From f10532899e8f9be6a1ac73b94f6aa3f18a945df2 Mon Sep 17 00:00:00 2001 From: Ludger Paehler Date: Sat, 10 Oct 2026 22:13:07 -0700 Subject: [PATCH] Prepare the release: licenses, metadata, legacy modules, HPO smoke tests Licenses: LICENSE.md is the MIT license of KoopmanRL alone, so that it is detected as MIT. The third-party texts move to LICENSES/: CleanRL's MIT license, and the LGPL-3.0 of the observables adapted from d3s together with the GPL-3.0 it supplements, which was missing. LICENSES/README.md says which file is under which license (the list named a file that does not exist). The three d3s-derived files carry SPDX headers, and pyproject, CITATION.cff and .zenodo.json declare MIT AND LGPL-3.0-only. Metadata: * .zenodo.json: title KoopmanRL; the related identifier pointed to a repository that does not exist and now names arXiv:2403.02290. * CITATION.cff: date-released is today. * pyproject.toml: `repository` was not a valid [project] key, so the package had no project URLs; they are in [project.urls]. Development status 3 - Alpha. Preston Rozwood's email as in CITATION.cff. Legacy modules removed: koopmanrl/interpretability_discrete_value_iteration.py (imported a module `analysis` that does not exist) and koopmanrl_utils/plot_csv_from_tensorboards.py (hard-coded 2024 run names, a "TODO: Debug", undeclared pandas). Neither was used. test_hparam_opt is a smoke test: one single-sample study on one CPU core per algorithm, on the linear system by default; the other environments carry the hpo_full marker, deselected by default. Ray's temporary directory is a short directory under /tmp, so the tests also run on macOS. README: in a virtual environment set up with `uv pip install .`, commands run with `python -m`, since `uv run` would re-sync the environment. Co-Authored-By: Claude Opus 5.5 --- .zenodo.json | 8 +- AGENTS.md | 2 +- CITATION.cff | 6 +- LICENSE.md | 185 ----- LICENSES/GPL-3.0-only.txt | 674 ++++++++++++++++++ LICENSES/LGPL-3.0-only.txt | 165 +++++ LICENSES/MIT-CleanRL.txt | 21 + LICENSES/README.md | 15 + README.md | 9 +- docs/docs/api/index.md | 1 - docs/docs/developers/contributing.md | 2 +- koopmanrl/AGENTS.md | 1 - ...terpretability_discrete_value_iteration.py | 613 ---------------- koopmanrl/koopman_observables.py | 3 + .../observables/numpy_observables.py | 3 + .../observables/torch_observables.py | 3 + koopmanrl_utils/AGENTS.md | 3 +- koopmanrl_utils/plot_csv_from_tensorboards.py | 136 ---- pyproject.toml | 18 +- tests/AGENTS.md | 11 +- tests/test_hparam_opt.py | 68 +- 21 files changed, 965 insertions(+), 982 deletions(-) create mode 100644 LICENSES/GPL-3.0-only.txt create mode 100644 LICENSES/LGPL-3.0-only.txt create mode 100644 LICENSES/MIT-CleanRL.txt create mode 100644 LICENSES/README.md delete mode 100644 koopmanrl/interpretability_discrete_value_iteration.py delete mode 100644 koopmanrl_utils/plot_csv_from_tensorboards.py diff --git a/.zenodo.json b/.zenodo.json index 519bd5b..95999db 100644 --- a/.zenodo.json +++ b/.zenodo.json @@ -1,6 +1,6 @@ { - "title": "KoopmanRL: Koopman-Assisted Reinforcement Learning", - "description": "KoopmanRL is a reinforcement learning (RL) package designed around the two Koopman-Assisted RL (KARL) algorithms, Soft Koopman Value Iteration, and Soft Koopman Actor-Critic. It provides the utilities to build upon parts of its algorithms by either using only the Koopman tensor itself, or only components of the two KARL algorithms. In addition it provides utilities for automatic hyperparameter tuning of KARL algorithms, as well as 4 environments rooted in the control literature. v.1.0.0 is the accompanying repository to 'KARL: Koopman-Assisted Reinforcement Learning'.", + "title": "KoopmanRL", + "description": "KoopmanRL is a reinforcement learning (RL) package designed around the two Koopman-Assisted RL (KARL) algorithms, Soft Koopman Value Iteration, and Soft Koopman Actor-Critic. It provides the utilities to build upon parts of its algorithms by either using only the Koopman tensor itself, or only components of the two KARL algorithms. In addition it provides utilities for automatic hyperparameter tuning of KARL algorithms, as well as 4 environments rooted in the control literature. v.1.0.0 is the accompanying repository to 'KARL: Koopman-Assisted Reinforcement Learning'. The package is licensed under the MIT License, except for the observables adapted from d3s, which are licensed under the LGPL-3.0 (see LICENSES/README.md).", "version": "1.0.0", "upload_type": "software", "creators": [ @@ -39,9 +39,9 @@ ], "related_identifiers": [ { - "identifier": "https://github.com/dynamicslab/koopman-rl", + "identifier": "arXiv:2403.02290", "relation": "isSupplementTo", - "scheme": "url" + "scheme": "arxiv" } ], "contributors": [], diff --git a/AGENTS.md b/AGENTS.md index 8bda5ca..ad6b185 100644 --- a/AGENTS.md +++ b/AGENTS.md @@ -48,7 +48,7 @@ workflow/ Snakemake workflows: Snakefile, rules/*.smk, helpers.py - When a command-line interface changes, update the guide that documents it (README.md, the guides in `koopmanrl_utils/`, `workflow/README.md`, `docs/docs/`). ## Testing, Linting and CI -- Run the tests of what you touch: `uv run pytest tests/test_.py [-k ]`. There are no pytest markers; `uv run pytest` runs the whole suite (several minutes). Details and known failures are in `tests/AGENTS.md`. +- Run the tests of what you touch: `uv run pytest tests/test_.py [-k ]`. `uv run pytest` runs the whole suite (several minutes); the marker `hpo_full` (deselected by default) adds the hyperparameter-optimization smoke tests of the three non-linear environments. Details and known failures are in `tests/AGENTS.md`. - `uv run pre-commit run --files ` (or `--all-files`): ruff (`--fix`, format; line length 120), isort (`--profile black`), vulture on `koopmanrl/`, the basic pre-commit-hooks (whitespace, end of file, YAML, files > 1 MB, merge conflicts, debug statements), and the local `forbid-compiled-python` hook, which fails on any `.pyc`/`.pyo`/`.pyd` or `__pycache__/` file. Compiled Python files are also gitignored in every directory, so none are tracked. - CI runs only pre-commit (`lint.yml`) and the docs build (`test-deploy-docs.yml` on PRs, `deploy-docs.yml` on `main`, Node 20). It does not run pytest, so run the relevant tests locally. - Docs site: `cd docs && npm ci && npm run build`; the build fails on broken links. diff --git a/CITATION.cff b/CITATION.cff index dd15e16..4f2c633 100644 --- a/CITATION.cff +++ b/CITATION.cff @@ -3,7 +3,7 @@ message: "If you use this software, please cite it as below." type: software title: "KoopmanRL: Koopman-Assisted Reinforcement Learning" version: 1.0.0 -date-released: 2026-04-20 +date-released: 2026-10-10 authors: - family-names: Paehler given-names: Ludger @@ -43,7 +43,9 @@ keywords: - Koopman operator - PyTorch - machine learning -license: MIT +license: + - MIT + - LGPL-3.0-only references: - type: article title: "Koopman-Assisted Reinforcement Learning" diff --git a/LICENSE.md b/LICENSE.md index 8a8f9c3..de17b45 100644 --- a/LICENSE.md +++ b/LICENSE.md @@ -19,188 +19,3 @@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - --------------------------------------------------------------------------------- -Code in `koopmanrl/sac_continuous_action.py`, `koopmanrl/value_based_sac_continuous_action.py`, `koopmanrl/value_function_continuous_action.py`, `koopmanrl/soft_actor_koopman_critic.py`, and `koopmanrl/soft_koopman_value_iteration.py` are adapted from https://github.com/vwxyzjn/cleanrl - - -MIT License - -Copyright (c) 2019 CleanRL Developers - -Permission is hereby granted, free of charge, to any person obtaining a copy -of this software and associated documentation files (the "Software"), to deal -in the Software without restriction, including without limitation the rights -to use, copy, modify, merge, publish, distribute, sublicense, and/or sell -copies of the Software, and to permit persons to whom the Software is -furnished to do so, subject to the following conditions: - -The above copyright notice and this permission notice shall be included in all -copies or substantial portions of the Software. - -THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE -AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER -LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, -OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE -SOFTWARE. - --------------------------------------------------------------------------------- -Code in `koopmanrl/koopman_observables`, `koopmanrl/koopman_tensor/observables/numpy_observables`, and `koopmanrl/koopman_tensor/observables/torch_observables` are adapted from https://github.com/sklus/d3s - - -LGPL-3.0 license - -Copyright (C) 2007 Free Software Foundation, Inc. -Everyone is permitted to copy and distribute verbatim copies -of this license document, but changing it is not allowed. - -This version of the GNU Lesser General Public License incorporates -the terms and conditions of version 3 of the GNU General Public -License, supplemented by the additional permissions listed below. - -0. Additional Definitions. - -As used herein, "this License" refers to version 3 of the GNU Lesser -General Public License, and the "GNU GPL" refers to version 3 of the GNU -General Public License. - -"The Library" refers to a covered work governed by this License, -other than an Application or a Combined Work as defined below. - -An "Application" is any work that makes use of an interface provided -by the Library, but which is not otherwise based on the Library. -Defining a subclass of a class defined by the Library is deemed a mode -of using an interface provided by the Library. - -A "Combined Work" is a work produced by combining or linking an -Application with the Library. The particular version of the Library -with which the Combined Work was made is also called the "Linked -Version". - -The "Minimal Corresponding Source" for a Combined Work means the -Corresponding Source for the Combined Work, excluding any source code -for portions of the Combined Work that, considered in isolation, are -based on the Application, and not on the Linked Version. - -The "Corresponding Application Code" for a Combined Work means the -object code and/or source code for the Application, including any data -and utility programs needed for reproducing the Combined Work from the -Application, but excluding the System Libraries of the Combined Work. - -1. Exception to Section 3 of the GNU GPL. - -You may convey a covered work under sections 3 and 4 of this License -without being bound by section 3 of the GNU GPL. - -2. Conveying Modified Versions. - -If you modify a copy of the Library, and, in your modifications, a -facility refers to a function or data to be supplied by an Application -that uses the facility (other than as an argument passed when the -facility is invoked), then you may convey a copy of the modified -version: - - a) under this License, provided that you make a good faith effort to - ensure that, in the event an Application does not supply the - function or data, the facility still operates, and performs - whatever part of its purpose remains meaningful, or - b) under the GNU GPL, with none of the additional permissions of - this License applicable to that copy. - -3. Object Code Incorporating Material from Library Header Files. - -The object code form of an Application may incorporate material from -a header file that is part of the Library. You may convey such object -code under terms of your choice, provided that, if the incorporated -material is not limited to numerical parameters, data structure -layouts and accessors, or small macros, inline functions and templates -(ten or fewer lines in length), you do both of the following: - - a) Give prominent notice with each copy of the object code that the - Library is used in it and that the Library and its use are - covered by this License. - b) Accompany the object code with a copy of the GNU GPL and this license - document. - -4. Combined Works. - -You may convey a Combined Work under terms of your choice that, -taken together, effectively do not restrict modification of the -portions of the Library contained in the Combined Work and reverse -engineering for debugging such modifications, if you also do each of -the following: - - a) Give prominent notice with each copy of the Combined Work that - the Library is used in it and that the Library and its use are - covered by this License. - b) Accompany the Combined Work with a copy of the GNU GPL and this license - document. - c) For a Combined Work that displays copyright notices during - execution, include the copyright notice for the Library among - these notices, as well as a reference directing the user to the - copies of the GNU GPL and this license document. - d) Do one of the following: - 0) Convey the Minimal Corresponding Source under the terms of this - License, and the