neuralop.datasets.load_darcy_flow_small

neuralop.datasets.load_darcy_flow_small(n_train, n_tests, batch_size, test_batch_sizes, test_resolutions=[16, 32], grid_boundaries=[[0, 1], [0, 1]], positional_encoding=True, encode_input=False, encode_output=True, encoding='channel-wise', channel_dim=1)[source]

Loads a small Darcy-Flow dataset

Training contains 1000 samples in resolution 16x16. Testing contains 100 samples at resolution 16x16 and 50 samples at resolution 32x32.

Parameters:

n_trainint

n_testsint

batch_sizeint

test_batch_sizesint list

test_resolutionsint list, default is [16, 32],

grid_boundariesint list, default is [[0,1],[0,1]],

positional_encodingbool, default is True

encode_inputbool, default is False

encode_outputbool, default is True

encoding‘channel-wise’

channel_dimint, default is 1

where to put the channel dimension, defaults size is 1 i.e: batch, channel, height, width

Returns:

training_dataloader, testing_dataloaders

training_dataloadertorch DataLoader

testing_dataloadersdict (key: DataLoader)