Regularization Python Packages

dropblock

Implementation of DropBlock: A regularization method for convolutional networks in PyTorch.

neural-structured-learning

Training neural models with structured signals.

networks

Library which can be used to build feed forward NN, Convolutional Nets, Linear Regression, and Logistic Regression Models.

l1tv

Denoising/reconstruction of piecewise constant signals using the L1TV model

pyclsp

A modular two-step convex optimization estimator for ill-posed problems

early-stopping-pytorch

Early stopping for PyTorch

tomobar

TOmographic MOdel-BAsed Reconstruction (ToMoBAR) software

pytmpinv

Tabular Matrix Problems via Pseudoinverse Estimation

ear-transformers

Code associated with the paper "Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists"

scproject

Transfer Learning of Gene Expression Signatures in Python.

pylppinv

Linear Programming via Regularized Least Squares

mlektic

A Python ML library to use and visualize gradient descent for linear and logistic regression optimization.

pytmpinvi

Interactive Tabular Matrix Problems via Pseudoinverse Estimation

pyunfold

Iterative unfolding for Python

egora

EgoRA — Entropy-Governed Orthogonality Regularization for Adaptation. Dynamic information-theoretic regularization for fine-tuning neural networks.

mumfordshah2d

Algorithms for edge preserving smoothing based on the Mumford-Shah model

condat-tv

Python wrapper for the fast TV denoising algorithm by Laurent Condat

l0learn

Efficient Algorithms for L0 Regularized Learning

torch-layer-normalization

Layer normalization in PyTorch

keras-targeted-dropout

Targeted dropout implemented in Keras

egora-diagnostics

Rotation-Retention Law analysis, knowledge mapping, and gradient alignment diagnostics for fine-tuned neural networks. Companion to the egora package.

egora-rrl

Rotation-Retention Law: threshold analysis, knowledge mapping, and gradient alignment diagnostics for fine-tuned neural networks.

lppinv

Linear Programming via Regularized Least Squares

stochdepth

Deep Networks with Stochastic Depth for PyTorch