{"id":13778431,"url":"https://github.com/herilalaina/mosaic_ml","last_synced_at":"2025-05-11T11:36:10.988Z","repository":{"id":98290018,"uuid":"133595063","full_name":"herilalaina/mosaic_ml","owner":"herilalaina","description":"AutoML with MCTS","archived":false,"fork":false,"pushed_at":"2022-05-04T08:37:06.000Z","size":322,"stargazers_count":14,"open_issues_count":1,"forks_count":6,"subscribers_count":4,"default_branch":"master","last_synced_at":"2024-08-03T18:12:50.798Z","etag":null,"topics":["automl","hyperparameter-optimization","pipeline-configuration"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/herilalaina.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}},"created_at":"2018-05-16T01:48:41.000Z","updated_at":"2024-01-23T20:03:24.000Z","dependencies_parsed_at":null,"dependency_job_id":"f2b2c9b4-09e5-4dbd-be7d-85b221e2e238","html_url":"https://github.com/herilalaina/mosaic_ml","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/herilalaina%2Fmosaic_ml","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/herilalaina%2Fmosaic_ml/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/herilalaina%2Fmosaic_ml/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/herilalaina%2Fmosaic_ml/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/herilalaina","download_url":"https://codeload.github.com/herilalaina/mosaic_ml/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":225048929,"owners_count":17412893,"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","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":["automl","hyperparameter-optimization","pipeline-configuration"],"created_at":"2024-08-03T18:00:53.769Z","updated_at":"2024-11-17T14:30:26.564Z","avatar_url":"https://github.com/herilalaina.png","language":"Python","funding_links":[],"categories":["2019"],"sub_categories":[],"readme":"# Automated Machine Learning with MCTS\n\n[![Build Status](https://api.travis-ci.org/herilalaina/mosaic_ml.svg?branch=master)](https://travis-ci.org/herilalaina/mosaic_ml)\n\nMosaic ML is a Python library for machine learning pipeline configuration\nusing Monte Carlo Tree Search.\n\nThe original paper can be found here: [https://www.ijcai.org/Proceedings/2019/457](https://www.ijcai.org/Proceedings/2019/457)\n\nAuthors: Herilalaina Rakotoarison, Marc Schoenauer and Michèle Sebag\n\n\n### Installation\n\n**Requirements**:\n* Python (3.5 or higher)\n* Numy\n* Cython\n* scipy\n* Mosaic (https://github.com/herilalaina/mosaic)\n\n**Installation**:\n```bash\npip install cython numpy scipy pytest\nsudo apt-get install build-essential swig\npip install git+https://github.com/herilalaina/mosaic@0.1\npip install git+https://github.com/herilalaina/mosaic_ml\n```\n\n### Usage\nThe entry script is ``python examples/run_mosaic_ml.py -h``.\n\n```\n--openml-task-id OPENML_TASK_ID\n                      OpenML Task ID (default 252)\n--overall-time-budget OVERALL_TIME_BUDGET\n                      Overall time budget in seconds (default 360)\n--eval-time-budget EVAL_TIME_BUDGET\n                      Time budget for each machine learning evaluation\n                      (default 100)\n--memory-limit MEMORY_LIMIT\n                      RAM Memory limit (default 3034)\n--seed SEED           Seed for reproducibility (default 42)\n--nb-init-metalearning NB_INIT_METALEARNING\n                      Number of initial configurations from Auto-Sklearn\n                      (default 25)\n--ensemble-size ENSEMBLE_SIZE\n                      Size of ensemble set (default 50)\n```\n\n\n**Mosaic ML** has three different components:\n* *vanilla*: MCTS for algorithm selection and Bayesian Optimization for hyperparameter tuning\n\n```bash\npython examples/run_mosaic_ml.py --nb-init-metalearning 0 --ensemble-size 1\n```\n\n* *metalearning*: initialize with a set of configurations fetched from [Auto-Sklearn](https://automl.github.io/auto-sklearn/master/index.html) then apply *vanilla setting*\n\n```bash\npython examples/run_mosaic_ml.py --nb-init-metalearning 25 --ensemble-size 1\n```\n\n* *ensemble (with metalearning)*: add an ensemble selection method ([Caruana et al, 04](https://www.cs.cornell.edu/~caruana/ctp/ct.papers/caruana.icml04.icdm06long.pdf)) in the top of the *metalearning* setting\n\n\n```bash\npython examples/run_mosaic_ml.py --nb-init-metalearning 25 --ensemble-size 50\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fherilalaina%2Fmosaic_ml","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fherilalaina%2Fmosaic_ml","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fherilalaina%2Fmosaic_ml/lists"}