# mlip-arena
MLIP Arena is an open-source platform for benchmarking machine learning interatomic potentials (MLIPs). The platform provides a unified interface for users to evaluate the performance of their models on a variety of tasks, including single-point density functional theory calculations and molecular dynamics simulations. The platform is designed to be extensible, allowing users to contribute new models, benchmarks, and training data to the platform.
## Contribute
### Add new MLIP models
If you have pretrained MLIP models that you would like to contribute to the MLIP Arena and show benchmark in real-time, please follow these steps:
1. Create a new [Hugging Face Model](https://huggingface.co/new) repository and upload the model file.
2. Follow the template to code the I/O interface for your model, and upload the script along with metadata to the MLIP Arena [here]().
3. CPU benchmarking will be performed automatically. Due to the limited amount GPU compute, if you would like to be considered for GPU benchmarking, please create a pull request to demonstrate the offline performance of your model (published paper or preprint). We will review and select the models to be benchmarked on GPU.
### Add new benchmarks
#### Molecular dynamics calculations
- [ ] [MD17](http://www.sgdml.org/#datasets)
- [ ] [MD22](http://www.sgdml.org/#datasets)
#### Single-point density functional theory calculations
- [ ] MPTrj
- [ ] QM9
- [ ] [Alexandria](https://alexandria.icams.rub.de/)
### Add new training datasets
[Hugging Face Auto-Train](https://huggingface.co/docs/hub/webhooks-guide-auto-retrain)
Raw data
{
"_id": null,
"home_page": null,
"name": "mlip-arena",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "pytorch, machine-learning-interatomic-potentials, huggingface, deep-learning, graph-neural-networks",
"author": null,
"author_email": "Yuan Chiang <cyrusyc@lbl.gov>",
"download_url": "https://files.pythonhosted.org/packages/48/34/f2e066d7b3a4a5b8547e21e7f51db3b711369cc34c25754d6107333c45b8/mlip_arena-0.0.1.tar.gz",
"platform": null,
"description": "# mlip-arena\n\nMLIP Arena is an open-source platform for benchmarking machine learning interatomic potentials (MLIPs). The platform provides a unified interface for users to evaluate the performance of their models on a variety of tasks, including single-point density functional theory calculations and molecular dynamics simulations. The platform is designed to be extensible, allowing users to contribute new models, benchmarks, and training data to the platform.\n\n## Contribute\n\n### Add new MLIP models\n\nIf you have pretrained MLIP models that you would like to contribute to the MLIP Arena and show benchmark in real-time, please follow these steps:\n\n1. Create a new [Hugging Face Model](https://huggingface.co/new) repository and upload the model file.\n2. Follow the template to code the I/O interface for your model, and upload the script along with metadata to the MLIP Arena [here]().\n3. CPU benchmarking will be performed automatically. Due to the limited amount GPU compute, if you would like to be considered for GPU benchmarking, please create a pull request to demonstrate the offline performance of your model (published paper or preprint). We will review and select the models to be benchmarked on GPU.\n\n### Add new benchmarks\n\n#### Molecular dynamics calculations\n\n- [ ] [MD17](http://www.sgdml.org/#datasets)\n- [ ] [MD22](http://www.sgdml.org/#datasets)\n\n\n#### Single-point density functional theory calculations\n\n- [ ] MPTrj\n- [ ] QM9\n- [ ] [Alexandria](https://alexandria.icams.rub.de/)\n\n### Add new training datasets\n\n[Hugging Face Auto-Train](https://huggingface.co/docs/hub/webhooks-guide-auto-retrain)\n\n\n\n",
"bugtrack_url": null,
"license": null,
"summary": null,
"version": "0.0.1",
"project_urls": {
"Homepage": "https://github.com/atomind-ai/mlip-arena",
"Issues": "https://github.com/atomind-ai/mlip-arena/issues"
},
"split_keywords": [
"pytorch",
" machine-learning-interatomic-potentials",
" huggingface",
" deep-learning",
" graph-neural-networks"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "4153fe7325308c20168ce3381057c07cc6ddd4cdeb219bdfa9da7bbe78ed6c59",
"md5": "9094de5432549cdc8076fe1c6e780539",
"sha256": "6edbe9c4557dcad008e7a94f27f08e33fe8d56c3b8cd2ab4cf0ec5bda4fa8832"
},
"downloads": -1,
"filename": "mlip_arena-0.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "9094de5432549cdc8076fe1c6e780539",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 5324,
"upload_time": "2024-03-25T21:42:58",
"upload_time_iso_8601": "2024-03-25T21:42:58.086140Z",
"url": "https://files.pythonhosted.org/packages/41/53/fe7325308c20168ce3381057c07cc6ddd4cdeb219bdfa9da7bbe78ed6c59/mlip_arena-0.0.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4834f2e066d7b3a4a5b8547e21e7f51db3b711369cc34c25754d6107333c45b8",
"md5": "cd970731477e412f03c85a1b21ff4cb3",
"sha256": "66638f9c2105ce81bec7c1fee1a5925f40412e9eb0a72b8dd03de5f6d223c4ee"
},
"downloads": -1,
"filename": "mlip_arena-0.0.1.tar.gz",
"has_sig": false,
"md5_digest": "cd970731477e412f03c85a1b21ff4cb3",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 3236,
"upload_time": "2024-03-25T21:42:59",
"upload_time_iso_8601": "2024-03-25T21:42:59.855476Z",
"url": "https://files.pythonhosted.org/packages/48/34/f2e066d7b3a4a5b8547e21e7f51db3b711369cc34c25754d6107333c45b8/mlip_arena-0.0.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-03-25 21:42:59",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "atomind-ai",
"github_project": "mlip-arena",
"github_not_found": true,
"lcname": "mlip-arena"
}
JSON
