hibou.data_preprocessing.split#

hibou.preprocessing.split(*args, **kwargs)#

This function split the inputs and the targets data into two datasets: training and testing.

Parameters#

inputspd.DataFrame: The inputs data.
targetspd.DataFrame: The targets data.
timestampspd.DataFrame: The timestamps of the data. Please note that as the inputs and targets data are supposed synchronized, you only need to pass one timestamps.
nb_trainint: The number of points into each training data sets.
nb_testint: The number of points into each testing data sets.
offsetint, optionnal: By default: 0. The number of points to be shifted in training and evaluation sets.
displaybool, optionnal: By default: False. If set on True, the function will plot the targets that are used for training and testing on a figure.

Note

The figure will not be showed, but saved into {OUTPUT_DIR}/figures.

tuple: A tuple that contains two elements: training data and testing data. Each element is itself a tuple that contains three elements: inputs, targets, timestamps.