| Name | matbench-discovery JSON |
| Version |
1.1.1
JSON |
| download |
| home_page | |
| Summary | A benchmark for machine learning energy models on inorganic crystal stability prediction from unrelaxed structures |
| upload_time | 2024-01-28 14:19:42 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | >=3.11 |
| license | MIT License Copyright (c) 2022 Janosh Riebesell 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. |
| keywords |
bayesian optimization
convex hull
high-throughput search
inorganic crystal stability
interatomic potential
machine learning
materials discovery
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
<h1 align="center">
<img src="https://github.com/janosh/matbench-discovery/raw/main/site/static/favicon.svg" alt="Logo" width="60px"><br>
Matbench Discovery
</h1>
<h4 align="center" class="toc-exclude">
[](https://arxiv.org/abs/2308.14920)
[](https://github.com/janosh/matbench-discovery/actions/workflows/test.yml)
[](https://github.com/janosh/matbench-discovery/actions/workflows/gh-pages.yml)
[](https://python.org/downloads)
[](https://pypi.org/project/matbench-discovery?logo=pypi&logoColor=white)
</h4>
> TL;DR: We benchmark ML models on crystal stability prediction from unrelaxed structures finding universal interatomic potentials (UIP) like [CHGNet](https://github.com/CederGroupHub/chgnet), [MACE](https://github.com/ACEsuit/mace) and [M3GNet](https://github.com/materialsvirtuallab/m3gnet) to be highly accurate, robust across chemistries and ready for production use in high-throughput materials discovery.
Matbench Discovery is an [interactive leaderboard](https://janosh.github.io/matbench-discovery/models) and associated [PyPI package](https://pypi.org/project/matbench-discovery) which together make it easy to rank ML energy models on a task designed to simulate a high-throughput discovery campaign for new stable inorganic crystals.
We've tested <slot name="model-count" />models covering multiple methodologies ranging from random forests with structure fingerprints to graph neural networks, from one-shot predictors to iterative Bayesian optimizers and interatomic potential relaxers.
<slot name="best-report" />
Our results show that ML models have become robust enough to deploy them as triaging steps to more effectively allocate compute in high-throughput DFT relaxations. This work provides valuable insights for anyone looking to build large-scale materials databases.
<slot name="metrics-table" />
We welcome contributions that add new models to the leaderboard through GitHub PRs. See the [contributing guide](https://janosh.github.io/matbench-discovery/contribute) for details.
If you're interested in joining this work, feel free to [open a GitHub discussion](https://github.com/janosh/matbench-discovery/discussions) or [send an email](mailto:janosh.riebesell@gmail.gov?subject=Collaborate%20on%20Matbench%20Discovery).
For detailed results and analysis, check out the [preprint](https://janosh.github.io/matbench-discovery/preprint).
Raw data
{
"_id": null,
"home_page": "",
"name": "matbench-discovery",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.11",
"maintainer_email": "",
"keywords": "Bayesian optimization,convex hull,high-throughput search,inorganic crystal stability,interatomic potential,machine learning,materials discovery",
"author": "",
"author_email": "Janosh Riebesell <janosh.riebesell@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/fe/0d/f10ee05815ba3ae0964a74478b0463ae702bd04d95be322a214642be8f1f/matbench-discovery-1.1.1.tar.gz",
"platform": null,