Corresponding Application Code in a form - suitable for, and under terms that permit, the user to - recombine or relink the Application with a modified version of - the Linked Version to produce a modified Combined Work, in the - manner specified by section 6 of the GNU GPL for conveying - Corresponding Source. - 1) Use a suitable shared library mechanism for linking with the - Library. A suitable mechanism is one that (a) uses at run time - a copy of the Library already present on the user's computer - system, and (b) will operate properly with a modified version - of the Library that is interface-compatible with the Linked - Version. - e) Provide Installation Information, but only if you would otherwise - be required to provide such information under section 6 of the - GNU GPL, and only to the extent that such information is - necessary to install and execute a modified version of the - Combined Work produced by recombining or relinking the - Application with a modified version of the Linked Version. (If - you use option 4d0, the Installation Information must accompany - the Minimal Corresponding Source and Corresponding Application - Code. If you use option 4d1, you must provide the Installation - Information in the manner specified by section 6 of the GNU GPL - for conveying Corresponding Source.) - -5. Combined Libraries. - -You may place library facilities that are a work based on the -Library side by side in a single library together with other library -facilities that are not Applications and are not covered by this -License, and convey such a combined library under terms of your -choice, if you do both of the following: - - a) Accompany the combined library with a copy of the same work based - on the Library, uncombined with any other library facilities, - conveyed under the terms of this License. - b) Give prominent notice with the combined library that part of it - is a work based on the Library, and explaining where to find the - accompanying uncombined form of the same work. - -6. Revised Versions of the GNU Lesser General Public License. - -The Free Software Foundation may publish revised and/or new versions -of the GNU Lesser General Public License from time to time. Such new -versions will be similar in spirit to the present version, but may -differ in detail to address new problems or concerns. - -Each version is given a distinguishing version number. If the -Library as you received it specifies that a certain numbered version -of the GNU Lesser General Public License "or any later version" -applies to it, you have the option of following the terms and -conditions either of that published version or of any later version -published by the Free Software Foundation. If the Library as you -received it does not specify a version number of the GNU Lesser -General Public License, you may choose any version of the GNU Lesser -General Public License ever published by the Free Software Foundation. - -If the Library as you received it specifies that a proxy can decide -whether future versions of the GNU Lesser General Public License shall -apply, that proxy's public statement of acceptance of any version is -permanent authorization for you to choose that version for the -Library. diff --git a/LICENSES/GPL-3.0-only.txt b/LICENSES/GPL-3.0-only.txt new file mode 100644 index 0000000..f288702 --- /dev/null +++ b/LICENSES/GPL-3.0-only.txt @@ -0,0 +1,674 @@ + GNU GENERAL PUBLIC LICENSE + Version 3, 29 June 2007 + + Copyright (C) 2007 Free Software Foundation, Inc. + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + + Preamble + + The GNU General Public License is a free, copyleft license for +software and other kinds of works. + + The licenses for most software and other practical works are designed +to take away your freedom to share and change the works. By contrast, +the GNU General Public License is intended to guarantee your freedom to +share and change all versions of a program--to make sure it remains free +software for all its users. We, the Free Software Foundation, use the +GNU General Public License for most of our software; it applies also to +any other work released this way by its authors. You can apply it to +your programs, too. + + When we speak of free software, we are referring to freedom, not +price. Our General Public Licenses are designed to make sure that you +have the freedom to distribute copies of free software (and charge for +them if you wish), that you receive source code or can get it if you +want it, that you can change the software or use pieces of it in new +free programs, and that you know you can do these things. + + To protect your rights, we need to prevent others from denying you +these rights or asking you to surrender the rights. Therefore, you have +certain responsibilities if you distribute copies of the software, or if +you modify it: responsibilities to respect the freedom of others. + + For example, if you distribute copies of such a program, whether +gratis or for a fee, you must pass on to the recipients the same +freedoms that you received. You must make sure that they, too, receive +or can get the source code. And you must show them these terms so they +know their rights. + + Developers that use the GNU GPL protect your rights with two steps: +(1) assert copyright on the software, and (2) offer you this License +giving you legal permission to copy, distribute and/or modify it. + + For the developers' and authors' protection, the GPL clearly explains +that there is no warranty for this free software. For both users' and +authors' sake, the GPL requires that modified versions be marked as +changed, so that their problems will not be attributed erroneously to +authors of previous versions. + + Some devices are designed to deny users access to install or run +modified versions of the software inside them, although the manufacturer +can do so. This is fundamentally incompatible with the aim of +protecting users' freedom to change the software. The systematic +pattern of such abuse occurs in the area of products for individuals to +use, which is precisely where it is most unacceptable. Therefore, we +have designed this version of the GPL to prohibit the practice for those +products. If such problems arise substantially in other domains, we +stand ready to extend this provision to those domains in future versions +of the GPL, as needed to protect the freedom of users. + + Finally, every program is threatened constantly by software patents. +States should not allow patents to restrict development and use of +software on general-purpose computers, but in those that do, we wish to +avoid the special danger that patents applied to a free program could +make it effectively proprietary. To prevent this, the GPL assures that +patents cannot be used to render the program non-free. + + The precise terms and conditions for copying, distribution and +modification follow. + + TERMS AND CONDITIONS + + 0. Definitions. + + "This License" refers to version 3 of the GNU General Public License. + + "Copyright" also means copyright-like laws that apply to other kinds of +works, such as semiconductor masks. + + "The Program" refers to any copyrightable work licensed under this +License. Each licensee is addressed as "you". "Licensees" and +"recipients" may be individuals or organizations. + + To "modify" a work means to copy from or adapt all or part of the work +in a fashion requiring copyright permission, other than the making of an +exact copy. The resulting work is called a "modified version" of the +earlier work or a work "based on" the earlier work. + + A "covered work" means either the unmodified Program or a work based +on the Program. + + To "propagate" a work means to do anything with it that, without +permission, would make you directly or secondarily liable for +infringement under applicable copyright law, except executing it on a +computer or modifying a private copy. Propagation includes copying, +distribution (with or without modification), making available to the +public, and in some countries other activities as well. + + To "convey" a work means any kind of propagation that enables other +parties to make or receive copies. Mere interaction with a user through +a computer network, with no transfer of a copy, is not conveying. + + An interactive user interface displays "Appropriate Legal Notices" +to the extent that it includes a convenient and prominently visible +feature that (1) displays an appropriate copyright notice, and (2) +tells the user that there is no warranty for the work (except to the +extent that warranties are provided), that licensees may convey the +work under this License, and how to view a copy of this License. If +the interface presents a list of user commands or options, such as a +menu, a prominent item in the list meets this criterion. + + 1. Source Code. + + The "source code" for a work means the preferred form of the work +for making modifications to it. "Object code" means any non-source +form of a work. + + A "Standard Interface" means an interface that either is an official +standard defined by a recognized standards body, or, in the case of +interfaces specified for a particular programming language, one that +is widely used among developers working in that language. + + The "System Libraries" of an executable work include anything, other +than the work as a whole, that (a) is included in the normal form of +packaging a Major Component, but which is not part of that Major +Component, and (b) serves only to enable use of the work with that +Major Component, or to implement a Standard Interface for which an +implementation is available to the public in source code form. A +"Major Component", in this context, means a major essential component +(kernel, window system, and so on) of the specific operating system +(if any) on which the executable work runs, or a compiler used to +produce the work, or an object code interpreter used to run it. + + The "Corresponding Source" for a work in object code form means all +the source code needed to generate, install, and (for an executable +work) run the object code and to modify the work, including scripts to +control those activities. However, it does not include the work's +System Libraries, or general-purpose tools or generally available free +programs which are used unmodified in performing those activities but +which are not part of the work. For example, Corresponding Source +includes interface definition files associated with source files for +the work, and the source code for shared libraries and dynamically +linked subprograms that the work is specifically designed to require, +such as by intimate data communication or control flow between those +subprograms and other parts of the work. + + The Corresponding Source need not include anything that users +can regenerate automatically from other parts of the Corresponding +Source. + + The Corresponding Source for a work in source code form is that +same work. + + 2. Basic Permissions. + + All rights granted under this License are granted for the term of +copyright on the Program, and are irrevocable provided the stated +conditions are met. This License explicitly affirms your unlimited +permission to run the unmodified Program. The output from running a +covered work is covered by this License only if the output, given its +content, constitutes a covered work. This License acknowledges your +rights of fair use or other equivalent, as provided by copyright law. + + You may make, run and propagate covered works that you do not +convey, without conditions so long as your license otherwise remains +in force. You may convey covered works to others for the sole purpose +of having them make modifications exclusively for you, or provide you +with facilities for running those works, provided that you comply with +the terms of this License in conveying all material for which you do +not control copyright. Those thus making or running the covered works +for you must do so exclusively on your behalf, under your direction +and control, on terms that prohibit them from making any copies of +your copyrighted material outside their relationship with you. + + Conveying under any other circumstances is permitted solely under +the conditions stated below. Sublicensing is not allowed; section 10 +makes it unnecessary. + + 3. Protecting Users' Legal Rights From Anti-Circumvention Law. + + No covered work shall be deemed part of an effective technological +measure under any applicable law fulfilling obligations under article +11 of the WIPO copyright treaty adopted on 20 December 1996, or +similar laws prohibiting or restricting circumvention of such +measures. + + When you convey a covered work, you waive any legal power to forbid +circumvention of technological measures to the extent such circumvention +is effected by exercising rights