Replace SequentialLayer with Sequential Model
The SequentialLayer
was introducing more problems than the ones it solved.
Unfortunately, the old checkpoints will not be compatible but they can be easily converted using:
old_model = tf.keras.models.load_model(".../old_checkpoint")
new_model = bob.learn.tensorflow.models.incepetion_resenet_v2.IncepetionResnetV2()
new_model.set_weights(old_model.get_weights)
new_model.save(".../new_checkpoint")