Deep learning for NeuroImaging in Python.

Note

This page is a reference documentation. It only explains the class signature, and not how to use it. Please refer to the gallery for the big picture.

class surfify.datasets._generic.GenericSurfDataset(root, patterns, subject_in_patterns, ico_order, targets, target_mapping=None, split='train', transforms=None, mask=None, contrastive=False, patch=False, n_max=None, withdraw_subjects=None, target_ico_order=None, size=3)[source]¶

A scalable neuroimaging dataset.

Parameters:

root : str

the location where are stored the data.

patterns : str or list of str

the regex that can be used to retrieve the images of interest or any data that can be retrieved by nibabel.load.

subject_in_patterns : int or list of int

the folder level where the subject identifiers can be retrieved.

ico_order : int

the input data ico order.

targets : str or list of str

the dataset will also return these tabular data. A ‘participants.tsv’ file containing subject information (including the requested targets) is expected at the root.

target_mapping : dict, default None

optionaly, define a dictionary specifying different replacement values for different existing values. See pandas DataFrame.replace documentation for more information.

split : str, default ‘train’

define the split to be considered. A ‘<split>.tsv’ file containg the subject to include us expected at the root.

transforms : callable or list of callable, default None

a function that can be called to augment the input images.

mask : array, default None

optionnaly, mask the input image.

contrastive : bool, default False

optionaly, create a contrastive dataset that will return a pair of augmented images.

patch : bool, default False

optionaly, return triangular patches.

n_max : int, default None

optionaly, keep only a subset of subjects (for debuging purposes).

withdraw_subjects : list of str, default None

optionaly, provide a list of subjects to remove from the dataset.

target_ico_order : int, default None

the desired ico order (data will be downsample to this resolution).

size : int, default 3

the patch size.