teacher.datasets

The teacher.datasets module includes the different databases used to run experiments with the teacher package.

Available datasets

This module includes load methods for the following datasets that are included:

• Adult: Ron Kohavi, “Scaling Up the Accuracy of Naive-Bayes Classifiers: a Decision-Tree Hybrid”, Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, 1996 [dataset]

• Beer: G. Castellano, C. Castiello, and A. M. Fanelli, “The FISDeT software: Application to beer style classification,” in Proc. IEEE Int. Conf. Fuzzy Syst., 2017, pp. 1–6.

• Breast: O. L. Mangasarian and W. H. Wolberg, “Cancer diagnosis via linear programming,” Dept. Comput. Sci., Univ. Wisconsin-Madison, Madison, WI, USA, Tech. Rep., 1990. [dataset]

• Compas: J. Skeem and J. Eno Louden, “Assessment of evidence on the quality of the correctional offender management profiling for alternative sanctions (COMPAS),” California Dept. Corrections Rehabilitation, 2007. [dataset]

• German: [dataset]

• Heloc: [dataset]

• Pima: [dataset]

Dataset format

The different methods return a dict with the following keys:

• name : str, Name of the dataset

• df : pandas.DataFrame Pandas DataFrame with the original data

• columns : list, columns of the DataFrame

• class_name : str, name of the class variable

• possible_outcomes : list, the values of the class column

• type_features : dict, the variables grouped by type

• features_type : dict, the type of each feature

• discrete : list, columns to be considered to have discrete values

• continuous : list, columns to be considered to have continuous values

• idx_features : dict, column name of each column once arranged in a NumPy array

• label_encoder : sklearn.preprocessing.LabelEncoder, label encoder for the discrete values

• X : numpy.ndarray, all columns except for the class

• y : numpy.ndarray, class column

Functions

load_adult()

Loads the adult dataset

load_beer()

Loads the beer dataset

load_breast()

Loads the breast dataset

load_compas()

Loads the compas dataset

load_heloc()

Loads the heloc dataset

load_pima()

Loads the pima dataset

---

teacher.datasets.load_adult(normalize=False)

Load and return the adult dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_basket(normalize=False, reduced=False)

Load and return the basket dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_beer(normalize=False)

Load and return the beer dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_breast(normalize=False)

Load and return the breast cancer dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_compas(normalize=False)

Load and return the COMPAS scores dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_flavia(normalize=False)

Load and return the FLAVIA dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_german(normalize=False)

Load and return the german credit dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_heloc(normalize=False)

Load and return the HELOC dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_iris(normalize=False)

Load and return the iris dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_phishing(normalize=False)

Load and return the phishing dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_pima(normalize=False)

Load and return the pima indians dataset.

Returns:

dataset

Return type:

dict

teacher.datasets.load_wine(normalize=False)

Load and return the wine dataset.

Returns:

dataset

Return type:

dict