7.1 Built-in modules

What are modules?

• an object that encapsulates state
• can be of any complexity (e.g. layer, or models consisting of layers)

Examples of {torch} modules:

• linear: nn_linear()
• convolutional: nn_conf1d(), nn_conf2d(), nn_conv_3d()
• recurrent: nn_lstm(), nn_gru()
• embedding: nn_embedding()
• multi-head attention: nn_multihead_attention()
• See torch documentation for others