Dataloader Module¶

Created on Sat Nov 28 16:40:41 2020 Abdullah BAS abdullah.bas@boun.edu.tr BME Bogazici University Istanbul / Uskudar @author: abas

class exceLoader.dataset(path, gtpath=None, responseCol=- 1, phase='train', preprocess=True, smooth=True, normalise=True, transform=True)¶

Bases: torch.utils.data.dataset.Dataset

This function initializes the dataset.

Parameters

• path (string) – Input path

• gtpath (string,optional) – Path for ground truths. Default to None

• responseCol (string,optional) – If ınput dataset has the response in it you can simply assign it to this value. Note: 0 for first, 1 for second column same as python. Take that into account

• phase (str, optional) – Defaults to ‘train’.

• preprocess (bool, optional) – Switch for preprocess. Defaults to True.

• smooth (bool, optional) – Switch for smoothing. Defaults to True.

• normalise (bool, optional) – Switch for normalise. Defaults to True.

• transform (bool, optional) – Switch for yeo-johnson power transformation. Defaults to True.