Saduf2019

I ran the code shared and face memory related issues, please find the [gist here](https://colab.research.google.com/gist/Saduf2019/75fdc0dad0344355c9b02a21b6263f07/untitled503.ipynb)

I ran the code on tf 2.4 and and face a different error on nightly, please find the [gist here](https://colab.research.google.com/gist/Saduf2019/e6435e077a9cd287a043bed1389da761/untitled507.ipynb)

@kaiyu-pony You can use "tf.compat.v1.disable_eager_execution()" to avoid the error, please refer to this [gist here](https://colab.research.google.com/gist/Saduf2019/eedfcb6027362d1c0dc5db5ce5aca460/untitled631.ipynb).

I am able to replicate this issue on tf-nightly (2.4.0-dev20200929), please find the [gist here](https://colab.research.google.com/gist/Saduf2019/2a7de31fddf7d8b5d942bfd7aefd9513/untitled418.ipynb).

@fengyang0317 While i run the code shared on tpu i face a different error, please find the [gist of the same](https://colab.sandbox.google.com/gist/Saduf2019/ff29cf83a0d2cd331dfca56867707cb9/tpu.ipynb)

I am able to replicate the issue reported, please find the [gist here](https://colab.sandbox.google.com/gist/Saduf2019/1ae9894150b4bb7f1ac95d8d33846ef7/untitled191.ipynb)

@Tessil Is this still an issue

I am able to replicate the issue reported, please find the [gist here](https://colab.research.google.com/gist/Saduf2019/7b0f4b070dd2eda6e0d68e9035e7b703/untitled478.ipynb)

@liasece Memory error indicates that the process has consumed all of the ram memory. You may want to reduce the batch size/image size and try again.

@liasece I ran the code shared and face a different error, please find [gist here](https://colab.research.google.com/gist/Saduf2019/9756c6ec6bcbb1ccd63d9497a1c92022/untitled472.ipynb). please share a colab gist with the error reported.