Untitled3.ipynb

untitled3.ipynb

 def build_model(self, learning_rate=0.001):
        """Build the model architecture."""
        base_model = tf.keras.applications.MobileNetV2(
            weights=self.base_model_weights,
            include_top=False,
            input_shape=self.input_shape
        )
        base_model.trainable = False

        # Create feature extraction branches
        branch1 = tf.keras.layers.GlobalAveragePooling2D()(base_model.output)
        branch2 = tf.keras.layers.GlobalMaxPooling2D()(base_model.output)

        # Merge branches
        merged = tf.keras.layers.Concatenate()([branch1, branch2])
        
        # Add dropout for regularization
        x = tf.keras.layers.Dropout(0.5)(merged)
        
        # Add dense layers with L2 regularization
        x = tf.keras.layers.Dense(
            512, 
            activation='relu',
            kernel_regularizer=tf.keras.regularizers.l2(0.01)
        )(x)
        x = tf.keras.layers.Dropout(0.3)(x)
        
        output = tf.keras.layers.Dense(1, activation='sigmoid')(x)

        self.model = tf.keras.Model(inputs=base_model.input, outputs=output)
        
        # Compile with mixed precision for faster training
        tf.keras.mixed_precision.set_global_policy('mixed_float16')
        optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
        
        self.model.compile(
            optimizer=optimizer,
            loss='binary_crossentropy',
            metrics=['accuracy', tf.keras.metrics.AUC(), tf.keras.metrics.Precision(), tf.keras.metrics.Recall()]
        )