Tensorflow "no matching distribution"

Hi everyone,

I am currently trying a Jetson Nano and succeeded in installing almost all the dependencies I need.
However, I am trying to install Tensorflow (2.0) on the device without success.
I have the latest image (R32.3.1) including Jetpack 4.3. I followed this tutorial: https://docs.nvidia.com/deeplearning/frameworks/install-tf-jetson-platform/index.html

The problem is when I try to install tensorflow itself with this line:

sudo pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v43 tensorflow-gpu

I directly get this error:

Looking in indexes: https://pypi.org/simple, https://developer.download.nvidia.com/compute/redist/jp/v43
ERROR: Could not find a version that satisfies the requirement tensorflow-gpu (from versions: none)
ERROR: No matching distribution found for tensorflow-gpu

Is there anything I could have done wrong at some point ? At
Here are some information about what is installed:

pip -V: pip 19.3.1 from /home/rd/.local/lib/python3.6/site-packages/pip (python 3.6)
Jetpack: 4.3

Thank you a lot !

Hi,

An alternative is to download the package here and manually install it with this command:

sudo pip3 install tensorflow_gpu-1.15.0+nv19.12-cp36-cp36m-linux_aarch64.whl

Would you mind to give it a try?
Thanks.

1 Like

Hi AastaLLL,

First of all, thank you a lot for your help.
Unfortunately, the given instruction failed too but in the meantime I realized the issue. There was a conflict with two 3.x Python version, making it impossible to install tensorflow 2… As it was a fresh install on a 16GB SD card, I took this opportunity to make a fresh install on a 32GB.

Again, thank you for your help,
Have a nice day !

can you please elaborate on this conflict? What conflicts and how did you resolve it? I have the same issue when trying to install tensorflow on my nano, havent managed to find a solution yet