Cann't install tensorflow due to h5py -Jetson Nano

Hi there ,
i’m using jetson Nano 4G with jetpack version 4.6.1
i have managed to download a python version 3.9.16 on my jetson so i’m using it now but it’s not my system python as the system python is still 3.6

i’m trying to install tensorflow but whatever i did it doesn’t seems to work , i followed the official installation guide and didn’t work , followed some random utube videos and they didn’t help either .

from what i managed to understand is that h5py library was originally for python version2 so they did some backend developing to make it compatible with python v3 and those made some conflicts ‘this was from a utube video which i followed his steps and didn’t work too’

for a specific error here what comes after running the command :
$ pip3.9 install tensorflow (note i want to download it on my python3.9 version not the 3.6 one)

here is the error :

Defaulting to user installation because normal site-packages is not writeable
Collecting tensorflow
Using cached tensorflow-2.16.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.metadata (4.1 kB)
Collecting absl-py>=1.0.0 (from tensorflow)
Using cached absl_py-2.1.0-py3-none-any.whl.metadata (2.3 kB)
Collecting astunparse>=1.6.0 (from tensorflow)
tuptools/", line 967, in run_command
File “/tmp/pip-build-env-da3v8uq0/overlay/lib/python3.9/site-packages/setuptools/_distutils/”, line 988, in run_command
File “/tmp/pip-install-u04dfqy4/h5py_331add77a6f14729abccad72c095445a/”, line 147, in run
raise Exception(
Exception: This version of h5py requires HDF5 >= 1.10.4 (got version (1, 10, 0) from environment variable or library)
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for h5py
Failed to build h5py
ERROR: Could not build wheels for h5py, which is required to install pyproject.toml-based projects


Welcome to the NVIDIA Developer forums! Your topic will be best served in the Jetson category.

I will move this post over for visibility.



For Jetson Nano, we only provide the python3.6 TensorFlow package.
The installation steps can be found in the below topic: