This subpackage contains classes that implement different types of blocks, layers, and entire models
that can be used to tackle different tasks. Most models share the same base interface
that is derived from the PyTorch-Lightning base interface (pl.LightningModule), meaning they are
meant to be used as part of PyTorch-Lightning training/prediction experiments, which are launched
using the hardpicks/main.py script.
Models are automatically instantiated and used as part of our training and evaluation pipelines, but they can also be created and used manually. See the this module for more information.
Like for the data subpackage, task-specific details are split into separate
folders; see fbp for first break picking.