Between Chapter 5, Convolutional Neural Networks, and Chapter 7, Training Neural Networks with Spark, a full example was presented regarding a CNN model's configuration and training. This was an example of image classification. The training data that was used came from the MNIST database. The training set contained 60,000 examples of handwritten digits, with each image labeled by an integer. Let's use the same example to show the visual facilities that are provided by DL4J for monitoring and debugging a network at training time.
At the end of training, you can programmatically save the generated model as a ZIP archive and throw the writeModel method of the ModelSerializer class (https://static.javadoc.io/org.deeplearning4j/deeplearning4j-nn/0.9.1/org/deeplearning4j/util/ModelSerializer.html):
ModelSerializer...
