou)

    objects_detected = results[0].boxes.cls.tolist()  # Clases detectadas.
    objects_conf = results[0].boxes.conf.tolist()  # Probabilidad de detección por clase detectada.

    objects_nested_list = pd.DataFrame({'Clase': objects_detected, 'Probabilidad': objects_conf})

    result_img = model.render()
    return result_img, objects_nested_list

def save_feedback(size, iou, conf, 
                object_count_detected,
                objects_list,
                user_text, feedback_text, check_status):
    try:
        # Aquí puede ir el código para almacenar los datos en una base de datos.
        return "Se guardó el feedback exitosamente."
    except Exception as err:
        print(err)
        return "Error al guardar el feedback."

# Configura la interfaz de Gradio
with gr.Blocks() as demo:
    gr.Markdown("# YOLOv8 Detección de postura")

    with gr.Row():
        iou_threshold = gr.Slider(label="NMS IoU Threshold (0.0 - 1.0)", minimum=0.0, maximum=1.0, value=0.8)
        conf_threshold = gr.Slider(label="Umbral o threshold (0.0 - 1.0)", minimum=0.0, maximum=1.0, value=0.9)
        size = gr.Dropdown(label="Tamaño de la imagen", choices=["640", "1280"], value="640")

    with gr.Row():
        uploaded_image = gr.Image(type="pil", label="Cargar imagen")
        result_image = gr.Image(label="Resultado")
    
    detect_btn = gr.Button("Detectar postura")
    
    objects_table = gr.DataFrame(label="Objetos detectados")
    feedback = gr.Textbox(label="Ingrese su feedback (máximo 100 caracteres)", max_lines=2)
    user_text = gr.Textbox(label="Ingrese el nombre del usuario (máximo 50 caracteres)", max_lines=2)
    check_status = gr.Checkbox(label="¿El resultado contiene la cantidad correcta de figuras detectadas?", value=False)
    save_btn = gr.Button("Guardar feedback")
    output_message = gr.Textbox(label="Mensaje de salida", interactive=False)

    detect_btn.click(fn=detect_objects, 
                     inputs=[uploaded_image, size, iou_threshold, conf_threshold], 
                     outputs=[result_image, objects_table])

    save_btn.click(fn=save_feedback, 
                   inputs=[size, iou_threshold, conf_threshold, 
                           objects_table, user_text, feedback, check_status], 
                   outputs=[output_message])

# Ejecuta la interfaz de Gradio
if __name__ == "__main__":
    demo.launch()