AI for Anomaly and Outlier detection (AI4AO)

AI4AO is a Python package that allows to build any of the scikit-learn supported Clustering and Classification algorithms based machine learning models in batches. This means that one can use yaml declarative syntax in order to write a configuration file, and based on the instructions in the configuration file, and the machine learning models will be constructed sequentially. This way many models can be built with a single configuration file with the results being arranged in an extremely modular way. AI4AO can be considered as a convenient wrapper for scikit-learn models.

• Developed by: Selvakumar Ulaganathan
• Website: www.selvai.com
• PyPI Project: https://pypi.org/project/ai4ao
• GitLab Source: https://gitlab.com/selvai/ai4ao
• Official Documentation: https://selvai.gitlab.io/ai4ao/

Usage

Define a configuration in config.yaml

    # config.yaml
    IsolationForest_0.01:
        project_name: timeseries_anomaly
        run_this_project: True
        multi_variate_model: True
        model: IsolationForest
        data:
            path: 'path-to-train-data.csv'
            test_data_path: 'path-to-train-data.csv'
            features_to_avoid: ['feat-to-avoid']
        hyperparams:
            contamination: 0.01
        results:
            path: 'results/isolation_forest_001/'
        remote_run: False

Run the model defined in config.yaml

    # example_script.py
    import ai4ao # import package 
    from ai4ao.models import SKLearnModel as Model # scikit-learn wrapper 

    # fit and evaluate model
    model = Model(plot_results=True)
    model.batch_fit(path_config='configs.yaml')

    # print models and metrics
    print(model.models)
    print(model.metrics())