Loss Functions Python Packages

pytorch-optimizer

optimizer & lr scheduler & loss function collections in PyTorch

auraloss

Collection of audio-focused loss functions in PyTorch

dists-pytorch

IQA: Deep Image Structure and Texture Similarity Metric

focal-loss

TensorFlow implementation of focal loss

torch-log-wmse

logWMSE, an audio quality metric & loss function with support for digital silence target. Useful for training and evaluating audio source separation systems.

torch-l1-snr

Variations of L1 SNR Loss function for training audio source separation machine learning models

pytorch-focalloss

PyTorch implementation of binary and multi-class focal loss functions

beta-divergence-metrics

PyTorch implementations of the beta divergence loss.

torch-log-wmse-audio-quality

logWMSE, an audio quality metric & loss function with support for digital silence target. Useful for training and evaluating audio source separation systems.

iqa-pytorch

Comparison of IQA models in Perceptual Optimization

balanced-loss

Easy to use class balanced cross entropy and focal loss implementation for Pytorch

pytorch-fairloss

PyTorch classification loss that extends cross-entropy with a differentiable argmax penalty, inspired by Penalized Logarithmic Loss (PLL)

bokbokbok

Custom Loss Functions and Evaluation Metrics for XGBoost and LightGBM

pylandscape

This project propose the loss landscape analysis as effective methodology to understand the robustness against natural perturbation of QNN.

gedml

Generalized Deep Metric Learning.

pietoolbelt

Toolbelt for PiePline training pipeline

object-condensation

Efficient implementation of object condensation losses for use in various projects

comport-ai

An HR predictive analytics tool for forecasting the likely range of a worker’s future job performance using multiple ANNs with custom loss functions.

losshub

LossHub: Loss Functions Library for Image Classification and Detection

treeboost-autograd

Easy Custom Losses for Tree Boosters using Pytorch

mtl-uncertainty-loss

PyTorch unofficial implementation for uncertainty-weighted multi-task loss (Kendall et al., 2018)

toploss

Topological Loss Engineering: differentiable, optimizer-free regularizers that embed 2024-2026 optimizer breakthroughs (Muon/XSAM/CWD/AdEMAMix/NTKMTL/SymNoise) directly into the loss.

distortion-residual

Differentiable Distortion Residual Level (DRL) metric for audio processing optimization

gluonfr

Gluon FR Toolkit