""" Source: https://github.com/Crispy13/crispy """
import tensorflow as tf
from tensorflow import keras
[docs]
def build_guided_model(model):
"""
Builds guided model
"""
model_copied = keras.models.clone_model(model)
model_copied.set_weights(model.get_weights())
@tf.custom_gradient
def guidedRelu(x):
def grad(dy):
return tf.cast(dy > 0, tf.float32) * tf.cast(x > 0, tf.float32) * dy
return tf.nn.relu(x), grad
layer_dict = [layer for layer in model_copied.layers[1:] if hasattr(layer, 'activation')]
for layer in layer_dict:
if layer.activation == tf.keras.activations.relu:
layer.activation = guidedRelu
return model_copied
[docs]
def guided_backprop_on_guided_model(model_no_softmax, img, layer_name):
"""
Returns guided backpropagation image.
Parameters
----------
model_no_softmax : a keras model object
img : an img to inspect with guided backprop.
layer_name : a string
a layer name for calculating gradients.
"""
guided_model = build_guided_model(model_no_softmax)
part_model = keras.Model(guided_model.inputs, guided_model.get_layer(layer_name).output)
with tf.GradientTape() as tape:
f32_img = tf.cast(img, tf.float32)
tape.watch(f32_img)
part_output = part_model(f32_img)
grads = tape.gradient(part_output, f32_img)[0].numpy()
# delete copied model
del part_model
del guided_model
return grads