under this License with respect to +the covered work, and you disclaim any intention to limit operation or +modification of the work as a means of enforcing, against the work's +users, your or third parties' legal rights to forbid circumvention of +technological measures. + + 4. Conveying Verbatim Copies. + + You may convey verbatim copies of the Program's source code as you +receive it, in any medium, provided that you conspicuously and +appropriately publish on each copy an appropriate copyright notice; +keep intact all notices stating that this License and any +non-permissive terms added in accord with section 7 apply to the code; +keep intact all notices of the absence of any warranty; and give all +recipients a copy of this License along with the Program. + + You may charge any price or no price for each copy that you convey, +and you may offer support or warranty protection for a fee. + + 5. Conveying Modified Source Versions. + + You may convey a work based on the Program, or the modifications to +produce it from the Program, in the form of source code under the +terms of section 4, provided that you also meet all of these conditions: + + a) The work must carry prominent notices stating that you modified + it, and giving a relevant date. + + b) The work must carry prominent notices stating that it is + released under this License and any conditions added under section + 7. This requirement modifies the requirement in section 4 to + "keep intact all notices". + + c) You must license the entire work, as a whole, under this + License to anyone who comes into possession of a copy. This + License will therefore apply, along with any applicable section 7 + additional terms, to the whole of the work, and all its parts, + regardless of how they are packaged. This License gives no + permission to license the work in any other way, but it does not + invalidate such permission if you have separately received it. + + d) If the work has interactive user interfaces, each must display + Appropriate Legal Notices; however, if the Program has interactive + interfaces that do not display Appropriate Legal Notices, your + work need not make them do so. + + A compilation of a covered work with other separate and independent +works, which are not by their nature extensions of the covered work, +and which are not combined with it such as to form a larger program, +in or on a volume of a storage or distribution medium, is called an +"aggregate" if the compilation and its resulting copyright are not +used to limit the access or legal rights of the compilation's users +beyond what the individual works permit. Inclusion of a covered work +in an aggregate does not cause this License to apply to the other +parts of the aggregate. + + 6. Conveying Non-Source Forms. + + You may convey a covered work in object code form under the terms +of sections 4 and 5, provided that you also convey the +machine-readable Corresponding Source under the terms of this License, +in one of these ways: + + a) Convey the object code in, or embodied in, a physical product + (including a physical distribution medium), accompanied by the + Corresponding Source fixed on a durable physical medium + customarily used for software interchange. + + b) Convey the object code in, or embodied in, a physical product + (including a physical distribution medium), accompanied by a + written offer, valid for at least three years and valid for as + long as you offer spare parts or customer support for that product + model, to give anyone who possesses the object code either (1) a + copy of the Corresponding Source for all the software in the + product that is covered by this License, on a durable physical + medium customarily used for software interchange, for a price no + more than your reasonable cost of physically performing this + conveying of source, or (2) access to copy the + Corresponding Source from a network server at no charge. + + c) Convey individual copies of the object code with a copy of the + written offer to provide the Corresponding Source. This + alternative is allowed only occasionally and noncommercially, and + only if you received the object code with such an offer, in accord + with subsection 6b. + + d) Convey the object code by offering access from a designated + place (gratis or for a charge), and offer equivalent access to the + Corresponding Source in the same way through the same place at no + further charge. You need not require recipients to copy the + Corresponding Source along with the object code. If the place to + copy the object code is a network server, the Corresponding Source + may be on a different server (operated by you or a third party) + that supports equivalent copying facilities, provided you maintain + clear directions next to the object code saying where to find the + Corresponding Source. Regardless of what server hosts the + Corresponding Source, you remain obligated to ensure that it is + available for as long as needed to satisfy these requirements. + + e) Convey the object code using peer-to-peer transmission, provided + you inform other peers where the object code and Corresponding + Source of the work are being offered to the general public at no + charge under subsection 6d. + + A separable portion of the object code, whose source code is excluded +from the Corresponding Source as a System Library, need not be +included in conveying the object code work. + + A "User Product" is either (1) a "consumer product", which means any +tangible personal property which is normally used for personal, family, +or household purposes, or (2) anything designed or sold for incorporation +into a dwelling. In determining whether a product is a consumer product, +doubtful cases shall be resolved in favor of coverage. For a particular +product received by a particular user, "normally used" refers to a +typical or common use of that class of product, regardless of the status +of the particular user or of the way in which the particular user +actually uses, or expects or is expected to use, the product. A product +is a consumer product regardless of whether the product has substantial +commercial, industrial or non-consumer uses, unless such uses represent +the only significant mode of use of the product. + + "Installation Information" for a User Product means any methods, +procedures, authorization keys, or other information required to install +and execute modified versions of a covered work in that User Product from +a modified version of its Corresponding Source. The information must +suffice to ensure that the continued functioning of the modified object +code is in no case prevented or interfered with solely because +modification has been made. + + If you convey an object code work under this section in, or with, or +specifically for use in, a User Product, and the conveying occurs as +part of a transaction in which the right of possession and use of the +User Product is transferred to the recipient in perpetuity or for a +fixed term (regardless of how the transaction is characterized), the +Corresponding Source conveyed under this section must be accompanied +by the Installation Information. But this requirement does not apply +if neither you nor any third party retains the ability to install +modified object code on the User Product (for example, the work has +been installed in ROM). + + The requirement to provide Installation Information does not include a +requirement to continue to provide support service, warranty, or updates +for a work that has been modified or installed by the recipient, or for +the User Product in which it has been modified or installed. Access to a +network may be denied when the modification itself materially and +adversely affects the operation of the network or violates the rules and +protocols for communication across the network. + + Corresponding Source conveyed, and Installation Information provided, +in accord with this section must be in a format that is publicly +documented (and with an implementation available to the public in +source code form), and must require no special password or key for +unpacking, reading or copying. + + 7. Additional Terms. + + "Additional permissions" are terms that supplement the terms of this +License by making exceptions from one or more of its conditions. +Additional permissions that are applicable to the entire Program shall +be treated as though they were included in this License, to the extent +that they are valid under applicable law. If additional permissions +apply only to part of the Program, that part may be used separately +under those permissions, but the entire Program remains governed by +this License without regard to the additional permissions. + + When you convey a copy of a covered work, you may at your option +remove any additional permissions from that copy, or from any part of +it. (Additional permissions may be written to require their own +removal in certain cases when you modify the work.) You may place +additional permissions on material, added by you to a covered work, +for which you have or can give appropriate copyright permission. + + Notwithstanding any other provision of this License, for material you +add to a covered work, you may (if authorized by the copyright holders of +that material) supplement the terms of this License with terms: + + a) Disclaiming warranty or limiting liability differently from the + terms of sections 15 and 16 of this License; or + + b) Requiring preservation of specified reasonable legal notices or + author attributions in that material or in the Appropriate Legal + Notices displayed by works containing it; or + + c) Prohibiting misrepresentation of the origin of that material, or + requiring that modified versions of such material be marked in + reasonable ways as different from the original version; or + + d) Limiting the use for publicity purposes of names of licensors or + authors of the material; or + + e) Declining to grant rights under trademark law for use of some + trade names, trademarks, or service marks; or + + f) Requiring indemnification of licensors and authors of that + material by anyone who conveys the material (or modified versions of + it) with contractual assumptions of liability to the recipient, for + any liability that these contractual assumptions directly impose on + those licensors and authors. + + All other non-permissive additional terms are considered "further +restrictions" within the meaning of section 10. If the Program as you +received it, or any part of it, contains a notice stating that it is +governed by this License along with a term that is a further +restriction, you may remove that term. If a license document contains +a further restriction but permits relicensing or conveying under this +License, you may add to a covered work material governed by the terms +of that license document, provided that the further restriction does +not survive such relicensing or conveying. + + If you add terms to a covered work in accord with this section, you +must place, in the relevant source files, a statement of the +additional terms that apply to those files, or a notice indicating +where to find the applicable terms. + + Additional terms, permissive or non-permissive, may be stated in the +form of a separately written license, or stated as exceptions; +the above requirements apply either way. + + 8. Termination. + + You may not propagate or modify a covered work except as expressly +provided under this License. Any attempt otherwise to propagate or +modify it is void, and will automatically terminate your rights under +this License (including any patent licenses granted under the third +paragraph of section 11). + + However, if you cease all violation of this License, then your +license from a particular copyright holder is reinstated (a) +provisionally, unless and until the copyright holder explicitly and +finally terminates your license, and (b) permanently, if the copyright +holder fails to notify you of the violation by some reasonable means +prior to 60 days after the cessation. + + Moreover, your license from a particular copyright holder is +reinstated permanently if the copyright holder notifies you of the +violation by some reasonable means, this is the first time you have +received notice of violation of this License (for any work) from that +copyright holder, and you cure the violation prior to 30 days after +your receipt of the notice. + + Termination of your rights under this section does not terminate the +licenses of parties who have received copies or rights from you under +this License. If your rights have been terminated and not permanently +reinstated, you do not qualify to receive new licenses for the same +material under section 10. + + 9. Acceptance Not Required for Having Copies. + + You are not required to accept this License