{
    "info": {
        "author": "AutoML Freiburg",
        "author_email": "zimmerl@informatik.uni-freiburg.de",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "License :: OSI Approved :: BSD License",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering",
            "Topic :: Scientific/Engineering :: Artificial Intelligence",
            "Topic :: Utilities"
        ],
        "description": "# Auto-PyTorch\n\nCopyright (C) 2019  [AutoML Group Freiburg](http://www.automl.org/)\n\nThis a very early pre-alpha version of our upcoming Auto-PyTorch.\nSo far, Auto-PyTorch supports featurized data (classification, regression) and image data (classification).\n\n## Installation\n\nClone repository\n\n```sh\n$ cd install/path\n$ git clone https://github.com/automl/Auto-PyTorch.git\n$ cd Auto-PyTorch\n```\nIf you want to contribute to this repository switch to our current develop branch\n\n```sh\n$ git checkout develop\n```\n\nInstall pytorch: \nhttps://pytorch.org/\n\nInstall Auto-PyTorch:\n\n```sh\n$ cat requirements.txt | xargs -n 1 -L 1 pip install\n$ python setup.py install\n```\n\n\n## Examples\n\nFor a detailed tutorial, please refer to the jupyter notebook in https://github.com/automl/Auto-PyTorch/tree/master/examples/basics.\n\nIn a nutshell:\n\n```py\nfrom autoPyTorch import AutoNetClassification\n\n# data and metric imports\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\nX, y = sklearn.datasets.load_digits(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n        sklearn.model_selection.train_test_split(X, y, random_state=1)\n\n# running Auto-PyTorch\nautoPyTorch = AutoNetClassification(\"tiny_cs\",  # config preset\n                                    log_level='info',\n                                    max_runtime=300,\n                                    min_budget=30,\n                                    max_budget=90)\n\nautoPyTorch.fit(X_train, y_train, validation_split=0.3)\ny_pred = autoPyTorch.predict(X_test)\n\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, y_pred))\n```\n\nMore examples with datasets:\n\n```sh\n$ cd examples/\n\n```\n\n## Configuration\n\nHow to configure Auto-PyTorch for your needs:\n\n```py\n\n# Print all possible configuration options.\nAutoNetClassification().print_help()\n\n# You can use the constructor to configure Auto-PyTorch.\nautoPyTorch = AutoNetClassification(log_level='info', max_runtime=300, min_budget=30, max_budget=90)\n\n# You can overwrite this configuration in each fit call.\nautoPyTorch.fit(X_train, y_train, log_level='debug', max_runtime=900, min_budget=50, max_budget=150)\n\n# You can use presets to configure the config space.\n# Available presets: full_cs, medium_cs (default), tiny_cs.\n# These are defined in autoPyTorch/core/presets.\n# tiny_cs is recommended if you want fast results with few resources.\n# full_cs is recommended if you have many resources and a very high search budget.\nautoPyTorch = AutoNetClassification(\"full_cs\")\n\n# Enable or disable components using the Auto-PyTorch config:\nautoPyTorch = AutoNetClassification(networks=[\"resnet\", \"shapedresnet\", \"mlpnet\", \"shapedmlpnet\"])\n\n# You can take a look at the search space.\n# Each hyperparameter belongs to a node in Auto-PyTorch's ML Pipeline.\n# The names of the hyperparameters are prefixed with the name of the node: NodeName:hyperparameter_name.\n# If a hyperparameter belongs to a component: NodeName:component_name:hyperparameter_name.\n# Call with the same arguments as fit.\nautoPyTorch.get_hyperparameter_search_space(X_train, y_train, validation_split=0.3)\n\n# You can configure the search space of every hyperparameter of every component:\nfrom autoPyTorch import HyperparameterSearchSpaceUpdates\nsearch_space_updates = HyperparameterSearchSpaceUpdates()\n\nsearch_space_updates.append(node_name=\"NetworkSelector\",\n                            hyperparameter=\"shapedresnet:activation\",\n                            value_range=[\"relu\", \"sigmoid\"])\nsearch_space_updates.append(node_name=\"NetworkSelector\",\n                            hyperparameter=\"shapedresnet:blocks_per_group\",\n                            value_range=[2,5],\n                            log=False)\nautoPyTorch = AutoNetClassification(hyperparameter_search_space_updates=search_space_updates)\n```\n\nEnable ensemble building (for featurized data):\n\n```py\nfrom autoPyTorch import AutoNetEnsemble\nautoPyTorchEnsemble = AutoNetEnsemble(AutoNetClassification, \"tiny_cs\", max_runtime=300, min_budget=30, max_budget=90)\n\n```\n\nDisable pynisher if you experience issues when using cuda:\n\n```py\nautoPyTorch = AutoNetClassification(\"tiny_cs\", log_level='info', max_runtime=300, min_budget=30, max_budget=90, cuda=True, use_pynisher=False)\n\n```\n\n## License\n\nThis program is free software: you can redistribute it and/or modify\nit under the terms of the Apache license 2.0 (please see the LICENSE file).\n\nThis program is distributed in the hope that it will be useful,\nbut WITHOUT ANY WARRANTY; without even the implied warranty of\nMERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\n\nYou should have received a copy of the Apache license 2.0\nalong with this program (see LICENSE file).