tf.image.random_crop  |  TensorFlow v2.16.1 (original) (raw)

tf.image.random_crop

Stay organized with collections Save and categorize content based on your preferences.

Randomly crops a tensor to a given size.

View aliases

Compat aliases for migration

SeeMigration guide for more details.

tf.compat.v1.image.random_crop, tf.compat.v1.random_crop

tf.image.random_crop(
    value, size, seed=None, name=None
)

Used in the notebooks

Slices a shape size portion out of value at a uniformly chosen offset. Requires value.shape >= size.

If a dimension should not be cropped, pass the full size of that dimension. For example, RGB images can be cropped withsize = [crop_height, crop_width, 3].

Example usage:

image = [[1, 2, 3], [4, 5, 6]] result = tf.image.random_crop(value=image, size=(1, 3)) result.shape.as_list() [1, 3]

For producing deterministic results given a seed value, usetf.image.stateless_random_crop. Unlike using the seed param withtf.image.random_* ops, tf.image.stateless_random_* ops guarantee the same results given the same seed independent of how many times the function is called, and independent of global seed settings (e.g. tf.random.set_seed).