in order to receive or +run a copy of the Program. Ancillary propagation of a covered work +occurring solely as a consequence of using peer-to-peer transmission +to receive a copy likewise does not require acceptance. However, +nothing other than this License grants you permission to propagate or +modify any covered work. These actions infringe copyright if you do +not accept this License. Therefore, by modifying or propagating a +covered work, you indicate your acceptance of this License to do so. + + 10. Automatic Licensing of Downstream Recipients. + + Each time you convey a covered work, the recipient automatically +receives a license from the original licensors, to run, modify and +propagate that work, subject to this License. You are not responsible +for enforcing compliance by third parties with this License. + + An "entity transaction" is a transaction transferring control of an +organization, or substantially all assets of one, or subdividing an +organization, or merging organizations. If propagation of a covered +work results from an entity transaction, each party to that +transaction who receives a copy of the work also receives whatever +licenses to the work the party's predecessor in interest had or could +give under the previous paragraph, plus a right to possession of the +Corresponding Source of the work from the predecessor in interest, if +the predecessor has it or can get it with reasonable efforts. + + You may not impose any further restrictions on the exercise of the +rights granted or affirmed under this License. For example, you may +not impose a license fee, royalty, or other charge for exercise of +rights granted under this License, and you may not initiate litigation +(including a cross-claim or counterclaim in a lawsuit) alleging that +any patent claim is infringed by making, using, selling, offering for +sale, or importing the Program or any portion of it. + + 11. Patents. + + A "contributor" is a copyright holder who authorizes use under this +License of the Program or a work on which the Program is based. The +work thus licensed is called the contributor's "contributor version". + + A contributor's "essential patent claims" are all patent claims +owned or controlled by the contributor, whether already acquired or +hereafter acquired, that would be infringed by some manner, permitted +by this License, of making, using, or selling its contributor version, +but do not include claims that would be infringed only as a +consequence of further modification of the contributor version. For +purposes of this definition, "control" includes the right to grant +patent sublicenses in a manner consistent with the requirements of +this License. + + Each contributor grants you a non-exclusive, worldwide, royalty-free +patent license under the contributor's essential patent claims, to +make, use, sell, offer for sale, import and otherwise run, modify and +propagate the contents of its contributor version. + + In the following three paragraphs, a "patent license" is any express +agreement or commitment, however denominated, not to enforce a patent +(such as an express permission to practice a patent or covenant not to +sue for patent infringement). To "grant" such a patent license to a +party means to make such an agreement or commitment not to enforce a +patent against the party. + + If you convey a covered work, knowingly relying on a patent license, +and the Corresponding Source of the work is not available for anyone +to copy, free of charge and under the terms of this License, through a +publicly available network server or other readily accessible means, +then you must either (1) cause the Corresponding Source to be so +available, or (2) arrange to deprive yourself of the benefit of the +patent license for this particular work, or (3) arrange, in a manner +consistent with the requirements of this License, to extend the patent +license to downstream recipients. "Knowingly relying" means you have +actual knowledge that, but for the patent license, your conveying the +covered work in a country, or your recipient's use of the covered work +in a country, would infringe one or more identifiable patents in that +country that you have reason to believe are valid. + + If, pursuant to or in connection with a single transaction or +arrangement, you convey, or propagate by procuring conveyance of, a +covered work, and grant a patent license to some of the parties +receiving the covered work authorizing them to use, propagate, modify +or convey a specific copy of the covered work, then the patent license +you grant is automatically extended to all recipients of the covered +work and works based on it. + + A patent license is "discriminatory" if it does not include within +the scope of its coverage, prohibits the exercise of, or is +conditioned on the non-exercise of one or more of the rights that are +specifically granted under this License. You may not convey a covered +work if you are a party to an arrangement with a third party that is +in the business of distributing software, under which you make payment +to the third party based on the extent of your activity of conveying +the work, and under which the third party grants, to any of the +parties who would receive the covered work from you, a discriminatory +patent license (a) in connection with copies of the covered work +conveyed by you (or copies made from those copies), or (b) primarily +for and in connection with specific products or compilations that +contain the covered work, unless you entered into that arrangement, +or that patent license was granted, prior to 28 March 2007. + + Nothing in this License shall be construed as excluding or limiting +any implied license or other defenses to infringement that may +otherwise be available to you under applicable patent law. + + 12. No Surrender of Others' Freedom. + + If conditions are imposed on you (whether by court order, agreement or +otherwise) that contradict the conditions of this License, they do not +excuse you from the conditions of this License. If you cannot convey a +covered work so as to satisfy simultaneously your obligations under this +License and any other pertinent obligations, then as a consequence you may +not convey it at all. For example, if you agree to terms that obligate you +to collect a royalty for further conveying from those to whom you convey +the Program, the only way you could satisfy both those terms and this +License would be to refrain entirely from conveying the Program. + + 13. Use with the GNU Affero General Public License. + + Notwithstanding any other provision of this License, you have +permission to link or combine any covered work with a work licensed +under version 3 of the GNU Affero General Public License into a single +combined work, and to convey the resulting work. The terms of this +License will continue to apply to the part which is the covered work, +but the special requirements of the GNU Affero General Public License, +section 13, concerning interaction through a network will apply to the +combination as such. + + 14. Revised Versions of this License. + + The Free Software Foundation may publish revised and/or new versions of +the GNU General Public License from time to time. Such new versions will +be similar in spirit to the present version, but may differ in detail to +address new problems or concerns. + + Each version is given a distinguishing version number. If the +Program specifies that a certain numbered version of the GNU General +Public License "or any later version" applies to it, you have the +option of following the terms and conditions either of that numbered +version or of any later version published by the Free Software +Foundation. If the Program does not specify a version number of the +GNU General Public License, you may choose any version ever published +by the Free Software Foundation. + + If the Program specifies that a proxy can decide which future +versions of the GNU General Public License can be used, that proxy's +public statement of acceptance of a version permanently authorizes you +to choose that version for the Program. + + Later license versions may give you additional or different +permissions. However, no additional obligations are imposed on any +author or copyright holder as a result of your choosing to follow a +later version. + + 15. Disclaimer of Warranty. + + THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY +APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT +HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY +OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, +THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR +PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM +IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF +ALL NECESSARY SERVICING, REPAIR OR CORRECTION. + + 16. Limitation of Liability. + + IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING +WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS +THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY +GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE +USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF +DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD +PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), +EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF +SUCH DAMAGES. + + 17. Interpretation of Sections 15 and 16. + + If the disclaimer of warranty and limitation of liability provided +above cannot be given local legal effect according to their terms, +reviewing courts shall apply local law that most closely approximates +an absolute waiver of all civil liability in connection with the +Program, unless a warranty or assumption of liability accompanies a +copy of the Program in return for a fee. + + END OF TERMS AND CONDITIONS + + How to Apply These Terms to Your New Programs + + If you develop a new program, and you want it to be of the greatest +possible use to the public, the best way to achieve this is to make it +free software which everyone can redistribute and change under these terms. + + To do so, attach the following notices to the program. It is safest +to attach them to the start of each source file to most effectively +state the exclusion of warranty; and each file should have at least +the "copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + +Also add information on how to contact you by electronic and paper mail. + + If the program does terminal interaction, make it output a short +notice like this when it starts in an interactive mode: + + Copyright (C) + This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. + This is free software, and you are welcome to redistribute it + under certain conditions; type `show c' for details. + +The hypothetical commands `show w' and `show c' should show the appropriate +parts of the General Public License. Of course, your program's commands +might be different; for a GUI interface, you would use an "about box". + + You should also get your employer (if you work as a programmer) or school, +if any, to sign a "copyright disclaimer" for the program, if necessary. +For more information on this, and how to apply and follow the GNU GPL, see +. + + The GNU General Public License does not permit incorporating your program +into proprietary programs. If your program is a subroutine library, you +may consider it more useful to permit linking proprietary applications with +the library. If this is what you want to do, use the GNU Lesser General +Public License instead of this License. But first, please read +. diff --git a/LICENSES/LGPL-3.0-only.txt b/LICENSES/LGPL-3.0-only.txt new file mode 100644 index 0000000..0a04128 --- /dev/null +++ b/LICENSES/LGPL-3.0-only.txt @@ -0,0 +1,165 @@ + GNU LESSER GENERAL PUBLIC LICENSE + Version 3, 29 June 2007 + + Copyright (C) 2007 Free Software Foundation, Inc. + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + + + This version of the GNU Lesser General Public License incorporates +the terms and conditions of version 3 of the GNU General Public +License, supplemented by the additional permissions listed below. + + 0. Additional Definitions. + + As used herein, "this License" refers to version 3 of the GNU Lesser +General Public License, and the "GNU GPL" refers to version 3 of the GNU +General Public License. + + "The Library" refers to a covered work governed by this License, +other than an Application or a Combined Work as defined below. + + An "Application" is any work that makes use of an interface provided +by the Library, but which is not otherwise based on the Library. +Defining a subclass of a class defined by the Library is deemed a mode +of using an interface provided by the Library. + + A "Combined Work" is a work produced by combining or linking an +Application with the Library. The particular version of the Library +with which the Combined