{"id":15288149,"url":"https://github.com/magnusax/automl","last_synced_at":"2026-01-20T03:03:59.222Z","repository":{"id":57433148,"uuid":"90233715","full_name":"magnusax/AutoML","owner":"magnusax","description":"The project aims to develop a customized ML framework on top of existing libraries ","archived":false,"fork":false,"pushed_at":"2018-08-09T18:06:27.000Z","size":820,"stargazers_count":2,"open_issues_count":2,"forks_count":0,"subscribers_count":3,"default_branch":"master","last_synced_at":"2024-09-20T09:12:10.988Z","etag":null,"topics":["data-science","machine-learning","machine-learning-algorithms","machine-learning-library","python","scikit-learn"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/magnusax.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2017-05-04T07:23:23.000Z","updated_at":"2018-08-21T13:27:18.000Z","dependencies_parsed_at":"2022-08-28T04:23:58.662Z","dependency_job_id":null,"html_url":"https://github.com/magnusax/AutoML","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/magnusax%2FAutoML","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/magnusax%2FAutoML/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/magnusax%2FAutoML/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/magnusax%2FAutoML/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/magnusax","download_url":"https://codeload.github.com/magnusax/AutoML/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247427012,"owners_count":20937214,"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":["data-science","machine-learning","machine-learning-algorithms","machine-learning-library","python","scikit-learn"],"created_at":"2024-09-30T15:44:23.778Z","updated_at":"2026-01-20T03:03:59.216Z","avatar_url":"https://github.com/magnusax.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# Gazer\n\n\u003e \"Standing on the shoulders of giants, gazing at the stars\"\n\nThe project aims to develop a customized \"automl\" framework, leveraging the power of existing popular libraries. \nAs time passes the intention of the author is to add non-standard capabilities such as advanced forms of ensemble\nbuilding and Markov Chain Monte Carlo (MCMC) like optimization algorithms (see further below for a brief overview of ongoing efforts).\nThe project is still in its infancy, and fairly basic functionality has been implemented; but it is currently \nunder active development (conditional on time and energy).\n\nInstall is now available through the python package index. Run `pip install gazer==0.1.dev1` to install development version.\nCurrently tested for python versions `\u003e=3.4` only. \n\n### Using the library\nThe entrance to functionality is the **GazerMetaLearner** object;\n    \n```python\nfrom gazer import GazerMetaLearner\nlearner = GazerMetaLearner()\n```\n\t\nSet parameter `method='all'` to initialize all available algorithms. Default is to initialize 3 algorithms randomly.\n\nPost initialization, (meta) algorithms are available in the `.clf` variable of the initialized object. To be more specific, the variable \nreturns a dictionary of (name, MetaClassifier) tuples, where the **MetaClassifier** object is a wrapper around a \"sklearn-like\"\nclassifier. The current version of the library also implements *xgboost* and *keras* classifiers.\n\n```python    \t\nlearner.clf = {name1: MetaClassifier1, name2: MetaClassifier2, ... , nameN: MetaClassifierN}\n```\n   \t\nTo inspect loaded algorithms, simply call `learner.names`.\n\nAssuming a standard supervised learning scenario with data available in the form of `X, y`, you can easily fit \navailable algorithms using default (i.e. reasonble) hyperparameter settings:\n\n```python\t\nlearner.fit(X, y)\n```\n\nThis method trains all initialized algorithms. Moreover, Random, Grid search, and Bayesian optimization methods\nare implemented and may be directly called from the `learner` instance. To see parameters, simply call:\n```python\nlearner.clf['name'].cv_params\n\n```\nThis returns a list containing one or more parameter dictionaries that may be edited in place.\n\n\n### Roadmap\nSome ideas are currently being implemented, e.g.\n* Markov Chain Monte Carlo scheme for optimization of (very-) large ensemble models \nfor large data sets where training times are relatively long.\n* Model perturbation and greedy adaptation\n* Advanced ensembling techniques\n  - See [these two papers from Caruana et al.](http://www.cs.cornell.edu/~caruana/ctp/ct.papers/caruana.icml04.icdm06long.pdf)\n\n\nML libraries which will be heavily relied upon are first and foremost:\n* (Done) Scikit-learn\n* (Done) Xgboost\n* (Done) Keras \n* (Coming) Pytorch (`\\ge 0.4`)\n\n\nTODO: add direct links to each github project mentioned on this page, \nto ensure that credit and recognition goes to developers\n\n\n### Comments\nAs this project is in its initial development phase, the code has not been tested properly yet. It thus follows\nthat it has not yet been uploaded to the PyPi repository yet. Project developer's contact info is available in the setup.py file. \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmagnusax%2Fautoml","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmagnusax%2Fautoml","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmagnusax%2Fautoml/lists"}