Generative adversarial networks(subset of Deep learning and Machine learning)

As a part of my personal work I have to use one of the research paper. You can find the paper from this link(https://arxiv.org/pdf/1612.07828v2.pdf). The implementation of the paper can be found from this link(https://paperswithcode.com/paper/learning-from-simulated-and-unsupervised). There are 8 implementations by various people. I find this implement in the link(https://github.com/mjdietzx/SimGAN) to be very helpful which you can also find it from Kaggle from this (https://www.kaggle.com/kmader/simgan-notebook/data?scriptVersionId=0). But the problem is tensor flow version used in that Kaggle notebook is expired so when i run that Kaggle notebook its gives error. So i need to fix that Kaggle notebook by make modification to code to that latest tensor flow and python. Modifications has to be done in Kaggle. itself.