Java Deep Learning Projects

Network training and evaluation

Now that training and test sets are prepared, we can start the training. However, before that, we define some hyperparameters for the dataset preparation:

private static final int EPOCH = 100;private static final int BATCH_SIZE = 128;private static final int SAVING_INTERVAL = 100;

Additionally, we specify the path where the trained model will be saved for future reuse:

private static final String SAVING_PATH = "bin/CatvsDog_VG16_TrainedModel_Epoch100_v1.zip";

Now we can start training the network. We will do the training combined such that training is carried out with the training set and validation is effected by using the validation set. Finally, the network will evaluate the network performance using the ...