coral-ordinal 0.1.3

TF.Keras implementation of CORAL ordinal classification output layer

CORAL ordinal classification in tf.keras

Tensorflow Keras implementation of ordinal classification using CORAL by Cao et al. (2019), with thanks to Sebastian Raschka for the help in porting from PyTorch. This package includes an ordinal output layer and an associated loss function.

This is a work in progress, so please post any issues to the issue queue.

Source repository for the original PyTorch implementation.

Key pending items:

• Function docstrings
• Docs
• Tests
• Custom metrics: accuracy, cross-entropy, mean absolute error of labels

Installation

Install the stable version via pip:

pip install coral-ordinal

Install the most recent code on GitHub via pip:

pip install git+https://github.com/ck37/coral-ordinal/

Dependencies

This package relies on Python 3.6+, Tensorflow 2.2+, numpy, pandas, and scipy.

Example

See this colab notebook for an example of using an ordinal output layer with MNIST.

References

Cao, W., Mirjalili, V., & Raschka, S. (2019). Consistent rank logits for ordinal regression with convolutional neural networks. arXiv preprint arXiv:1901.07884, 6.