Is this the correct way to implement k-fold cross-validation in training a deep learning model?

Question

I understand the concept of k-fold cross validation, and is trying to apply it here. Is the following pseudocode correct?

encoder = Encoder.load_model(enc_fil