INNER CODE UNIT · Python

d

ankonzoid/LearningX · basic_ML/classification/decision_tree.py:118

        N, d = len(df), len(df_header) - 1
        X = np.array(df.drop(['y'], axis=1))  # extract X by dropping y column
        y_raw = np.array(df['y'])  # extract y

        y_classes_raw = sorted(list(set(y_raw)))
        y_classes_dict = {}
        for i, y_class_i in enumerate(y_classes_raw):
            y_classes_dict[y_class_i] = i

        y = np.array([y_classes_dict[y_i] for y_i in y_raw])
        y_classes = list(set(y))
        assert X.shape == (N, d)  # check X.shape
        assert y.shape == (N,)  # check y.shape
        if mode == "clf":
            assert y.dtype in ['int64']  # check y are integers
        elif mode == "regr":
            assert y.dtype in ['int64', 'float64']  # check y are integers/floats
        else:

View source record →