"description": "<h1 align=\"center\">\n <img src=\"https://github.com/janosh/matbench-discovery/raw/main/site/static/favicon.svg\" alt=\"Logo\" width=\"60px\"><br>\n Matbench Discovery\n</h1>\n\n<h4 align=\"center\" class=\"toc-exclude\">\n\n[](https://arxiv.org/abs/2308.14920)\n[](https://github.com/janosh/matbench-discovery/actions/workflows/test.yml)\n[](https://github.com/janosh/matbench-discovery/actions/workflows/gh-pages.yml)\n[](https://python.org/downloads)\n[](https://pypi.org/project/matbench-discovery?logo=pypi&logoColor=white)\n\n</h4>\n\n> TL;DR: We benchmark ML models on crystal stability prediction from unrelaxed structures finding universal interatomic potentials (UIP) like [CHGNet](https://github.com/CederGroupHub/chgnet), [MACE](https://github.com/ACEsuit/mace) and [M3GNet](https://github.com/materialsvirtuallab/m3gnet) to be highly accurate, robust across chemistries and ready for production use in high-throughput materials discovery.\n\nMatbench Discovery is an [interactive leaderboard](https://janosh.github.io/matbench-discovery/models) and associated [PyPI package](https://pypi.org/project/matbench-discovery) which together make it easy to rank ML energy models on a task designed to simulate a high-throughput discovery campaign for new stable inorganic crystals.\n\nWe've tested <slot name=\"model-count\" />models covering multiple methodologies ranging from random forests with structure fingerprints to graph neural networks, from one-shot predictors to iterative Bayesian optimizers and interatomic potential relaxers.\n\n<slot name=\"best-report\" />\n\nOur results show that ML models have become robust enough to deploy them as triaging steps to more effectively allocate compute in high-throughput DFT relaxations. This work provides valuable insights for anyone looking to build large-scale materials databases.\n\n<slot name=\"metrics-table\" />\n\nWe welcome contributions that add new models to the leaderboard through GitHub PRs. See the [contributing guide](https://janosh.github.io/matbench-discovery/contribute) for details.\n\nIf you're interested in joining this work, feel free to [open a GitHub discussion](https://github.com/janosh/matbench-discovery/discussions) or [send an email](mailto:janosh.riebesell@gmail.gov?subject=Collaborate%20on%20Matbench%20Discovery).\n\nFor detailed results and analysis, check out the [preprint](https://janosh.github.io/matbench-discovery/preprint).\n",
"bugtrack_url": null,
"license": "MIT License Copyright (c) 2022 Janosh Riebesell 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. ",
"summary": "A benchmark for machine learning energy models on inorganic crystal stability prediction from unrelaxed structures",
"version": "1.1.1",
"project_urls": {
"Homepage": "https://janosh.github.io/matbench-discovery",
"Package": "https://pypi.org/project/matbench-discovery",
"Repo": "https://github.com/janosh/matbench-discovery"
},
"split_keywords": [
"bayesian optimization",
"convex hull",
"high-throughput search",
"inorganic crystal stability",
"interatomic potential",
"machine learning",
"materials discovery"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "cc379171cb5c9479b01218bcb71e63ef5d7ee2599ad38de0d9f564e3b582f4da",
"md5": "552a0a7a3d7ae7a43012ceec1ec62724",
"sha256": "5468e0e82e84af08b21defad6f8dce9a734e070e1063710a9eb3909aa5c5fd50"
},
"downloads": -1,
"filename": "matbench_discovery-1.1.1-py2.py3-none-any.whl",
"has_sig": false,
"md5_digest": "552a0a7a3d7ae7a43012ceec1ec62724",
"packagetype": "bdist_wheel",
"python_version": "py2.py3",
"requires_python": ">=3.11",
"size": 33348,
"upload_time": "2024-01-28T14:19:40",
"upload_time_iso_8601": "2024-01-28T14:19:40.514757Z",
"url": "https://files.pythonhosted.org/packages/cc/37/9171cb5c9479b01218bcb71e63ef5d7ee2599ad38de0d9f564e3b582f4da/matbench_discovery-1.1.1-py2.py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fe0df10ee05815ba3ae0964a74478b0463ae702bd04d95be322a214642be8f1f",
"md5": "4fd1810fd4cf5cc02b5fb73668f131bc",
"sha256": "b801ed8be776c123bf296d34e95b013428de170e3e0ede2f0309a03f7c954ece"
},
"downloads": -1,
"filename": "matbench-discovery-1.1.1.tar.gz",
"has_sig": false,
"md5_digest": "4fd1810fd4cf5cc02b5fb73668f131bc",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.11",
"size": 40581,
"upload_time": "2024-01-28T14:19:42",
"upload_time_iso_8601": "2024-01-28T14:19:42.274878Z",
"url": "https://files.pythonhosted.org/packages/fe/0d/f10ee05815ba3ae0964a74478b0463ae702bd04d95be322a214642be8f1f/matbench-discovery-1.1.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-01-28 14:19:42",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "janosh",
"github_project": "matbench-discovery",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "matbench-discovery"
}
