Install tensorflow2.1 and pytorch1.4 on Jetson nano

Based on the latest JetPack 4.3 Mirror image .
Please refer to the boot configuration Blog

First step ： in pip Source

1. Root creation .pip file ：mkdir ~/.pip
2. create a file pip.conf：vim .pip/pip.conf
3. Click on “i” key , Enter edit mode , Copy information ：

sudo mkdir ~/.pip
cd ~/.pip
sudo touch pip.conf
sudo nano ~/.pip/pip.conf
# Write the following configurations 
[global] 
index-url = http://pypi.douban.com/simple 
[install] 
trusted-host=pypi.douban.com

The second step ： Installation dependent

sudo apt-get install libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev
sudo apt-get install python3-pip
sudo pip3 install -U pip
sudo pip3 install -U numpy grpcio absl-py py-cpuinfo psutil portpicker six mock requests gast h5py astor termcolor protobuf keras-applications keras-preprocessing wrapt google-pasta

The third step ： install tensorflow2.1:

$ sudo pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v43 tensorflow==2.1.0+nv20.3

Among them, installation 2.1 Version will appear scipy 1.4.1 There is a problem with the compilation, which makes it impossible to install ,
But this scipy1.4.1 I've compiled .

link ：https://pan.baidu.com/s/1VdsNq-ETnIlN5mG8rJXDKw
Extraction code ：6cnb

It is recommended to install first scipy 1.4.1 Then install tensorflow2.1 So that we don't make a mistake

Step four ： Install one by the way keras:

sudo pip3 install keras

Step five ： install pytorch1.4:

wget https://nvidia.box.com/shared/static/ncgzus5o23uck9i5oth2n8n06k340l6k.whl -O torch-1.4.0-cp36-cp36m-linux_aarch64.whl
sudo apt-get install python3-pip libopenblas-base
sudo pip3 install Cython
sudo pip3 install numpy torch-1.4.0-cp36-cp36m-linux_aarch64.whl