MnistSvhn

class multivae.data.datasets.MnistSvhn(data_path, split='train', download=False, data_multiplication=5, **kwargs)[source]

A paired MnistSvhn dataset.

Parameters:

  • path (str) – The path where the data is saved.

  • split (str) – Either β€˜train’ or β€˜test’.

  • download (bool) – Whether to download the data or not. Default to True.

  • data_multiplication (int)

  • **kwargs – transform_mnist (Transform) : a transformation to apply to MNIST. If none specified, a simple ToTensor() is applied. transform_svhn (Transform) : a transformation to apply to SVHN. If none specified, a simple ToTensor() is applied.