Neural Network Projects with Python

Model compilation

Before we start training our model, we need to define the parameters of the training process, which is done via the compile method.

There are three different parameters we need to define for the training process:

• Optimizer: Let's use the adam optimizer, which is a popular optimizer in Keras. For most datasets, the adam optimizer will work well without much tuning.
• Loss function: We'll use binary_crossentropy as our loss function since the problem at hand is a binary classification problem.
• Metrics: We'll use accuracy (that is, the percentage of correctly classified samples) as our evaluation metric.

Then, we can run the compile() function as follows:

# Compile the modelmodel.compile(optimizer='adam', loss='binary_crossentropy', ...