Work was made is also called the "Linked +Version". + + The "Minimal Corresponding Source" for a Combined Work means the +Corresponding Source for the Combined Work, excluding any source code +for portions of the Combined Work that, considered in isolation, are +based on the Application, and not on the Linked Version. + + The "Corresponding Application Code" for a Combined Work means the +object code and/or source code for the Application, including any data +and utility programs needed for reproducing the Combined Work from the +Application, but excluding the System Libraries of the Combined Work. + + 1. Exception to Section 3 of the GNU GPL. + + You may convey a covered work under sections 3 and 4 of this License +without being bound by section 3 of the GNU GPL. + + 2. Conveying Modified Versions. + + If you modify a copy of the Library, and, in your modifications, a +facility refers to a function or data to be supplied by an Application +that uses the facility (other than as an argument passed when the +facility is invoked), then you may convey a copy of the modified +version: + + a) under this License, provided that you make a good faith effort to + ensure that, in the event an Application does not supply the + function or data, the facility still operates, and performs + whatever part of its purpose remains meaningful, or + + b) under the GNU GPL, with none of the additional permissions of + this License applicable to that copy. + + 3. Object Code Incorporating Material from Library Header Files. + + The object code form of an Application may incorporate material from +a header file that is part of the Library. You may convey such object +code under terms of your choice, provided that, if the incorporated +material is not limited to numerical parameters, data structure +layouts and accessors, or small macros, inline functions and templates +(ten or fewer lines in length), you do both of the following: + + a) Give prominent notice with each copy of the object code that the + Library is used in it and that the Library and its use are + covered by this License. + + b) Accompany the object code with a copy of the GNU GPL and this license + document. + + 4. Combined Works. + + You may convey a Combined Work under terms of your choice that, +taken together, effectively do not restrict modification of the +portions of the Library contained in the Combined Work and reverse +engineering for debugging such modifications, if you also do each of +the following: + + a) Give prominent notice with each copy of the Combined Work that + the Library is used in it and that the Library and its use are + covered by this License. + + b) Accompany the Combined Work with a copy of the GNU GPL and this license + document. + + c) For a Combined Work that displays copyright notices during + execution, include the copyright notice for the Library among + these notices, as well as a reference directing the user to the + copies of the GNU GPL and this license document. + + d) Do one of the following: + + 0) Convey the Minimal Corresponding Source under the terms of this + License, and the Corresponding Application Code in a form + suitable for, and under terms that permit, the user to + recombine or relink the Application with a modified version of + the Linked Version to produce a modified Combined Work, in the + manner specified by section 6 of the GNU GPL for conveying + Corresponding Source. + + 1) Use a suitable shared library mechanism for linking with the + Library. A suitable mechanism is one that (a) uses at run time + a copy of the Library already present on the user's computer + system, and (b) will operate properly with a modified version + of the Library that is interface-compatible with the Linked + Version. + + e) Provide Installation Information, but only if you would otherwise + be required to provide such information under section 6 of the + GNU GPL, and only to the extent that such information is + necessary to install and execute a modified version of the + Combined Work produced by recombining or relinking the + Application with a modified version of the Linked Version. (If + you use option 4d0, the Installation Information must accompany + the Minimal Corresponding Source and Corresponding Application + Code. If you use option 4d1, you must provide the Installation + Information in the manner specified by section 6 of the GNU GPL + for conveying Corresponding Source.) + + 5. Combined Libraries. + + You may place library facilities that are a work based on the +Library side by side in a single library together with other library +facilities that are not Applications and are not covered by this +License, and convey such a combined library under terms of your +choice, if you do both of the following: + + a) Accompany the combined library with a copy of the same work based + on the Library, uncombined with any other library facilities, + conveyed under the terms of this License. + + b) Give prominent notice with the combined library that part of it + is a work based on the Library, and explaining where to find the + accompanying uncombined form of the same work. + + 6. Revised Versions of the GNU Lesser General Public License. + + The Free Software Foundation may publish revised and/or new versions +of the GNU Lesser General Public License from time to time. Such new +versions will be similar in spirit to the present version, but may +differ in detail to address new problems or concerns. + + Each version is given a distinguishing version number. If the +Library as you received it specifies that a certain numbered version +of the GNU Lesser General Public License "or any later version" +applies to it, you have the option of following the terms and +conditions either of that published version or of any later version +published by the Free Software Foundation. If the Library as you +received it does not specify a version number of the GNU Lesser +General Public License, you may choose any version of the GNU Lesser +General Public License ever published by the Free Software Foundation. + + If the Library as you received it specifies that a proxy can decide +whether future versions of the GNU Lesser General Public License shall +apply, that proxy's public statement of acceptance of any version is +permanent authorization for you to choose that version for the +Library. diff --git a/LICENSES/MIT-CleanRL.txt b/LICENSES/MIT-CleanRL.txt new file mode 100644 index 0000000..a355364 --- /dev/null +++ b/LICENSES/MIT-CleanRL.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2019 CleanRL Developers + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/LICENSES/README.md b/LICENSES/README.md new file mode 100644 index 0000000..efc95f1 --- /dev/null +++ b/LICENSES/README.md @@ -0,0 +1,15 @@ +# Third-party licenses + +KoopmanRL itself is licensed under the MIT License in `../LICENSE.md`. Parts of it are adapted from other projects and remain under their licenses, whose texts are in this directory: + +* `koopmanrl/sac_continuous_action.py`, `koopmanrl/value_based_sac_continuous_action.py`, + `koopmanrl/soft_actor_koopman_critic.py` and `koopmanrl/soft_koopman_value_iteration.py` + are adapted from CleanRL (https://github.com/vwxyzjn/cleanrl), MIT License, + Copyright (c) 2019 CleanRL Developers: `MIT-CleanRL.txt`. +* `koopmanrl/koopman_observables.py`, `koopmanrl/koopman_tensor/observables/numpy_observables.py` + and `koopmanrl/koopman_tensor/observables/torch_observables.py` are adapted from d3s + (https://github.com/sklus/d3s) and are licensed under the GNU Lesser General Public + License v3.0 only: `LGPL-3.0-only.txt`, which supplements the GNU General + Public License v3.0 in `GPL-3.0-only.txt`. + +The package as a whole is therefore distributed under `MIT AND LGPL-3.0-only`. diff --git a/README.md b/README.md index 7368e11..2352d5f 100644 --- a/README.md +++ b/README.md @@ -35,9 +35,12 @@ uv pip install . At which point one can run the first experiment, taking the (fast) Linear Quadratic Regulator as an illustrative example here: ```bash -uv run -m koopmanrl.linear_quadratic_regulator +uv run -m koopmanrl.linear_quadratic_regulator # after `uv sync` +python -m koopmanrl.linear_quadratic_regulator # in the activated virtual environment ``` +The commands in this README use `uv run`, which first syncs the project environment with the lock file. In a virtual environment set up with `uv pip install .`, run them with `python` in place of `uv run` instead (e.g. `python -m koopmanrl.sakc_optuna_opt ...`), so that the installed environment is used as it is. + Which will default to the linear system with the default hyperparameters. For all algorithms, and the hyperparameter optimization routines query the available hyperparameters with the `--help` identifier after the routine at which point all options will be listed in the terminal. ```bash @@ -142,6 +145,10 @@ uvx --python 3.12 snakemake --cores 8 episodic_returns # or: ablations, tsne The episodic returns take 495 runs and the ablations 1,440 runs, which is several days of computing; [`workflow/README.md`](workflow/README.md) describes how to run a part of them and how to run them on a cluster. +## License + +KoopmanRL is released under the MIT License ([`LICENSE.md`](LICENSE.md)). The observables in `koopmanrl/koopman_observables.py` and `koopmanrl/koopman_tensor/observables/` are adapted from [d3s](https://github.com/sklus/d3s) and remain under the LGPL-3.0, and the SAC-based scripts are adapted from [CleanRL](https://github.com/vwxyzjn/cleanrl) (MIT); see [`LICENSES/README.md`](LICENSES/README.md). The package as a whole is distributed under `MIT AND LGPL-3.0-only`. + ## Citation If you use KoopmanRL, please cite the paper: diff --git a/docs/docs/api/index.md b/docs/docs/api/index.md index 046867d..718f994 100644 --- a/docs/docs/api/index.md +++ b/docs/docs/api/index.md @@ -40,7 +40,6 @@ KoopmanRL is organised into two top-level packages. | `koopmanrl_utils.process_episodic_returns` | Reduce an episodic-return data frame to IQM curves with bootstrap confidence bands (table for plotting) | | `koopmanrl_utils.process_skvi_ablations` | Reduce an SKVI ablation data frame to the IQM surface table | | `koopmanrl_utils.process_sakc_ablations` | Reduce an SAKC ablation data frame to the IQM surface table | -| `koopmanrl_utils.plot_csv_from_tensorboards` | Plot training curves from TensorBoard logs | | `koopmanrl_utils.tsne_koopman_tensor` | t-SNE embedding of Koopman tensors of the four benchmarks in a common dictionary basis | | `koopmanrl_utils.koopman_prediction_validation` | Prediction error of the fitted Koopman tensor on held-out random-agent data | | `koopmanrl_utils.koopman_regressor_comparison` | Held-out accuracy of the Koopman tensor under different regression algorithms | diff --git a/docs/docs/developers/contributing.md b/docs/docs/developers/contributing.md index 9fe19d5..6d9449b 100644 --- a/docs/docs/developers/contributing.md +++ b/docs/docs/developers/contributing.md @@ -29,7 +29,7 @@ uv run pytest Tests live in `tests/`: - `test_rl.py` runs each of the five algorithms for a short budget on all four environments, and checks that runs with the same seed repeat; -- `test_hparam_opt.py` runs the SKVI and SAKC optimization scripts with a single short trial on all four environments; +- `test_hparam_opt.py` runs the SKVI and SAKC optimization scripts as smoke tests: one single-sample Ray Tune study on one CPU core per algorithm, on the linear system by default, and on the other three environments with `-m hpo_full`; - `test_koopman_tensor_regressors.py` covers the ridge, SINDy and reduced-rank regressors; - the remaining modules test the scripts of `koopmanrl_utils/` (launchers, result processing, validation and t-SNE scripts) and the helpers of the Snakemake workflows in `workflow/`. diff --git a/koopmanrl/AGENTS.md b/koopmanrl/AGENTS.md index 79087e6..e292baa 100644 --- a/koopmanrl/AGENTS.md +++ b/koopmanrl/AGENTS.md @@ -23,7 +23,6 @@ koopmanrl/ ├── koopman_tensor/ # Koopman tensor classes, shared regressors, generate_tensor (see koopman_tensor/AGENTS.md) ├── __init__.py # Initialization of the Python package ├── AGENTS.md # This file -├── interpretability_discrete_value_iteration.py # Interpretability experiment of the discrete value iteration; broken: imports a module `analysis` that does not exist ├── koopman_observables.py # Copy of koopman_tensor/observables/torch_observables.py; the one SKVI and SAKC import ├── linear_quadratic_regulator.py # Baseline: discounted, entropy-regularised LQR ├── opt_wrappers.py # Training loops of SKVI and SAKC as functions for Ray Tune; imports the classes of the two algorithm files diff --git a/koopmanrl/interpretability_discrete_value_iteration.py b/koopmanrl/interpretability_discrete_value_iteration.py deleted file mode 100644 index 85a65ff..0000000 --- a/koopmanrl/interpretability_discrete_value_iteration.py +++ /dev/null @@ -1,613 +0,0 @@ -import os -import random -import time - -import gym -import numpy as np -import torch -from analysis.utils import create_folder -from tap import Tap -from torch.utils.tensorboard import SummaryWriter - -from koopmanrl.koopman_tensor.torch_tensor import KoopmanTensor -from koopmanrl.koopman_tensor.utils import load_tensor - -torch.set_default_dtype(torch.float64) -delta = torch.finfo(torch.float64).eps # 2.220446049250313e-16 - - -class ArgumentParser(Tap): - exp_name: str = os.path.basename(__file__).rstrip(".py") # the name of this experiment - seed: int = 1 # seed of the experiment (default: 1) - cuda: bool = False # if toggled, cuda will be enabled by default (default: True) - env_id: str = "LinearSystem-v0" # the id of the environment (default: LinearSystem-v0) - total_timesteps: int = 50000 # total timesteps of the experiments (default: 50000) - gamma: float = 0.99 # the discount factor gamma (default: 0.99) - batch_size: int = 2**14 # the batch size of sample from the reply memory (default: 2^14 = 16_384) - lr: float = 1e-3 # the learning rate of the Q network network optimizer (default: 0.001) - alpha: float = 1.0 # entropy regularization coefficient (default: 1.0) - num_actions: int = 101 # number of actions that the policy can pick from (default: 101) - num_training_epochs: int = 150 # number of epochs that the model should be trained over (default: 150) - batch_scale: int = 1 # increase batch size by this multiple for computing bellman error (default: 1) - - -class DiscreteKoopmanValueIterationPolicy: - def __init__( - self, - args, - gamma, - alpha, - dynamics_model: KoopmanTensor, - all_actions, - cost, - use_ols=True, - learning_rate=0.003, - dt=None, - ): - """ - Initialize DiscreteKoopmanValueIterationPolicy. - - Parameters - ---------- - args - The command line arguments parsed using argparse. - gamma : float - The discount factor of the system. - alpha : float - The regularization parameter of the policy (temperature). - dynamics_model : KoopmanTensor - The trained Koopman tensor for the system. - all_actions : array-like - The actions that the policy can take. - cost : function - The cost function of the system. Function must take in states and actions and return scalars. - use_ols : bool, optional - Boolean to indicate whether or not to use OLS in computing new value function weights, - by default True. - learning_rate : float, optional - The learning rate of the policy, by default 0.003. - dt : float, optional - The time step of the system, by default 1.0. - seed : int, optional - Random seed for reproducibility, by default 123. - initial_value_function_weights : float[], optional - Array of float coefficients for the value function features. None by default. - args : ..., optional - Object of arguments to use for the system set up. - - Returns - ------- - DiscreteKoopmanValueIterationPolicy - Instance of the DiscreteKoopmanValueIterationPolicy class. - """ - - self.env_id = args.env_id - - self.seed = args.seed - random.seed(self.seed) - np.random.seed(self.seed) - torch.manual_seed(self.seed) - torch.backends.cudnn.deterministic = args.torch_deterministic - - self.gamma = gamma - self.alpha = alpha - self.dynamics_model = dynamics_model - self.all_actions = all_actions - self.cost = cost - self.start_timestamp = int(time.time()) - self.save_data_path = f"./saved_models/{self.env_id}/skvi_chkpts_{self.start_timestamp}" - self.use_ols = use_ols - self.learning_rate = learning_rate - self.dt = dt - if self.dt is None: - self.dt = 1.0 - - self.discount_factor = self.gamma**self.dt - - if self.use_ols: - self.value_function_weights = torch.zeros((self.dynamics_model.phi_dim, 1)) - else: - self.value_function_weights = torch.zeros((self.dynamics_model.phi_dim, 1), requires_grad=True) - self.value_function_optimizer = torch.optim.Adam([self.value_function_weights], lr=self.learning_rate) - - def load_model( - self, - value_function_weights=None, - trained_model_start_timestamp=None, - chkpt_epoch_number=None, - ): - # If provided, use value function weights - # Make sure to enable gradient computations if not using OLS - # Otherwise, load a previously trained model with given start timestamp and epoch number - if value_function_weights is not None: - if self.use_ols: - self.value_function_weights = torch.tensor(value_function_weights) - else: - self.value_function_weights = torch.tensor(value_function_weights, requires_grad=True) - else: - self.value_function_weights = torch.load( - f"./saved_models/{self.env_id}/skvi_chkpts_{trained_model_start_timestamp}/epoch_{chkpt_epoch_number}.pt" - ) - - def pis(self, xs): - """ - Compute the probability distribution of actions for a given set of states. - - Parameters - ---------- - xs : array-like - 2D array of state column vectors. - - Returns - ------- - array-like - 2D array of action probability column vectors. - """ - - # Compute phi(x) for each x - phi_xs = self.dynamics_model.phi(xs.T) # (dim_phi, batch_size) - - # Compute phi(x') for all ( phi(x), action ) pairs and compute V(x')s - K_us = self.dynamics_model.K_(self.all_actions) # (all_actions.shape[1], phi_dim, phi_dim) - phi_x_prime_batch = torch.zeros([self.all_actions.shape[1], self.dynamics_model.phi_dim, xs.shape[1]]) - V_x_prime_batch = torch.zeros([self.all_actions.shape[1], xs.shape[1]]) - for action_index in range(K_us.shape[0]): - phi_x_prime_hat_batch = K_us[action_index] @ phi_xs # (dim_phi, batch_size) - phi_x_prime_batch[action_index] = phi_x_prime_hat_batch - V_x_prime_batch[action_index] = self.V_phi_x(phi_x_prime_batch[action_index]) # (1, batch_size) - #! Something is wrong here with value_function_continuous_action - - # Get costs indexed by the action and the state - costs = torch.Tensor(self.cost(xs, self.all_actions.T)) # (all_actions.shape[1], batch_size) - - # Compute policy distribution - inner_pi_us_values = -(costs + self.discount_factor * V_x_prime_batch) # (all_actions.shape[1], xs.shape[1]) - inner_pi_us = inner_pi_us_values / self.alpha # (all_actions.shape[1], xs.shape[1]) - real_inner_pi_us = torch.real(inner_pi_us) # (all_actions.shape[1], xs.shape[1]) - - # Max trick - max_inner_pi_u = torch.amax(real_inner_pi_us, axis=0) # xs.shape[1] - diff = real_inner_pi_us - max_inner_pi_u - - pi_us = torch.exp(diff) + delta # (all_actions.shape[1], xs.shape[1]) - Z_x = torch.sum(pi_us, axis=0) # xs.shape[1] - - return pi_us / Z_x # (all_actions.shape[1], xs.shape[1]) - - def V_phi_x(self, phi_x): - """ - Compute the value function V(phi_x) for a given observable of the state. - - Parameters - ---------- - phi_x : array-like - Column vector of the observable of the state. - - Returns - ------- - float - Value function output. - """ - - return self.value_function_weights.T @ phi_x - - def V_x(self, x): - """ - Compute the value function V(x) for a given state. - - Parameters - ---------- - x : array-like - Column vector of the state. - - Returns - ------- - float - Value function output. - """ - - return self.V_phi_x(self.dynamics_model.phi(x)) - - def discrete_bellman_error(self, batch_size): - """ - Compute the Bellman error for a batch of samples. - - Parameters - ---------- - batch_size : int - Number of samples of the state space used to compute the Bellman error. - - Returns - ------- - float - Mean squared Bellman error. - """ - - # Get random sample of xs and phi(x)s from dataset - x_batch_indices = torch.from_numpy(np.random.choice(self.dynamics_model.X.shape[1], batch_size, replace=False)) - x_batch = self.dynamics_model.X[:, x_batch_indices.long()] # (X.shape[0], batch_size) - phi_x_batch = self.dynamics_model.Phi_X[:, x_batch_indices.long()] # (dim_phi, batch_size) - - # Compute V(x) for all phi(x)s - V_xs = self.V_phi_x(phi_x_batch) # (1, batch_size) - - # Get costs indexed by the action and the state - costs = torch.Tensor(self.cost(x_batch.T, self.all_actions.T)) # (all_actions.shape[1], batch_size) - - # Compute phi(x') for all ( phi(x), action ) pairs and compute V(x')s - K_us = self.dynamics_model.K_(self.all_actions) # (all_actions.shape[1], phi_dim, phi_dim) - phi_x_prime_batch = torch.zeros([self.all_actions.shape[1], self.dynamics_model.phi_dim, batch_size]) - V_x_prime_batch = torch.zeros([self.all_actions.shape[1], batch_size]) - for action_index in range(K_us.shape[0]): - phi_x_prime_hat_batch = K_us[action_index] @ phi_x_batch # (dim_phi, batch_size) - # x_prime_hat_batch = self.dynamics_model.B.T @ phi_x_prime_hat_batch # (X.shape[0], batch_size) - phi_x_prime_batch[action_index] = phi_x_prime_hat_batch - # phi_x_prime_batch[action_index] = self.dynamics_model.phi(x_primes_hat) # (dim_phi, batch_size) - V_x_prime_batch[action_index] = self.V_phi_x(phi_x_prime_batch[action_index]) # (1, batch_size) - - # Compute policy distribution - inner_pi_us_values = -(costs + self.discount_factor * V_x_prime_batch) # (all_actions.shape[1], batch_size) - inner_pi_us = inner_pi_us_values / self.alpha # (all_actions.shape[1], batch_size) - real_inner_pi_us = torch.real(inner_pi_us) # (all_actions.shape[1], batch_size) - - # Max trick - max_inner_pi_u = torch.amax(real_inner_pi_us, axis=0) # (batch_size,) - diff = real_inner_pi_us - max_inner_pi_u # (all_actions.shape[1], batch_size) - - # Softmax distribution - pi_us = torch.exp(diff) + delta # (all_actions.shape[1], batch_size) - Z_x = torch.sum(pi_us, axis=0) # (batch_size,) - pis_response = pi_us / Z_x # (all_actions.shape[1], batch_size) - - # Compute log probabilities - log_pis = torch.log(pis_response) # (all_actions.shape[1], batch_size) - - # Compute expectation - expectation_u = torch.sum( - (costs + self.alpha * log_pis + self.discount_factor * V_x_prime_batch) * pis_response, axis=0 - ).reshape(1, -1) # (1, batch_size) - - # Compute mean squared error - squared_error = torch.pow(V_xs - expectation_u, 2) # (1, batch_size) - mean_squared_error = torch.mean(squared_error) # scalar - - return mean_squared_error - - def get_action_and_log_prob(self, x, sample_size=None, is_greedy=False): - """ - Compute the action given the current state. - - Parameters - ---------- - x : array_like - State of the system as a column vector. - sample_size : int or None, optional - How many actions to sample. None gives 1 sample. - is_greedy : bool, optional - If True, select the action with maximum probability greedily. - If False, sample actions based on the probability distribution. - - Returns - ------- - actions : array - Selected actions from the value iteration policy. - log_probabilities : array - Logarithm of the probabilities corresponding to the selected actions. - - Notes - ----- - This function computes the action to be taken given the current state `x`. - If `sample_size` is provided, it selects multiple actions based on the - policy distribution. If `is_greedy` is True, it selects the action with - the maximum probability greedily; otherwise, it samples actions according - to the probability distribution defined by the policy. - """ - - if sample_size is None: - sample_size = self.dynamics_model.u_column_dim - - pis_response = self.pis(x)[:, 0] - - if is_greedy: - selected_indices = torch.ones(sample_size, dtype=torch.int8) * torch.argmax(pis_response) - else: - selected_indices = torch.from_numpy( - np.random.choice(np.arange(len(pis_response)), size=sample_size, p=pis_response) - ) - - return (self.all_actions[0][selected_indices.long()], torch.log(pis_response[selected_indices.long()])) - - def get_action(self, x, sample_size=None, is_greedy=False): - """ - Compute the action given the current state. - - Parameters - ---------- - x : array_like - State of the system as a column vector. - sample_size : int or None, optional - How many actions to sample. None gives 1 sample. - is_greedy : bool, optional - If True, select the action with maximum probability greedily. - If False, sample actions based on the probability distribution. - - Returns - ------- - action : array - Selected action(s) from the value iteration policy. - - Notes - ----- - This function computes the action to be taken given the current state `x`. - If `sample_size` is provided, it selects multiple actions based on the - policy distribution. If `is_greedy` is True, it selects the action with - the maximum probability greedily; otherwise, it samples actions according - to the probability distribution defined by the policy. - """ - - return self.get_action_and_log_prob(x, sample_size, is_greedy)[0] - - def train( - self, - training_epochs, - batch_size=2**14, - batch_scale=1, - epsilon=1e-2, - gammas=[], - gamma_increment_amount=0.0, - how_often_to_chkpt=250, - ): - """ - Train the value iteration model. - - Parameters - ---------- - training_epochs : int - Number of epochs for which to train the model. - batch_size : int, optional - Sample of states for computing the value function weights. - batch_scale : int, optional - Scale factor that is multiplied by batch_size