Jetson Nano with CUDA 11 and JetPack (TensorFlow2 compatibility)

[dusty_nv]

@ryan.byrne it appears that somehow, a new protobuf 4 package appeared on the PyPi index for Python 3.6 which is in fact only for Python 3.7+ - instead, install protobuf<4. These TensorFlow wheels are for Python 3.6.

Here are the steps I just ran in a fresh environment to install TensorFlow wheel in JetPack 4.6:

Failed building wheel of h5py

@barton @gustavoflor3s I was just able to do git checkout 3.1.0 on h5py, maybe try again now? However, I did not need to manually clone/patch h5py anymore when doing a TensorFlow install in a fresh environment (in nvcr.io/nvidia/l4t-base:r32.7.1 container) : apt-get update apt-get install -y python3-pip pkg-config apt-get install -y libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran ln -s /usr/include/locale.h /usr/include/xlocale.h pip3 i…