mldas.explore.prepare.prepare_loader

mldas.explore.prepare. prepare_loader ( datapath , img_size , stride , sample_size , batch_size , nrand = None , adjust = False , shuffle = True , verbose = True )

Create custom data loader with unlabeled, single-channel, data images directly extracted from raw data files.

Parameters

datapath str

Path to raw data repository

img_size int

Size of squared image

stride int

Sliding interval

sample_size int

Number of input MAT data files to use

batch_size int

Batch size to use in DataLoader

nrand int

Specify total number of images in final set

adjust bool

Do brightness adjustment

shuffle bool

Shuffling images in set

verbose bool

Do verbose

Returns

train_loader torch.utils.data.DataLoader

Data loader