Checkpointing model

Training a deep neural network is an expensive process in terms of time, storage, and resources. Retraining a network each time we want to use it is preposterous and impractical. The good news is that we can use a mechanism to automatically save the best versions of a network during the training process.

In this recipe, we'll talk about such a mechanism, known as checkpointing.

How to do it…

Follow these steps to learn about the different modalities of checkpointing you have at your disposal in TensorFlow:

1. Import the modules we will be using:

   import numpy as np
   import tensorflow as tf
   from sklearn.model_selection import train_test_split
   from sklearn.preprocessing import LabelBinarizer
   from tensorflow.keras.callbacks import ModelCheckpoint
   from tensorflow.keras.datasets import fashion_mnist as fm
   from tensorflow.keras.layers import *
   from tensorflow.keras.models import *

2. Define a function that will load Fashion-MNIST into tf.data...