import tensorflow as tf
new_model = tf.keras.models.load_model('saved_model/my_model_KNOCK_2_RMS')
new_model.get_weights()
How can I retrieve the Bias matrix as I retrieve the weights? Or is there a different way to get the bias matrix?
model.get_weights()
returns all the variables of your network, biases included.
You can iterate over the variables
attribute and filter on the name of the tf.Variable
to get only the biases.
biases = [var for var in new_model.variables if "bias" in var.name]