공순이의 블로그

Rank = n-dimension tensor Rank = 0: 0 Rank = 1: [0, 1, 2] Rank = 2: [[0, 1, 2], [1, 2, 3]] Shape = length of each rank Shape = []: 0 Shape = [3]: [0, 1, 2] Shape = [2, 3]: [[0, 1, 2], [1, 2, 3]] Shape = [3, 1, 2]: [[[5, 7]], [[3, 4]], [[8. 2]]]

tensorflow 2020. 8. 26. 08:57

I was afraid of the installation was broken, but it was due to the TF version. So, getting TF 1.x like behavior in TF 2.0 import tensorflow.compat.v1 as tf tf.disable_v2_behavior() Reference: https://stackoverflow.com/questions/55142951/tensorflow-2-0-attributeerror-module-tensorflow-has-no-attribute-session

tensorflow 2020. 8. 26. 08:41