Hands-On Deep Learning with R

Generating random images

After training, we will use the model to create random images, as follows:

1. The next step is to create a matrix of random variables. The number of random variables should be set at the batch size times the size of the input shape for our generator model. The dimensions are then set so that the number of rows is equal to the batch size and the number of columns is equal to the length of the input shape defined here. In this case, 20 is used as the batch size and 100 is used as the length of the vector to be passed to the generator model. Both of these values are modifiable. Increasing either or both provides more data to the model, which could improve performance but will also increase runtime. We create our matrix ...