You need to use tensorflow version <= 2.2.0. There are a few posts here that face issues with getting this version of tensorflow to work. If you use anaconda then if you look at my post it may help; “C1W3 Run the TensorFlow.js Converter on The Saved Keras Model Error.”
@Osama_Saad_Farouk, anaconda will not let you install tensorflow 2.2. You have to use conda to install tensorflow 2.1 and then use pip to install tensorflow 2.2. Also, make sure you do this in the anaconda prompt as I stated in the following response.
Try running this line by line individually in anaconda’s conda prompt, not in jupyter lab:
conda create -n tf210 tensorflow=2.1.0 python=3.7
conda activate tf210
pip install tensorflowjs==2.0.0
pip install tensorflow-intel==0.0.1
pip install tensorflow-cpu==2.1.0
pip install tensorflow-estimator==2.1.0rc0
pip install tensorflow==2.2.0
pip install matplotlib
You should still get errors and package compatibility errors but ignore those. So, you can ignore the gast and tensorboard package errors. Then go to anaconda navigator, install jupyter lab in the tf210 environment, and complete the assignment. Do not install anything in the assignment. Comment out the pip installs in the assignment, and you should see that you are using tensorflow 2.2.0.
If this doesn’t work, then change the first line to conda create -n tf210 tensorflow=2.0.0 python=3.0. If that doesn’t work, then I am sorry but I don’t have any more suggestions.