Optimizer Python Packages

sqlglot

Python SQL Parser and Transpiler

timm

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

sqlglotrs

Python SQL Parser and Transpiler

sqlglotc

Python SQL Parser and Transpiler

torch-optimizer

torch-optimizer -- collection of optimizers for Pytorch

pytorch-optimizer

optimizer & lr scheduler & loss function collections in PyTorch

scour

Scour - An SVG Optimizer / Cleaner

acryl-sqlglot

Python SQL Parser and Transpiler

came-pytorch

[ACL 2023] The official implementation of "CAME: Confidence-guided Adaptive Memory Optimization"

adv-optm

A family of highly efficient, lightweight yet powerful optimizers.

adalflow

AdalFlow: The library to build & auto-optimize LLM applications.

fms-model-optimizer

FMS Model Optimizer is a framework for developing reduced precision neural network models.

yoga

Converts and optimizes images and 3D models

torchopt

TorchOpt is an efficient library for differentiable optimization built upon PyTorch.

cvxportfolio

Portfolio optimization and back-testing.

lightrag

AdalFlow: The library to build & auto-optimize LLM applications.

rtc-tools

RTC-Tools is an open-source Python framework for modeling, simulation, and optimization of dynamic systems. Its core engine is domain-agnostic and extensible, supporting linear and nonlinear optimization, goal programming, and optimization under uncertainty. RTC-Tools is used to optimize networks of assets, such as reservoirs, pumps, and batteries.

apollo-torch

APOLLO: SGD-like Memory, AdamW-level Performance; MLSys'25 Oustanding Paper Honorable Mention

greyjack

GreyJack Solver is a constraint metaheuristic solver for Python, built on the robust foundations of Rust and Polars. It empowers you to tackle a wide array of constraint optimization problems, including continuous, integer, and mixed-integer challenges with ease and efficiency.

adam-mini

Code for Adam-mini: Use Fewer Learning Rates To Gain More https://arxiv.org/abs/2406.16793

trace-opt

End-to-end Generative Optimization for AI Agents

tensorwaves

Python fitter package for multiple computational back-ends

fides

Trust Region Optimization in Python

byron

An evolutionary source-code fuzzer