Auto-sklearn

Auto-sklearn is an open-source Automated Machine Learning (AutoML) tool built on top of scikit-learn. It automatically selects the best machine learning models, optimizes hyperparameters using Bayesian optimization, and builds ensembles for classification and regression tasks. Auto-sklearn simplifies the creation of accurate machine learning pipelines by automating data preprocessing, model selection, evaluation, and optimization with minimal user intervention.

Features of Auto-sklearn:

• Automates model selection and hyperparameter tuning for supervised learning tasks.
• Supports both classification and regression problems.
• Handles preprocessing tasks such as missing value imputation, scaling, and encoding.
• Performs cross-validation and provides reliable performance metrics.
• Builds ensemble models to improve prediction accuracy and robustness.
• Supports parallel computation for faster and efficient model optimization.
• Fully compatible with scikit-learn pipelines and ecosystem.
• Open-source and distributed under the BSD License.

Auto-sklearn Usage

$ sudo su
$ source /opt/autosklearn39-env/bin/activate
$ python -c "import autosklearn; print(autosklearn.__version__)"     
$ pip show auto-sklearn | grep Version