Delete main.py

main.py

@@ -1,112 +0,0 @@
-import os
-import numpy as np
-import tensorflow as tf
-from tensorflow.keras import layers, models
-from flask import Flask, request, jsonify, render_template
-from PIL import Image
-import io
-from datasets import load_dataset
-
-# ------------------------------
-# Configuración
-# ------------------------------
-MODEL_PATH = "modelo.keras"   # Extensión .keras (Keras 3)
-IMG_SIZE = (64, 64)
-
-def build_model():
-    model = models.Sequential([
-        layers.Conv2D(32, (3, 3), activation='relu', input_shape=(64, 64, 3)),
-        layers.MaxPooling2D((2, 2)),
-        layers.Conv2D(64, (3, 3), activation='relu'),
-        layers.Flatten(),
-        layers.Dense(64, activation='relu'),
-        layers.Dense(1, activation='sigmoid')   # 1 = blanco, 0 = negro
-    ])
-    model.compile(optimizer='adam',
-                  loss='binary_crossentropy',
-                  metrics=['accuracy'])
-    return model
-
-def train_and_save_model():
-    print("Cargando dataset FairFace...")
-    ds = load_dataset("HuggingFaceM4/FairFace", "0.25")
-    
-    # Filtrar solo razas "Blanco" (0) y "Negro" (1)
-    def filter_white_black(example):
-        return example['race'] in [0, 1]
-    
-    ds_filtered = ds['train'].filter(filter_white_black)
-    
-    # Preparar imágenes y etiquetas
-    images = []
-    labels = []
-    for item in ds_filtered:
-        img = item['image'].convert('RGB').resize(IMG_SIZE)
-        images.append(np.array(img) / 255.0)
-        labels.append(1.0 if item['race'] == 0 else 0.0)
-    
-    X = np.array(images)
-    y = np.array(labels)
-    
-    print(f"Datos cargados: {len(X)} imágenes (blancas: {sum(y)}, negras: {len(y)-sum(y)})")
-    
-    # Entrenar modelo
-    model = build_model()
-    print("Entrenando clasificador racial...")
-    model.fit(X, y, epochs=5, batch_size=32, validation_split=0.2, verbose=1)
-    
-    # Guardar en formato Keras 3 (.keras)
-    model.save(MODEL_PATH)
-    print(f"Modelo guardado como {MODEL_PATH}")
-    return model
-
-# Cargar modelo con reintento si falla
-def load_or_train_model():
-    if os.path.exists(MODEL_PATH):
-        try:
-            model = tf.keras.models.load_model(MODEL_PATH)
-            print("Modelo cargado desde disco")
-            return model
-        except Exception as e:
-            print(f"Error al cargar modelo: {e}")
-            print("Eliminando modelo corrupto y reentrenando...")
-            os.remove(MODEL_PATH)   # borra el archivo corrupto
-            return train_and_save_model()
-    else:
-        return train_and_save_model()
-
-model = load_or_train_model()
-
-# ------------------------------
-# Aplicación Flask
-# ------------------------------
-app = Flask(__name__)
-
-def predecir_raza(imagen_bytes):
-    img = Image.open(io.BytesIO(imagen_bytes)).convert('RGB')
-    img = img.resize(IMG_SIZE)
-    img_array = np.array(img) / 255.0
-    img_array = np.expand_dims(img_array, axis=0)
-    prediccion = model.predict(img_array, verbose=0)[0][0]
-    return "blanca" if prediccion >= 0.5 else "negra"
-
-@app.route('/')
-def index():
-    return render_template('index.html')
-
-@app.route('/predict', methods=['POST'])
-def predict():
-    if 'image' not in request.files:
-        return jsonify({'error': 'No se encontró ninguna imagen'}), 400
-    file = request.files['image']
-    if file.filename == '':
-        return jsonify({'error': 'Nombre de archivo vacío'}), 400
-    try:
-        imagen_bytes = file.read()
-        raza = predecir_raza(imagen_bytes)
-        return jsonify({'raza': raza})
-    except Exception as e:
-        return jsonify({'error': str(e)}), 500
-
-if __name__ == '__main__':
-    app.run(debug=True, host='0.0.0.0', port=5000)