DIRESA - distance-regularized Siamese twin autoencoder

DIRESA is a Python package for dimension reduction based on TensorFlow. The distance-regularized Siamese twin autoencoder architecture is designed to preserve distance (ordering) in latent space while capturing the non-linearities in the datasets.

Introduction

• DIRESA architecture

• Install DIRESA

Tutorial

• Start tutorial

• DIRESA model

• Evaluate

• Convolutional example

• Custom encoder/decoder

Module reference

• models module

• loss module

• callback module

• toolbox module