mnist_corrupted

• Description:

MNISTCorrupted is a dataset generated by adding 15 corruptions to the test images in the MNIST dataset. This dataset wraps the static, corrupted MNIST test images uploaded by the original authors

• Additional Documentation: Explore on Papers With Code

• Homepage: https://github.com/google-research/mnist-c

• Source code: tfds.image_classification.MNISTCorrupted

• Versions:

  • 1.0.0 (default): New split API

• Download size: 235.23 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'test'	10,000
'train'	60,000

• Feature structure:

FeaturesDict({
    'image': Image(shape=(28, 28, 1), dtype=uint8),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=10),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
image	Image	(28, 28, 1)	uint8
label	ClassLabel		int64

• Supervised keys (See as_supervised doc): ('image', 'label')

• Citation:

@article{mu2019mnist,
  title={MNIST-C: A Robustness Benchmark for Computer Vision},
  author={Mu, Norman and Gilmer, Justin},
  journal={arXiv preprint arXiv:1906.02337},
  year={2019}
}

mnist_corrupted/identity (default config)

• Config description: Corruption method: identity

• Dataset size: 21.00 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/shot_noise

• Config description: Corruption method: shot_noise

• Dataset size: 15.05 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/impulse_noise

• Config description: Corruption method: impulse_noise

• Dataset size: 24.61 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/glass_blur

• Config description: Corruption method: glass_blur

• Dataset size: 35.91 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/motion_blur

• Config description: Corruption method: motion_blur

• Dataset size: 32.45 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/shear

• Config description: Corruption method: shear

• Dataset size: 27.73 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/scale

• Config description: Corruption method: scale

• Dataset size: 17.41 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/rotate

• Config description: Corruption method: rotate

• Dataset size: 27.09 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/brightness

• Config description: Corruption method: brightness

• Dataset size: 20.52 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/translate

• Config description: Corruption method: translate

• Dataset size: 20.88 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/stripe

• Config description: Corruption method: stripe

• Dataset size: 23.09 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/fog

• Config description: Corruption method: fog

• Dataset size: 35.76 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/spatter

• Config description: Corruption method: spatter

• Dataset size: 29.67 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/dotted_line

• Config description: Corruption method: dotted_line

• Dataset size: 24.45 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/zigzag

• Config description: Corruption method: zigzag

• Dataset size: 25.91 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

mnist_corrupted/canny_edges

• Config description: Corruption method: canny_edges

• Dataset size: 13.23 MiB

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):