Optuna Python Packages

cmaes

Python library for CMA Evolution Strategy.

optuna-dashboard

Real-time Web Dashboard for Optuna.

class-resolver

🔍 Lookup classes and instantiate them with style

neptune

📘 The experiment tracker for foundation model training

neptune-client

📘 The experiment tracker for foundation model training

irspack

Train, evaluate, and optimize implicit feedback-based recommender systems.

martin-binance

Free trading system for crypto exchanges SPOT market. Adaptive customizable reverse grid strategy based on martingale.

jupyterlab-optuna

Real-time Web Dashboard for Optuna.

vllm-optimizer

Benchmark-driven configuration optimization for vLLM inference servers

aiaccel

AIST Toolkit for Accelerating Machine Learning Research

abliterix

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.

quoptuna

Fairness-aware, explainable AutoML for quantum + classical ML — 21 models, one Optuna search, SHAP & AI reports. uvx quoptuna

cobra-rfic

AI-assisted RFIC optimization framework combining ONNX surrogate models, Qucs-S/Xyce circuit simulation, and AWS Palace fine-tuning for open-source electronic design automation (EDA).

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.

hyperparameter-tuning

Hyperparameter tuning with minimal boilerplate

certified-pruner

Stop bad runs early, with a conformal bound on how often you stop a good one

rf-plateau-hpo

How many trees in a Random Forest? Adaptive n_estimators tuning via plateau search + Optuna — code for the IEEE Access 2026 paper

quantjourney-bt

QuantJourney Backtester

ablate-llm

🔓 Ablate — directional ablation (abliteration) toolkit for open-source LLMs. Automatic censorship/refusal removal via residual-stream direction ablation, with KL-guided search, an LLM-judge harness, and one-call push to the Hub.

mltb2

Machine Learning Toolbox 2

biomimetic-lattice-pipeline

From synchrotron micro-CT of tooth enamel to biomimetic lattices: parametric CAD, finite-element analysis, and closed-loop design optimization.

vtune

Benchmark-driven configuration optimization for vLLM inference servers

surestop

Stop bad runs early, with a conformal bound on how often you stop a good one

tuneta

Intelligently optimizes technical indicators and optionally selects the least intercorrelated for use in machine learning models