Machine Learning Terms


By Andres Segura-Tinoco

Hyperparameters specify the model's architecture and how it will be trained. Therefore, these hyperparameters are not learned during training phase. In the case of neural networks, some common hyperparameters are:

  • Learning rate
  • Batch size
  • Number of epochs to train the model
  • Number of hidden layers
  • Number of neurons in each layer
  • Activations functions