Model Fitting

Once a model has been initialized and layers have been added to the ANN, the model must be configured with an optimizer, losses, and any evaluation metrics through the compilation process. A model can be compiled using the model's compile method, as follows:

model.compile(optimizer='adam', loss='binary_crossentropy', \
              metrics=['accuracy'])

Optimizers can be chosen by simply naming the optimizer as the argument. The following optimizers are available as default for Keras models:

• Stochastic gradient descent (SGD): This updates the weights for each example in the dataset. You can find more information about SGD here: https://keras.io/api/optimizers/sgd/.
• RMSprop: This is an adaptive optimizer that varies the weights during training by using a decaying average of the gradients at each update. You can find more information about RMSprop...