AttributeError: module 'tensorflow.compat.v2' has no attribute 'depth_to_space'

Question

I am trying to run a code, which was written with tensorflow version 1.4.0 I am running my code on google colab which gives in tensorflow version 2.x with it.

To run my code, I am using backward compatibility like: replacing import tensorflow as tf with

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()

which works for certain things, but at a point it comes down to this error

AttributeError: module 'tensorflow.compat.v2' has no attribute 'depth_to_space'

as you can see in this image

What I am not able to understand is this method 'depth_to_space' is in both versions of tensorflow 1.x and 2.x, then why is my version of tensorflow not getting it? Here's the link to the method in version 1.x: https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/nn/depth_to_space

Please help me understand what's the cause of this error.

Thanks, Pranay

Answer 1

I tried with Tensorflow 1.15 .Its working fine. Looks like your Tensorflow version bit old,

Below sample code tested with Tf 1.15 without any error.

import tensorflow as tf
print(tf.__version__)
x = [[[[1, 2, 3, 4]]]]

tf.nn.depth_to_space(x, 2, data_format='NHWC', name=None)