{"id":31232883,"url":"https://github.com/biogeek/hackathon_indabax_2025_mlip","last_synced_at":"2025-09-22T13:01:09.339Z","repository":{"id":303562928,"uuid":"1013774790","full_name":"BioGeek/hackathon_IndabaX_2025_mlip","owner":"BioGeek","description":null,"archived":false,"fork":false,"pushed_at":"2025-07-08T08:24:16.000Z","size":1197,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-07-08T09:29:00.914Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/BioGeek.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-07-04T12:50:30.000Z","updated_at":"2025-07-08T08:24:19.000Z","dependencies_parsed_at":"2025-07-08T09:30:49.569Z","dependency_job_id":"d52c8354-0e8d-4237-a80a-5221134cd655","html_url":"https://github.com/BioGeek/hackathon_IndabaX_2025_mlip","commit_stats":null,"previous_names":["biogeek/hackathon_indabax_2025_mlip"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/BioGeek/hackathon_IndabaX_2025_mlip","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/BioGeek%2Fhackathon_IndabaX_2025_mlip","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/BioGeek%2Fhackathon_IndabaX_2025_mlip/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/BioGeek%2Fhackathon_IndabaX_2025_mlip/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/BioGeek%2Fhackathon_IndabaX_2025_mlip/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/BioGeek","download_url":"https://codeload.github.com/BioGeek/hackathon_IndabaX_2025_mlip/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/BioGeek%2Fhackathon_IndabaX_2025_mlip/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":276407032,"owners_count":25636974,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-22T02:00:08.972Z","response_time":79,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-09-22T13:00:23.148Z","updated_at":"2025-09-22T13:01:09.310Z","avatar_url":"https://github.com/BioGeek.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Hackathon Challenge Description\n\nIn this challenge, we train a Machine Learning Interatomic Potential (MLIP)\nmodel using the [mlip](https://github.com/instadeepai/mlip) library.\n\nThe model will be trained on the dataset located in `train.xyz`. This dataset consists\nof 500 conformations of a molecule called 3-(benzyloxy)pyridin-2-amine (abbreviated \nas 3BPA) sampled with Molecular Dynamics at a temperature of 300 Kelvin.\n\nFor a detailed tutorial on how to train an MLIP model on such a dataset, see\n[this](https://github.com/instadeepai/mlip/blob/main/tutorials/model_training_tutorial.ipynb)\nJupyter notebook provided with the mlip library. As explained in the notebook, one can\nthen save a final model in a zip archive format supported by the library. Also\nsee the deep-dive tutorial \non [model training](https://instadeepai.github.io/mlip/user_guide/training.html) and\non the [models](https://instadeepai.github.io/mlip/user_guide/models.html) \nthemselves to understand how to adapt the hyperparameters of the training\nprocess and the models, respectively.\n\nThe model will be tested on its ability to predict the energies of new conformations\nof the same molecule. However, to test the generalization\ncapabilities of the model, these conformations are sampled at higher temperature,\ni.e., 1200 Kelvin. The test conformations are located in the\n`test_public.xyz` file. You can predict energies for them with a model saved in the\nzip format with the mlip library's batched inference functionality, described\n[here](https://instadeepai.github.io/mlip/user_guide/simulations.html#batched-inference)\nin the mlip documentation or explained in section 2 of \n[mlip's simulation tutorial](https://github.com/instadeepai/mlip/blob/main/tutorials/simulation_tutorial.ipynb).\nThe target energies are located in the `test_public.csv` file. The metric the \npredictions will be scored on is root-mean-square error (RMSE).\n\nThe private test set for which to submit predictions is located in the\n`test_private.xyz` file. It contains more conformations, also sampled at the higher\ntemperature.\n\nBuild the Docker image with the following command:\n\n```bash\ndocker build -t mlip-hackathon .\n``` \n\nThen, run the container with the following command:\n\n```bash\ndocker run -p 8888:8888 --gpus all -v \"$(pwd)\":/app mlip-hackathon\n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbiogeek%2Fhackathon_indabax_2025_mlip","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbiogeek%2Fhackathon_indabax_2025_mlip","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbiogeek%2Fhackathon_indabax_2025_mlip/lists"}