for computing the Bellman error. - epsilon : float, optional - End the training process if the Bellman error < epsilon. - gammas : list of float, optional - Array of gammas to try in case of iterating on the discounting factors. - gamma_increment_amount : float, optional - Amount by which to increment gamma until it reaches 0.99. If 0.0, no incrementing. - how_often_to_chkpt : int, optional - Number of training iterations to do before saving model weights and training data. - - Notes - ----- - This function updates the class parameters without returning anything. - After running this function, you can call `policy.get_action(x)` to get an action using the trained policy. - """ - - # Create all directories needed to save data - create_folder(f"{self.save_data_path}/training_data") - - # Save original gamma and set gamma to first in array - original_gamma = self.gamma - if len(gammas) > 0: - self.gamma = gammas[0] - self.discount_factor = self.gamma**self.dt - - # Compute initial Bellman error - BE = self.discrete_bellman_error(batch_size=batch_size * batch_scale).detach().numpy() - bellman_errors = [BE] - print(f"Initial Bellman error: {BE}") - - step = 0 - gamma_iteration_condition = self.gamma <= 0.99 or self.gamma == 1 - while gamma_iteration_condition: - print(f"gamma for iteration #{step + 1}: {self.gamma}") - self.discount_factor = self.gamma**self.dt - - for epoch in range(training_epochs): - # Get random batch of X and Phi_X from tensor training data - x_batch_indices = torch.from_numpy( - np.random.choice(self.dynamics_model.X.shape[1], batch_size, replace=False) - ) - x_batch = self.dynamics_model.X[:, x_batch_indices.long()] # (X.shape[0], batch_size) - phi_x_batch = self.dynamics_model.Phi_X[:, x_batch_indices.long()] # (dim_phi, batch_size) - - # Compute costs indexed by the action and the state - costs = torch.Tensor(self.cost(x_batch.T, self.all_actions.T)) # (all_actions.shape[1], batch_size) - - # Compute V(x')s - K_us = self.dynamics_model.K_(self.all_actions) # (all_actions.shape[1], phi_dim, phi_dim) - phi_x_prime_batch = torch.zeros((self.all_actions.shape[1], self.dynamics_model.phi_dim, batch_size)) - V_x_prime_batch = torch.zeros((self.all_actions.shape[1], batch_size)) - for action_index in range(phi_x_prime_batch.shape[0]): - phi_x_prime_hat_batch = K_us[action_index] @ phi_x_batch # (phi_dim, batch_size) - phi_x_prime_batch[action_index] = phi_x_prime_hat_batch - V_x_prime_batch[action_index] = self.V_phi_x(phi_x_prime_batch[action_index]) # (1, batch_size) - - # Compute policy distribution - inner_pi_us_values = -( - costs + self.discount_factor * V_x_prime_batch - ) # (all_actions.shape[1], batch_size) - inner_pi_us = inner_pi_us_values / self.alpha # (all_actions.shape[1], batch_size) - real_inner_pi_us = torch.real(inner_pi_us) # (all_actions.shape[1], batch_size) - - # Max trick - max_inner_pi_u = torch.amax(real_inner_pi_us, axis=0) # (batch_size,) - diff = real_inner_pi_us - max_inner_pi_u # (all_actions.shape[1], batch_size) - - # Softmax distribution - pi_us = torch.exp(diff) + delta # (all_actions.shape[1], batch_size) - Z_x = torch.sum(pi_us, axis=0) # (batch_size,) - pis_response = pi_us / Z_x # (all_actions.shape[1], batch_size) - - # Compute log pi - log_pis = torch.log(pis_response) # (all_actions.shape[1], batch_size) - - # Compute expectations - expectation_term_1 = torch.sum( - (costs + self.alpha * log_pis + self.discount_factor * V_x_prime_batch) * pis_response, dim=0 - ).reshape(1, -1) # (1, batch_size) - - # Optimize value function weights - if self.use_ols: - # OLS as in Lewis - self.value_function_weights = torch.linalg.lstsq(phi_x_batch.T, expectation_term_1.T).solution - else: - # Compute loss - loss = torch.pow(V_x_prime_batch - expectation_term_1, 2).mean() - - # Backpropogation for value function weights - self.value_function_optimizer.zero_grad() - loss.backward() - self.value_function_optimizer.step() - - # Recompute Bellman error - BE = self.discrete_bellman_error(batch_size=batch_size * batch_scale).detach().numpy() - bellman_errors.append(BE) - - # Print epoch number - print(f"Epoch number: {epoch + 1}") - - # Every so often, print out and save the model weights and bellman errors - if (epoch + 1) % how_often_to_chkpt == 0: - torch.save(self.value_function_weights, f"{self.save_data_path}/policy.pt") - torch.save(bellman_errors, f"{self.save_data_path}/training_data/bellman_errors.pt") - print(f"Bellman error at epoch {epoch + 1}: {BE}") - - if BE <= epsilon: - break - - step += 1 - - if len(gammas) == 0 and gamma_increment_amount == 0: - gamma_iteration_condition = False - break - - if self.gamma == 0.99: - break - - if len(gammas) > 0: - self.gamma = gammas[step] - else: - self.gamma += gamma_increment_amount - - if self.gamma > 0.99: - self.gamma = 0.99 - - gamma_iteration_condition = self.gamma <= 0.99 - - self.gamma = original_gamma - self.discount_factor = self.gamma**self.dt - - -def make_env(env_id, seed, idx, capture_video, run_name): - def thunk(): - env = gym.make(env_id) - env = gym.wrappers.RecordEpisodeStatistics(env) - if capture_video: - if idx == 0: - env = gym.wrappers.RecordVideo(env, f"videos/{run_name}") - env.seed(seed) - env.action_space.seed(seed) - env.observation_space.seed(seed) - return env - - return thunk - - -def main(): - args = ArgumentParser().parse_args() - run_name = f"{args.env_id}__{args.exp_name}__{args.seed}__{int(time.time())}" - - writer = SummaryWriter(f"runs/{run_name}") - writer.add_text( - "hyperparameters", - "|param|value|\n|-|-|\n%s" % ("\n".join([f"|{key}|{value}|" for key, value in vars(args).items()])), - ) - - device = torch.device("cuda" if torch.cuda.is_available() and args.cuda else "cpu") - - # TRY NOT TO MODIFY: seeding - random.seed(args.seed) - np.random.seed(args.seed) - torch.manual_seed(args.seed) - torch.backends.cudnn.deterministic = args.torch_deterministic - - envs = gym.vector.SyncVectorEnv([make_env(args.env_id, args.seed, 0, False, run_name)]) - - koopman_tensor = load_tensor(args.env_id, "path_based_tensor") - - try: - dt = envs.envs[0].dt - except Exception: - dt = None - - # Construct set of all possible actions - all_actions = torch.from_numpy( - np.linspace(start=envs.single_action_space.low, stop=envs.single_action_space.high, num=args.num_actions) - ).T - - value_function_weights = torch.tensor( - [ - [-333.7974], # 1 - [22.5883], # x - [-8.0066], # y - [-157.5718], # z - [267.9301], # x*x - [-80.5217], # x*y - [-27.1598], # x*z - [173.2158], # y*y - [6.2852], # y*z - [149.4211], # z*z - ] - ) - value_iteration_policy = DiscreteKoopmanValueIterationPolicy( - args=args, - gamma=args.gamma, - alpha=args.alpha, - dynamics_model=koopman_tensor, - all_actions=all_actions, - cost=envs.envs[0].vectorized_cost_fn, - use_ols=True, - learning_rate=args.lr, - dt=dt, - ) - value_iteration_policy.load_model( - value_function_weights=value_function_weights, trained_model_start_timestamp=None, chkpt_epoch_number=None - ) - - envs.single_observation_space.dtype = np.float64 - start_time = time.time() - - # TRY NOT TO MODIFY: start the game - obs = envs.reset() - for global_step in range(args.total_timesteps): - # ALGO LOGIC: put action logic here - actions = value_iteration_policy.get_action(torch.Tensor(obs).to(device)) - actions = actions.detach().cpu().numpy() - - # TRY NOT TO MODIFY: execute the game and log data. - next_obs, rewards, dones, infos = envs.step(actions) - - # TRY NOT TO MODIFY: record rewards for plotting purposes - for info in infos: - if "episode" in info.keys(): - print(f"global_step={global_step}, episodic_return={info['episode']['r']}") - writer.add_scalar("charts/episodic_return", info["episode"]["r"], global_step) - writer.add_scalar("charts/episodic_length", info["episode"]["l"], global_step) - break - - # TRY NOT TO MODIFY: CRUCIAL step easy to overlook - obs = next_obs - - # Write data - if global_step % 100 == 0: - sps = int(global_step / (time.time() - start_time)) - print("Steps per second (SPS):", sps) - writer.add_scalar("charts/SPS", sps, global_step) - - envs.close() - writer.close() - - -if __name__ == "__main__": - main() diff --git a/koopmanrl/koopman_observables.py b/koopmanrl/koopman_observables.py index 77fa7fe..a907d09 100644 --- a/koopmanrl/koopman_observables.py +++ b/koopmanrl/koopman_observables.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: LGPL-3.0-only +# Adapted from d3s (https://github.com/sklus/d3s), licensed under the GNU Lesser General Public License v3.0; +# see LICENSES/README.md. The rest of KoopmanRL is MIT-licensed (LICENSE.md). import copy import math diff --git a/koopmanrl/koopman_tensor/observables/numpy_observables.py b/koopmanrl/koopman_tensor/observables/numpy_observables.py index ff29293..b442d9e 100644 --- a/koopmanrl/koopman_tensor/observables/numpy_observables.py +++ b/koopmanrl/koopman_tensor/observables/numpy_observables.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: LGPL-3.0-only +# Adapted from d3s (https://github.com/sklus/d3s), licensed under the GNU Lesser General Public License v3.0; +# see LICENSES/README.md. The rest of KoopmanRL is MIT-licensed (LICENSE.md). import math import numpy as np diff --git a/koopmanrl/koopman_tensor/observables/torch_observables.py b/koopmanrl/koopman_tensor/observables/torch_observables.py index 77fa7fe..a907d09 100644 --- a/koopmanrl/koopman_tensor/observables/torch_observables.py +++ b/koopmanrl/koopman_tensor/observables/torch_observables.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: LGPL-3.0-only +# Adapted from d3s (https://github.com/sklus/d3s), licensed under the GNU Lesser General Public License v3.0; +# see LICENSES/README.md. The rest of KoopmanRL is MIT-licensed (LICENSE.md). import copy import math diff --git a/koopmanrl_utils/AGENTS.md b/koopmanrl_utils/AGENTS.md index f7c667d..20d7e1b 100644 --- a/koopmanrl_utils/AGENTS.md +++ b/koopmanrl_utils/AGENTS.md @@ -8,7 +8,7 @@ All scripts need to be run from the root of the repository, as modules: - `uv run -m koopmanrl_utils.` (the directory has no `__init__.py`; it is imported as a namespace package) -Most scripts have a `tap` parser and list their options with `--help`. `run_skvi_optimization.py` and `run_sakc_optimization.py` require `--storage_dir` and start four full hyperparameter optimizations as soon as they are given one; only `--help` is safe to run. Paths are never defaulted to a machine: `dataframe_creator.py` requires `--target_dir` and `--storage_dir`, and the `process_*` scripts `--root_dir` and `--output_dir`. `interpret_koopman.py` is an unfinished stub (it reads `args.skvi_stored_weights`, which its parser does not define; `--paths_to_files` takes a JSON file in the layout of the template `interpret_koopman.json`), and `plot_csv_from_tensorboards.py` defaults to run names of 2024. +Most scripts have a `tap` parser and list their options with `--help`. `run_skvi_optimization.py` and `run_sakc_optimization.py` require `--storage_dir` and start four full hyperparameter optimizations as soon as they are given one; only `--help` is safe to run. Paths are never defaulted to a machine: `dataframe_creator.py` requires `--target_dir` and `--storage_dir`, and the `process_*` scripts `--root_dir` and `--output_dir`. `interpret_koopman.py` is an unfinished stub (it reads `args.skvi_stored_weights`, which its parser does not define; `--paths_to_files` takes a JSON file in the layout of the template `interpret_koopman.json`). ## Directory Structure @@ -23,7 +23,6 @@ koopmanrl_utils/ ├── interpret_koopman.py # Unfinished stub for interpreting a stored Koopman tensor; does not run ├── koopman_prediction_validation.py # Held-out one-step and multi-step prediction error of the Koopman tensor ├── koopman_regressor_comparison.py # Held-out error of the Koopman tensor under 13 regression algorithms; TikZ figures -├── plot_csv_from_tensorboards.py # Ingests Tensorboard results and generates csv files ├── process_episodic_returns.py # Writes the pgfplots tables (.dat, or .csv) of the episodic returns from the JSON data frames ├── process_sakc_ablations.py # Writes the pgfplots tables of the SAKC ablation from the JSON data frames ├── process_skvi_ablations.py # Writes the pgfplots tables of the SKVI ablation from the JSON data frames diff --git a/koopmanrl_utils/plot_csv_from_tensorboards.py b/koopmanrl_utils/plot_csv_from_tensorboards.py deleted file mode 100644 index 73f73d4..0000000 --- a/koopmanrl_utils/plot_csv_from_tensorboards.py +++ /dev/null @@ -1,136 +0,0 @@ -# /// script -# requires-python = "== 3.10" -# dependencies = [ -# "os", -# "numpy", -# "pandas", -# "koopmanrl", -# "matplotlib", -# "typed-argument-parser", -# "tensorboard" -# ] -# /// - -import os - -import matplotlib.pyplot as plt -import numpy as np -import pandas as pd -from tap import Tap -from tensorboard.backend.event_processing.event_accumulator import EventAccumulator - -from koopmanrl.utils