oracle

ORACLE is a general framework for hierarchical classification.

Modules

architectures	Top-level module for defining various neural network architectures for hierarchical classification.
constants	Module containing constant mappings and configurations for the ORACLE project.
custom_datasets	Custom datasets for the ORACLE framework.
loss	Top level module for defining the Weighted Hierarchical Cross Entropy Loss function for hierarchical classification tasks.
presets	Top-level module for defining presets and utility functions for model selection, data loading, and training configurations in the ORACLE project.
pretrained	Module for working with pretrained models in the ORACLE framework.
taxonomies	Top level module for defining taxonomies used in hierarchical classification tasks.
test	Interface for testing saved models in the ORACLE framework.
tester	Module for testing hierarchical models in the ORACLE framework.
train	Interface for training models in the ORACLE framework.
trainer	Module for training hierarchical models in the ORACLE framework.
utils	Utility scripts for the ORACLE framework.
visualization	Module for visualization functions in the ORACLE framework.