morl-baselines - MORL Baselines Release v1.2.0

What's Changed

• Doc/update version by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/127
• Remove Sparsity from metrics logging by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/124
• Fix retrieval of observation and action spaces when using wrappers by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/129
• MORecordEpisodeStatistics-logging-fix by @1Kraks in https://github.com/LucasAlegre/morl-baselines/pull/142
• Refactor PCN save method by @1Kraks in https://github.com/LucasAlegre/morl-baselines/pull/143
• Chore/update precommit by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/147
• Add MOSAC Discrete, weights saving+loading for MORL-D, net_arch specification for MOSAC by @JaydenTeoh in https://github.com/LucasAlegre/morl-baselines/pull/134
• Add PGMORL 3D and include sparsity coefficient from paper by @JaydenTeoh in https://github.com/LucasAlegre/morl-baselines/pull/135
• Speed up tensor creation in pcn.py by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/152
• Add missing load function to pcn.py by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/150
• Add proper termination functions for GPI-PD LunarLander and Humanoid by @JaydenTeoh in https://github.com/LucasAlegre/morl-baselines/pull/136
• Fix device mismatch bug by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/154
• Fix PCN documentation by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/155
• Fix sorting individuals in performance buffer pgmorl by @Henweiz in https://github.com/LucasAlegre/morl-baselines/pull/156
• Fix CAPQL documentation by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/157
• Use weights_only=False in torch.load by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/158
• Provide helpful error message when pycddlib is not installed by @nkirschi in https://github.com/LucasAlegre/morl-baselines/pull/162
• Update Gymnasium's FrameStackObservation and GrayscaleObservation imports by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/163
• Allow changing experiment name for GPI-LS by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/164
• Update morldlunarlander_restore.py by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/165
• Release Version 1.2.0 by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/166

New Contributors

• @1Kraks made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/142
• @JaydenTeoh made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/134
• @nkirschi made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/152
• @Henweiz made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/156

Full Changelog: https://github.com/LucasAlegre/morl-baselines/compare/v1.1.0...v1.2.0

- Python
Published by LucasAlegre 7 months ago

morl-baselines - MORL Baselines Release v1.1.0: Gymnasium 1.0, MORL/D, HPO, bug fixes & improvements

What's Changed

• Move util functions to make more sense by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/62
• Implement fast pareto and convex hull pruning by @wilrop in https://github.com/LucasAlegre/morl-baselines/pull/60
• Support Gymnasium 0.29 by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/64
• Add minecart-deterministic-v0 to envs with known pfs by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/65
• Support for image-based observations in GPI-LS and Envelope by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/63
• Implement a fix to #69 by @wilrop in https://github.com/LucasAlegre/morl-baselines/pull/70
• Feature/hpo by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/74
• Use deterministic policies when evaluating PCN by @vaidas-sl in https://github.com/LucasAlegre/morl-baselines/pull/75
• Add bibtex for NeurIPS paper by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/77
• Update README.md by @eltociear in https://github.com/LucasAlegre/morl-baselines/pull/80
• Log all training parameters given to the algorithm by @wilrop in https://github.com/LucasAlegre/morl-baselines/pull/83
• Add support for continuous action spaces to PCN by @vaidas-sl in https://github.com/LucasAlegre/morl-baselines/pull/82
• Fix bug where tolist was called on a float by @wilrop in https://github.com/LucasAlegre/morl-baselines/pull/86
• Adding cardinality metric by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/87
• Remove dropout at evaluation time on GPI-LS by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/90
• MORL/D by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/89
• Chore/remove tensorboard stuff make gym logging optional by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/94
• Make number of sampled weights used to compute utility metrics parameterizable by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/95
• Dev eupg by @omidsbhn in https://github.com/LucasAlegre/morl-baselines/pull/99
• Fix PGMORL example by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/104
• Bump actions/download-artifact from 2 to 4.1.7 in /.github/workflows by @dependabot in https://github.com/LucasAlegre/morl-baselines/pull/116
• Fix EUPG's inconsistent expansion of observations by @timondesch in https://github.com/LucasAlegre/morl-baselines/pull/119
• Migration to gymnasium 1.0 by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/109
• Add save_freq option to CAPQL train method by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/122
• In GPI-LS for continuous action, use GPI only for selecting weights as default by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/126
• Add python 3.12 support by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/125

New Contributors

• @vaidas-sl made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/75
• @eltociear made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/80
• @omidsbhn made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/99
• @dependabot made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/116
• @timondesch made their first contribution in https://github.com/LucasAlegre/morl-baselines/pull/119

Full Changelog: https://github.com/LucasAlegre/morl-baselines/compare/1.0.0...v1.1.0

- Python
Published by ffelten over 1 year ago

morl-baselines - MORL-Baselines 1.0.0

This release marks the first stable version of MORL-Baselines. After having thoroughly tested the algorithms on various environments fixing bugs for the past few weeks. We feel the library is stable enough to deserve a proper release.

Features

• Over 10 MORL algorithms supported under the MO-Gymnasium API (multi & single policy, under SER and ESR criteria);
• Automated reporting to Weights and Biases dashboards... of various metrics (see screenshot below);
• Clean, documented, and tested code, and this is enforced by our CI hooks;
• Utility functions to help researchers build new algorithms, e.g. ParetoArchive, NatureCNN, PrioritizedReplayBuffer;
• Performances have been tested and reported in a reproducible manner: see #43 and https://wandb.ai/openrlbenchmark/MORL-Baselines.

Example of our dashboards: Pareto front and multi-objective metrics are visible in real-time.

- Python
Published by ffelten over 2 years ago

morl-baselines - 1.0.0-rc2 bugfixes and enhancements

What's Changed

• Change PQL to linearly decaying exploration by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/48
• Refactor random seed by @LucasAlegre in https://github.com/LucasAlegre/morl-baselines/pull/49
• Recover from solver error in OLS by @ffelten in https://github.com/LucasAlegre/morl-baselines/pull/51

Full Changelog: https://github.com/LucasAlegre/morl-baselines/compare/1.0.0-rc1...1.0.0-rc2

- Python
Published by ffelten almost 3 years ago

morl-baselines - 1.0.0-rc1 Stabilizing and performance assessment

First release candidate aiming at stabilizing and reporting the performances of the algorithms in the codebase. We aim to fix bugs as we encounter them when assessing performances and bumping RC numbers along the way. Once we have finished the performance assessments, we should be able to release 1.0.0.

- Python
Published by ffelten almost 3 years ago