Version 3.0.0

[f-aguzzi]

This is the roadmap to Version 3.0.0:

• put a virtual method called train into BaseActionClass to improve uniformity of all reducers and classifiers
• make a new report generation system based on dictionaries

classDiagram
    BaseDataModel <|-- LLDFDataModel
    BaseDataModel <|-- ComponentDataModel
    ComponentDataModel <|.. PCADataModel
    ComponentDataModel <|.. LDADataModel
    
    class BaseDataModel {
        x_data
        x_train
        y
        load_from_file()
        export_to_file()
    }

    <<abstract>> ComponentDataModel
    class ComponentDataModel {
        n_components
    }

    class PCADataModel {
        array_scores
    }

    class LDADataModel {

    }

    class LLDFDataModel {
        tables
    }

    BaseActionClass <|.. BaseReducer
    BaseActionClass <|.. BaseClassifier

    BaseDataModel *-- BaseActionClass

    <<abstract>> BaseActionClass
    class BaseActionClass {
        train()
        settings
        data ~BaseDataModel~
        model
        from_file()
        import_model()
        export_model()
    }

    BaseReducer <|.. PCA
    BaseReducer <|.. LDA
    BaseClassifier <|.. LDA

    <<abstract>> BaseReducer
    class BaseReducer {
        components
        rescaled_data
        export_data()
        reduce()
    }

    <<abstract>> BaseClassifier
    class BaseClassifier {
        predict()
    }

    BaseClassifier <| .. LDA
    BaseClassifier <| .. LR
    BaseClassifier <| .. SVM
    BaseClassifier <| .. KNN
    BaseClassifier <| .. PLSDA

    class PCA {
        pca_stats()
        export_model()
    }

    class LLDF {
        fuse_data()
    }

Loading