tapas.datasets.utils.encode_data

tapas.datasets.utils.encode_data(dataset, infer_ranges=False)

Convert raw data to an np.ndarray with continuous features normalised and categorical features one-hot encoded.

Parameters

• dataset (TabularDataset) – Tabular dataset to encode.

• infer_ranges (bool) – If false, will use ranges provided in metadata, otherwise, will use input data to infer ranges of the continuous variables and will update metadata in-place with the new ranges.

Returns

encoded_data – Encoded data (normalised and one-hot encoded).

Return type

np.ndarray