Hyperparameter Tuning Python Packages

wandb

The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.

skypilot

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

scikit-optimize

Sequential model-based optimization with a `scipy.optimize` interface

wandb-osh

A convenient way to trigger synchronizations to wandb / Weights & Biases if your compute nodes don't have internet!

skypilot-nightly

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

arm-mango

Parallel Hyperparameter Tuning in Python

nni

An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.

smac

SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization

hypertune

A library for performing hyperparameter optimization

google-vizier

Python-based research interface for blackbox and hyperparameter optimization, based on the internal Google Vizier Service.

evalml

EvalML is an AutoML library written in python.

tune-sklearn

A drop-in replacement for Scikit-Learn’s GridSearchCV / RandomizedSearchCV -- but with cutting edge hyperparameter tuning techniques.

kubeflow-katib

Automated Machine Learning on Kubernetes

gamma-facet

Human-explainable AI.

kubeflow-katib-api

Automated Machine Learning on Kubernetes

pyruns

A lightweight Python CLI for batch experiments: auto-generate tasks from configs, run in parallel, and track results with real-time monitoring.

tune

An abstraction layer for parameter tuning

lale

Library for Semi-Automated Data Science

layer

Metadata store for Production ML

polystores

A library for performing hyperparameter optimization

octis

OCTIS: Comparing Topic Models is Simple! A python package to optimize and evaluate topic models (accepted at EACL2021 demo track)

omega-lock

Catch overfitting before you ship it: a walk-forward / held-out gate for any optimizer (Optuna, sweeps, prompt search). The highest score isn't deployable until it survives data it never saw. CLI + Python.

syne-tune

Large scale and asynchronous Hyperparameter and Architecture Optimization at your fingertips.

rl-zoo3

A training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.