Deep Learning with Keras

Understanding the power of deep learning

Another test that we can run to better understand the power of deep learning and ConvNet is to reduce the size of the training set and observe the consequent decay in performance. One way to do this is to split the training set of 50,000 examples into two different sets:

• The proper training set used for training our model will progressively reduce its size of (5,900, 3,000, 1,800, 600, and 300) examples
• The validation set used to estimate how well our model has been trained will consist of the remaining examples

Our test set is always fixed and it consists of 10,000 examples.

With this setup, we compare the just-defined deep learning ConvNet against the first example of neural network defined in ...