\n\n## Reference\n\n```\n@incollection{mendoza-automlbook18a,\n  author    = {Hector Mendoza and Aaron Klein and Matthias Feurer and Jost Tobias Springenberg and Matthias Urban and Michael Burkart and Max Dippel and Marius Lindauer and Frank Hutter},\n  title     = {Towards Automatically-Tuned Deep Neural Networks},\n  year      = {2018},\n  month     = dec,\n  editor    = {Hutter, Frank and Kotthoff, Lars and Vanschoren, Joaquin},\n  booktitle = {AutoML: Methods, Sytems, Challenges},\n  publisher = {Springer},\n  chapter   = {7},\n  pages     = {141--156},\n  note      = {To appear.},\n}\n```\n\n**Note**: Previously, the name of the project was AutoNet. Since this was too generic, we changed the name to AutoPyTorch. AutoNet 2.0 in the reference mention above is indeed AutoPyTorch.\n\n\n## Contact\n\nAuto-PyTorch is developed by the [AutoML Group of the University of Freiburg](http://www.automl.org/).\n\n\n",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/automl/Auto-PyTorch",
        "keywords": "machine learning algorithm configuration hyperparameter optimization tuning neural architecture deep learning",
        "license": "3-clause BSD",
        "maintainer": "",
        "maintainer_email": "",
        "name": "autoPyTorch",
        "package_url": "https://pypi.org/project/autoPyTorch/",
        "platform": "Linux",
        "project_url": "https://pypi.org/project/autoPyTorch/",
        "project_urls": {
            "Homepage": "https://github.com/automl/Auto-PyTorch"
        },
        "release_url": "https://pypi.org/project/autoPyTorch/0.0.2/",
        "requires_dist": [
            "setuptools",
            "Cython",
            "netifaces",
            "numpy",
            "pandas",
            "scipy",
            "statsmodels",
            "scikit-learn (>=0.20.0)",
            "imblearn",
            "ConfigSpace",
            "pynisher",
            "hpbandster",
            "fasteners",
            "torch",
            "torchvision",
            "tensorboard-logger",
            "openml"
        ],
        "requires_python": ">=3",
        "summary": "Auto-PyTorch searches neural architectures using BO-HB",
        "version": "0.0.2"
    },
    "last_serial": 5976486,
    "releases": {
        "0.0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "b7659a9480c3d20f16bab869b98b1d31",
                    "sha256": "7a4176ef4678391f17255470d52205570905c70a6a1c8673d7ff38e9d7ff4a05"
                },
                "downloads": -1,
                "filename": "autoPyTorch-0.0.2-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "b7659a9480c3d20f16bab869b98b1d31",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": ">=3",
                "size": 312235,
                "upload_time": "2019-10-15T11:49:38",
                "url": "https://files.pythonhosted.org/packages/51/1f/b883a924de574ab2478ceb72232a22bc2502f3a472ac1cec62df49ad8e32/autoPyTorch-0.0.2-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "f16785c9f8312c246cc4955c331f4db3",
                    "sha256": "ed185e341888324fcf95964ba26af5ba531f4355c1f3b042ca0d2cacbbdb2e33"
                },
                "downloads": -1,
                "filename": "autoPyTorch-0.0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "f16785c9f8312c246cc4955c331f4db3",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3",
                "size": 208440,
                "upload_time": "2019-10-15T11:49:42",
                "url": "https://files.pythonhosted.org/packages/a4/8f/f1b8c1cb2ff3f23569aa957278ae28a6f4215b5e4f993d845ec97038f2f6/autoPyTorch-0.0.2.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "b7659a9480c3d20f16bab869b98b1d31",
                "sha256": "7a4176ef4678391f17255470d52205570905c70a6a1c8673d7ff38e9d7ff4a05"
            },
            "downloads": -1,
            "filename": "autoPyTorch-0.0.2-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "b7659a9480c3d20f16bab869b98b1d31",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3",
            "size": 312235,
            "upload_time": "2019-10-15T11:49:38",
            "url": "https://files.pythonhosted.org/packages/51/1f/b883a924de574ab2478ceb72232a22bc2502f3a472ac1cec62df49ad8e32/autoPyTorch-0.0.2-py3-none-any.whl"
        },
        {
            "comment_text": "",
            "digests": {
                "md5": "f16785c9f8312c246cc4955c331f4db3",
                "sha256": "ed185e341888324fcf95964ba26af5ba531f4355c1f3b042ca0d2cacbbdb2e33"
            },
            "downloads": -1,
            "filename": "autoPyTorch-0.0.2.tar.gz",
            "has_sig": false,
            "md5_digest": "f16785c9f8312c246cc4955c331f4db3",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": ">=3",
            "size": 208440,
            "upload_time": "2019-10-15T11:49:42",
            "url": "https://files.pythonhosted.org/packages/a4/8f/f1b8c1cb2ff3f23569aa957278ae28a6f4215b5e4f993d845ec97038f2f6/autoPyTorch-0.0.2.tar.gz"
        }
    ]
}