Python: Advanced Guide to Artificial Intelligence

Keras-based MLP for MNIST classification

Now let's build the same MLP network with Keras, a high-level library for TensorFlow. We keep all the parameters the same as we used for the TensorFlow example in this chapter, for example, the activation function for the hidden layers is kept as the ReLU function.

1. Import the required modules from the Keras:

import kerasfrom keras.models import Sequentialfrom keras.layers import Densefrom keras.optimizers import SGD

2. Define the hyper-parameters (we assume that the dataset has already been loaded into the X_train, Y_train, X_test, and Y_test variables):

num_layers = 2num_neurons = []for i in range(num_layers):   num_neurons.append(256)learning_rate = 0.01n_epochs = 50batch_size = 100

3. Create a sequential ...