Decision Making Python Packages

torchrl

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

1.3M 3K 450

emukit

A Python-based toolbox of various methods in decision making, uncertainty quantification and statistical emulation: multi-fidelity, experimental design, Bayesian optimisation, Bayesian quadrature, etc.

51K 655 132

torchrl-nightly

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

37K 3K 450

scikit-decide

AI framework for Reinforcement Learning, Automated Planning and Scheduling

4K 194 30

flatland-rl

The Flatland Framework is a multi-purpose environment to tackle problems around resilient resource allocation under uncertainty. It is designed to be a flexible and method agnostic to solve a wide range of problems in the field of operations research and reinforcement learning.

3K 63 18

magi-core

A council of LLMs for hard decisions, inspired by the MAGI of Neon Genesis Evangelion. Deliberate, and synthesise — blind peer review.

2K 1 0

abtree

High-performance asynchronous behavior tree framework with behavior forest collaboration,

969 2 0

arbitration-graphs

Hierarchical behavior models for complex decision-making and behavior generation in robotics

940 30 1

false-feet-algorithm

False Feet Algorithm: Multi-branch hypothesis evaluation with inverse resource allocation. Solves complex problems through adaptive reasoning—includes mystery solver demo + generic framework for any domain.

820 0 0

insight137-eap

Entropy Attunement Protocol - 4D entropy profiling for behavioral prediction and AI safety

767 1 0

znum

Z number for fuzzy calculations

679 0 0

ns-gym

A framework for modeling non-stationary Markov decision processes and the key decision making problems in these environments

610 11 2

scikit-psl

Probabilistic Scoring List classifier

580 18 5

spice-runtime

Spice core runtime framework

477 130 7

maze-rl

Maze Applied Reinforcement Learning Framework

474 288 12

eospython

EOS is a set of functions (intended to be a package) that encompases everything within a centralized Earth Observation Satellite scheduling system in terms of Scenario generation, problem setup, solution approach, decision maker preference integration.

471 9 2