{"id":31021354,"url":"https://github.com/sisl/crux.jl","last_synced_at":"2025-09-13T11:20:52.341Z","repository":{"id":45534605,"uuid":"310735135","full_name":"sisl/Crux.jl","owner":"sisl","description":"Julia library for deep reinforcement learning","archived":false,"fork":false,"pushed_at":"2025-09-11T02:32:45.000Z","size":13951,"stargazers_count":63,"open_issues_count":7,"forks_count":17,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-09-11T05:54:06.347Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Julia","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sisl.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2020-11-07T00:28:30.000Z","updated_at":"2025-09-11T02:32:48.000Z","dependencies_parsed_at":"2023-11-30T20:27:09.308Z","dependency_job_id":"fd3bf739-73c6-4390-94b2-a019b0817691","html_url":"https://github.com/sisl/Crux.jl","commit_stats":null,"previous_names":["sisl/crux.jl","ancorso/crux.jl"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/sisl/Crux.jl","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FCrux.jl","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FCrux.jl/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FCrux.jl/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FCrux.jl/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sisl","download_url":"https://codeload.github.com/sisl/Crux.jl/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FCrux.jl/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":274955742,"owners_count":25380669,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-13T02:00:10.085Z","response_time":70,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-09-13T11:20:46.600Z","updated_at":"2025-09-13T11:20:52.254Z","avatar_url":"https://github.com/sisl.png","language":"Julia","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Crux.jl\n\n[![Documentation](https://img.shields.io/badge/docs-latest-blue.svg)](https://sisl.github.io/Crux.jl/dev/)\n[![Build Status](https://github.com/sisl/Crux.jl/actions/workflows/CI.yml/badge.svg)](https://github.com/sisl/Crux.jl/actions/workflows/CI.yml)\n[![Code Coverage](https://codecov.io/gh/sisl/Crux.jl/branch/master/graph/badge.svg)](https://codecov.io/gh/sisl/Crux.jl)\n\nDeep RL library with concise implementations of popular algorithms. Implemented using [Flux.jl](https://github.com/FluxML/Flux.jl) and fits into the [POMDPs.jl](https://github.com/JuliaPOMDP/POMDPs.jl) interface.\n\nSupports CPU and GPU computation and implements deep reinforcement learning, imitation learning, batch RL, adversarial RL, and continual learning algorithms. See the [documentation]() for more details.\n\n### Reinforcement Learning\n* [Deep Q-Learning (DQN)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/dqn.jl)\n  * Prioritized Experience Replay\n* [Soft Q-Learning](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/softq.jl)\n* [REINFORCE](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/reinforce.jl)\n* [Proximal Policy Optimization (PPO)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/ppo.jl)\n* [Lagrange-Constrained PPO](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/ppo.jl)\n* [Advantage Actor Critic (A2C)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/a2c.jl)\n* [Deep Deterministic Policy Gradient (DDPG)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/ddpg.jl)\n* [Twin Delayed DDPG (TD3)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/td3.jl)\n* [Soft Actor Critic (SAC)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/rl/sac.jl)\n\n### Imitation Learning\n* [Behavioral Cloning](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/bc.jl)\n* [Generative Adversarial Imitation Learning (GAIL) w/ On-Policy and Off-Policy Versions](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/gail.jl)\n* [Adversarial Value Moment Imitation Learning (AdVIL)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/AdVIL.jl)\n* [Adversarial Reward-moment Imitation Learning (AdRIL)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/AdRIL.jl)\n* [Soft Q Imitation Learning (SQIL)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/sqil.jl)\n* [Adversarial Soft Advantage Fitting (ASAF)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/asaf.jl)\n* [Inverse Q-Learning (IQLearn)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/il/iqlearn.jl)\n\n### Batch RL\n* [Batch Soft Actor Critic (BatchSAC)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/batch/sac.jl)\n* [Conservative Q-Learning (CQL)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/batch/cql.jl)\n\n### Adversarial RL\n* [Robust Adversarial RL (RARL)](https://github.com/sisl/Crux.jl/blob/master/src/model_free/adversarial/rarl.jl)\n\n### Continual Learning\n* [Experience Replay](https://github.com/sisl/Crux.jl/blob/master/src/model_free/cl/experience_replay.jl)\n\n\n## Usage\n\nAn example usage of the `REINFORCE` algorithm with a simple Flux network for the Cart Pole problem is shown here:\n\n```julia\nusing Crux, POMDPGym\n\n# Problem setup\nmdp = GymPOMDP(:CartPole)\nas = actions(mdp)\nS = state_space(mdp)\n\n# Flux network: Map states to actions\nA() = DiscreteNetwork(Chain(Dense(dim(S)..., 64, relu), Dense(64, length(as))), as)\n\n# Setup REINFORCE solver\nsolver_reinforce = REINFORCE(S=S, π=A())\n\n# Solve the `mdp` to get the `policy`\npolicy_reinforce = solve(solver_reinforce, mdp)\n```\n\n\n## Installation\n\nTo install the package, run:\n```\n] add Crux\n```\n\nSee the [installation documentation]() for more details on how to install POMDPGym for more environment.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisl%2Fcrux.jl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsisl%2Fcrux.jl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisl%2Fcrux.jl/lists"}