Open nspeer12 opened 4 years ago
got the same error as well. I am guessing there is not much difference with the voice output if tf 1.x is continued in this. A quick fix for now I did was:
%tensorflow_version 1.x
import tensorflow as tf
print(tf.__version__)
Hoping there will be an update in the classes
Tensorflow 1 is being used in hparams.py. Tensorflow 1 is not a specified requirement, nor is it supported anymore. Tensorflow.contrib.haparms has been moved to from tensorboard.plugins.hparams, however it will need to be reimplemented because the classes are not the same.