Google I/O is a wrap! Catch up on TensorFlow sessions View sessions


TensorFlow 1 version View source on GitHub

Randomly flips an image vertically (upside down).

With a 1 in 2 chance, outputs the contents of image flipped along the first dimension, which is height. Otherwise output the image as-is. When passing a batch of images, each image will be randomly flipped independent of other images.

Example usage:

Randomly flip a single image.

import numpy as np

image = np.array([[[1], [2]], [[3], [4]]])
tf.image.random_flip_up_down(image, 3).numpy().tolist()
[[[3], [4]], [[1], [2]]]

Randomly flip multiple images.

images = np.array( ... [ ... [[[1], [2]], [[3], [4]]], ... [[[5], [6]], [[7], [8]]] ... ]) tf.image.random_flip_up_down(images, 4).numpy().tolist() [[[[3], [4]], [[1], [2]]], [[[5], [6]], [[7], [8]]]]

image 4-D Tensor of shape [batch, height, width, channels] or 3-D Tensor of shape [height, width, channels].
seed A Python integer. Used to create a random seed. See tf.compat.v1.set_random_seed for behavior.

A tensor of the same type and shape as image.

ValueError if the shape of image not supported.