import create_folder - -plt.style.use("ggplot") - -# TODO: Debug, and make sure it runs as intended. - - -class ArgumentParser(Tap): - path: str = "./runs" # Directory in which the files are - files: list[str] = [ - "FluidFlow-v0__interpretability_discrete_value_iteration__1__1707412846", - "FluidFlow-v0__interpretability_discrete_value_iteration__1__1707413302", - "FluidFlow-v0__interpretability_discrete_value_iteration__1__1707413512", - "FluidFlow-v0__interpretability_discrete_value_iteration__1__1707413673", - ] # List of file names which are to be plotted - - -def collect_episodic_returns(tensorboard_file_directory: str, tensorboard_file_name: str): - """ - Collects episodic returns and corresponding environment steps from a TensorBoard file. - - Parameters: - - tensorboard_file_directory (str): The directory containing the TensorBoard file. - - tensorboard_file_name (str): The name of the TensorBoard file. - - Returns: - Tuple[List[float], List[int]]: A tuple containing two lists - - 1. episodic_returns (List[float]): List of episodic return values. - 2. steps (List[int]): List of corresponding step values. - """ - - summary_iterator = EventAccumulator(os.path.join(tensorboard_file_directory, tensorboard_file_name)).Reload() - - scalar_name = "charts/episodic_return" - - steps = [e.step for e in summary_iterator.Scalars(scalar_name)] - episodic_returns = [e.value for e in summary_iterator.Scalars(scalar_name)] - - return episodic_returns, steps - - -def main(): - args = ArgumentParser().parse_args() - - # Initialize Pandas dataframe - df = pd.DataFrame() - - # For capturing file seeds - file_seeds = [] - - # For each tensorboard file in the pre-defined list above, - for i, file_name in enumerate(args.files): - # Collect seed number and environment id - summary_iterator = EventAccumulator(os.path.join(args.path, file_name)).Reload() - hyperparams = str(summary_iterator.Tensors("hyperparameters/text_summary")[0]).split("|") - # file_seed = i+1 #! Hardcoded seed value for testing - file_seed = hyperparams[hyperparams.index("seed") + 1] - file_seeds.append(f"Seed {file_seed}") - env_id = hyperparams[hyperparams.index("env_id") + 1] - - # Collect the episodic returns and environment steps - # and save the data into numpy files for use in other places - episodic_returns, steps = collect_episodic_returns(args.path, file_name) - create_folder(f"./analysis/{file_name}") - np.save(f"./analysis/{file_name}/episodic_returns.npy", episodic_returns) - np.save(f"./analysis/{file_name}/steps.npy", steps) - - # Print information about the data - print(f"{file_name}'s episodic returns length: {len(episodic_returns)}") - print(f"{file_name}'s steps length: {len(steps)}") - print(f"{file_name}'s steps per episode: {steps[0] + 1}") - - # Compute mean episodic return - print(f"{file_name}'s mean episodic return: {np.mean(episodic_returns)}\n") - - # Store information in dataframe - try: - # Try to write to all entries at once - df.loc[steps, f"Seed {file_seed}"] = episodic_returns - except Exception: - # We end up here if there is an error accessing the index above - # Loop through the steps and returns and add them to the dataframe - for step, episodic_return in zip(steps, episodic_returns): - df.loc[step, f"Seed {file_seed}"] = episodic_return - - # Save dataframe to local CSV - df.to_csv(f"./analysis/{env_id}_performance.csv") - - # Plot the path data from the dataframe - plt.plot(df) - plt.title("Episodic Returns For Various Runs") - plt.xlabel("Step in Environment") - plt.ylabel("Episodic Return") - plt.legend(file_seeds) - plt.show() - - # Plot the average and standard deviation with two sigma - mean_episodic_returns = df.mean(axis=1) - std_episodic_returns = df.std(axis=1) - two_sigma_upper = mean_episodic_returns + 2 * std_episodic_returns - two_sigma_lower = mean_episodic_returns - 2 * std_episodic_returns - - plt.plot(mean_episodic_returns, label="Mean Episodic Return") - plt.fill_between( - mean_episodic_returns.index, two_sigma_lower, two_sigma_upper, color="gray", alpha=0.3, label="Two Sigma" - ) - plt.title("Average Episodic Returns with Two Sigma Bounds") - plt.xlabel("Step in Environment") - plt.ylabel("Episodic Return") - plt.legend() - plt.show() - - -if __name__ == "__main__": - main() diff --git a/pyproject.toml b/pyproject.toml index 0910082..86454f0 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -3,15 +3,16 @@ name = "koopmanrl" description = "KoopmanRL is a reinforcement learning (RL) package designed around the two Koopman-Assisted RL (KARL) algorithms, Soft Koopman Value Iteration, and Soft Koopman Actor-Critic. It provides the utilities to build upon parts of its algorithms by either using only the Koopman tensor itself, or only components of the two KARL algorithms. In addition it provides utilities for automatic hyperparameter tuning of KARL algorithms, as well as 4 environments rooted in the control literature. v.1.0.0 is the accompanying repository to 'KARL: Koopman-Assisted Reinforcement Learning'." authors = [ {name = "Ludger Paehler", email = "paehlerludger@gmail.com"}, - {name = "Preston Rozwood", email = "pwr36@cornell.edu"}, + {name = "Preston Rozwood", email = "pwr@cornell.edu"}, {name = "Edward J. Mehrez", email = "ejm322@cornell.edu"}, ] requires-python = "==3.10.*" version = "1.0.0" -license = "MIT" +license = "MIT AND LGPL-3.0-only" +license-files = ["LICENSE.md", "LICENSES/*.txt"] readme = "README.md" classifiers = [ - "Development Status :: 2 - Pre-Alpha", + "Development Status :: 3 - Alpha", "Intended Audience :: Science/Research", "Operating System :: MacOS", "Operating System :: Unix", @@ -24,7 +25,6 @@ classifiers = [ "Topic :: Scientific/Engineering :: Physics", "Topic :: Software Development :: Libraries", ] -repository = "https://github.com/dynamicslab/KoopmanRL" keywords = [ "Reinforcement Learning", "Koopman Operator", @@ -48,6 +48,12 @@ dependencies = [ "typed-argument-parser>=1.11.0", ] + +[project.urls] +Repository = "https://github.com/dynamicslab/KoopmanRL" +Documentation = "https://dynamicslab.github.io/KoopmanRL/" +Paper = "https://arxiv.org/abs/2403.02290" + [tool.ruff] line-length = 120 target-version = "py310" @@ -73,6 +79,10 @@ testpaths = ["tests"] python_files = ["test_*.py"] python_classes = ["Test*"] python_functions = ["test_*"] +markers = [ + "hpo_full: hyperparameter-optimization smoke tests on the environments other than the linear system (deselected by default)", +] +addopts = "-m 'not hpo_full'" [tool.hatch.build.targets.sdist] include = [ diff --git a/tests/AGENTS.md b/tests/AGENTS.md index 4a01e41..22e041d 100644 --- a/tests/AGENTS.md +++ b/tests/AGENTS.md @@ -7,10 +7,11 @@ The pytest suite of KoopmanRL. `pyproject.toml` sets `testpaths = ["tests"]` and ```bash uv run pytest tests/test_.py # the tests of what you touch uv run pytest tests/test_.py -k # a single test -uv run pytest --deselect tests/test_hparam_opt.py # everything else (several minutes) +uv run pytest # everything (several minutes), HPO smoke tests on the linear system only +uv run pytest tests/test_hparam_opt.py -m hpo_full # HPO smoke tests of the other three environments ``` -There are no pytest markers, so there is no fast/slow split to select with `-m`. +`pyproject.toml` registers one marker, `hpo_full`, and deselects it by default (`addopts = "-m 'not hpo_full'"`). Pass `-m hpo_full` to run only those tests, or `-m ""` to run everything. ## Directory Structure @@ -19,7 +20,7 @@ tests/ ├── AGENTS.md # This file ├── utils.py # run_module: `uv run -m ` from the repository root, 300 s timeout ├── test_rl.py # Short runs of LQR, both SACs, SKVI and SAKC on the four environments; same-seed runs repeat -├── test_hparam_opt.py # One-sample runs of skvi_optuna_opt and sakc_optuna_opt on the four environments (Ray Tune) +├── test_hparam_opt.py # Smoke tests of skvi_optuna_opt and sakc_optuna_opt (one single-sample Ray Tune study each) ├── test_koopman_tensor_regressors.py # koopman_tensor/regressors.py and the four copies of the tensor class ├── test_koopman_prediction_validation.py # koopmanrl_utils/koopman_prediction_validation.py ├── test_koopman_regressor_comparison.py # koopmanrl_utils/koopman_regressor_comparison.py @@ -39,8 +40,6 @@ tests/ * Most tests run a module as a subprocess through `tests/utils.py::run_module`, with the repository root as working directory. `test_rl.py` therefore writes `runs/` and `saved_models/` into the repository root (both gitignored). * The reproducibility tests in `test_rl.py` run each module twice in separate `tmp_path` folders with `sys.executable` and compare the logged returns; they rely on the `gelsd` solves of SKVI and SAKC (see `koopmanrl/AGENTS.md`). -* `test_hparam_opt.py` needs care: - * It asks Ray for `--cpu_cores_per_trial=16`, so on a machine with fewer than 16 cores no trial is ever scheduled and the test runs into its 600 s timeout. - * It sets `RAY_TMPDIR` to pytest's `tmp_path`. On macOS that path makes Ray's AF_UNIX socket path exceed the OS limit (about 104 bytes) and the test fails at startup. Deselect it there. +* `test_hparam_opt.py` is a smoke test, not a hyperparameter search: each test runs one study with `--num_samples=1`, `--max_concurrent=1` and `--cpu_cores_per_trial=1`, and checks that the best configuration is written to `--storage_dir`. By default only `LinearSystem-v0` runs (two tests, about 1.5 minutes); the other environments carry the `hpo_full` marker. Ray's temporary directory is a short directory under `/tmp`, because pytest's `tmp_path` makes Ray's AF_UNIX socket paths too long on macOS. * `test_workflow_helpers.py` imports `workflow/helpers.py` by path. Snakemake is not in the project environment, so the rule files themselves are checked with a dry run (`uvx --python 3.12 snakemake -n `), not by pytest. * CI does not run pytest; run the tests of the code you change locally before opening a pull request. diff --git a/tests/test_hparam_opt.py b/tests/test_hparam_opt.py index a66ba9c..6238068 100644 --- a/tests/test_hparam_opt.py +++ b/tests/test_hparam_opt.py @@ -1,52 +1,70 @@ +"""Smoke tests of the hyperparameter optimizations `koopmanrl.sakc_optuna_opt` and `koopmanrl.skvi_optuna_opt`. + +Each test runs one Ray Tune study with a single sample on one CPU core and checks that the best configuration is +written. By default only the linear system is run, one study per algorithm; the other three environments are marked +`hpo_full` and deselected by `addopts` in pyproject.toml. Run them with `uv run pytest tests/test_hparam_opt.py -m +hpo_full`, or everything with `-m ""`. +""" + import os +import shutil +import tempfile import pytest from tests.utils import run_module TOTAL_TIMESTEPS = 1000 -ENVS = ["LinearSystem-v0", "FluidFlow-v0", "Lorenz-v0", "DoubleWell-v0"] +FULL = pytest.mark.hpo_full +ENVS = [ + "LinearSystem-v0", + pytest.param("FluidFlow-v0", marks=FULL), + pytest.param("Lorenz-v0", marks=FULL), + pytest.param("DoubleWell-v0", marks=FULL), +] -@pytest.mark.parametrize("env_id", ENVS) -def test_sakc_opt(env_id, tmp_path): +@pytest.fixture +def ray_env(): + """Environment for the study, with Ray's temporary directory at a short path. + + Ray places AF_UNIX sockets under RAY_TMPDIR, whose paths may not exceed about 104 bytes on macOS; pytest's tmp_path + is too long for that, so a short directory under /tmp is used where it exists. + """ + ray_tmpdir = tempfile.mkdtemp(prefix="ray", dir="/tmp" if os.path.isdir("/tmp") else None) env = os.environ.copy() - env["RAY_TMPDIR"] = str(tmp_path) - result = run_module( - "koopmanrl.sakc_optuna_opt", + env["RAY_TMPDIR"] = ray_tmpdir + yield env + shutil.rmtree(ray_tmpdir, ignore_errors=True) + + +def run_study(module, env_id, storage_dir, env): + return run_module( + module, [ "--env_id", env_id, "--num_samples=1", + "--max_concurrent=1", f"--total_timesteps={TOTAL_TIMESTEPS}", - "--cpu_cores_per_trial=16", - f"--storage_dir={tmp_path}", + "--cpu_cores_per_trial=1", + f"--storage_dir={storage_dir}", "--output_file=best_config", ], timeout=600, env=env, ) + + +@pytest.mark.parametrize("env_id", ENVS) +def test_sakc_opt(env_id, tmp_path, ray_env): + result = run_study("koopmanrl.sakc_optuna_opt", env_id, tmp_path, ray_env) assert result.returncode == 0, result.stderr assert (tmp_path / "best_config.json").is_file() @pytest.mark.parametrize("env_id", ENVS) -def test_skvi_opt(env_id, tmp_path): - env = os.environ.copy() - env["RAY_TMPDIR"] = str(tmp_path) - result = run_module( - "koopmanrl.skvi_optuna_opt", - [ - "--env_id", - env_id, - "--num_samples=1", - f"--total_timesteps={TOTAL_TIMESTEPS}", - "--cpu_cores_per_trial=16", - f"--storage_dir={tmp_path}", - "--output_file=best_config", - ], - timeout=600, - env=env, - ) +def test_skvi_opt(env_id, tmp_path, ray_env): + result = run_study("koopmanrl.skvi_optuna_opt", env_id, tmp_path, ray_env) assert result.returncode == 0, result.stderr assert (tmp_path / "best_config.json").is_file()