I just installed Tensorflow2.0 for Python 3.6+JetPack4.5 on my Jetson Nano, by following the instructions here.
I modified them slightly because I wanted to have it installed in a virtual environment, so I did the following:
*) installed TF dependencies
sudo apt-get install libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran sudo apt-get install python3-pip
*) created the virtual environment:
sudo apt-get install virtualenv python3 -m virtualenv -p python3 tf_env source envs/tf_env/bin/activate
*) from within the new environment, I installed TF by:
pip install testresources setuptools==49.6.0 pip install numpy==1.16.1 future==0.18.2 mock==3.0.5 h5py==2.10.0 keras_preprocessing==1.1.1 keras_applications==1.0.8 gast==0.2.2 futures protobuf pybind11 pip install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v45 tensorflow
Tensorflow seems to be installed normally, I can successfully import it from Python within the environment.
At this point, I wanted to install the object detection API, by following the official instructions.
I cloned the repo:
git clone https://github.com/tensorflow/models.git
cd models/research # Compile protos. protoc object_detection/protos/*.proto --python_out=.
but at this point I got an error, since it seems protobuf is not installed
-bash: protoc: command not found
So, I’ve been following the protobuf installation guide, but the ‘make check’ step fails.
Is this the right way to install the object detection API, or is there another one?