Unable to install pynq with pip in a conda environment on ZCU104

Support
1

Hi, I’m a student who is new to Pynq. I would like to use Pynq on ZCU104 while managing my environment with conda. But I was unable to install pynq with pip.

Here are the detailed steps:

  1. Boot with the precompiled v3.0.1 image found on PYNQ supported boards and PYNQ pre-built images | PYNQ
  2. Install miniconda (linux/arm64) with installer found on Download Now | Anaconda
  3. Create a conda environment with python 3.10
  4. Deactivate base environment & activate the new one
  5. Install ipykernel, jupyter, pytorch & torchvision with conda
  6. Install pynq with pip
1 Like
2

When I tried to perform step 6, it always ended with ModuleNotFoundError: No module named ‘pynqutils’

image
image1405×430 18.8 KB

Here are the things I’ve tried, but none of them worked

  1. Run the commands with sudo
  2. Install pynqmetadata & pynqutils with pip first, as described on failure to upgrade pynq in virtual python environment · Issue #1432 · Xilinx/PYNQ · GitHub) (the installation was successful)
  3. Run the command sudo pip3 install --upgrade --upgrade-strategy only-if-needed pynq as described on pip3 install fails ModuleNotFound error · Issue #19 · Xilinx/PYNQ-HelloWorld · GitHub

Is there any way to solve this issue ?
Many thanks !

1 Like
3

Hi,
Is your issue similar to this previous request?

ModuleNotFoundError: No module named ‘pynq.utils’

4

Hi matthew,

Thanks for replying.

My issue is similar to that, but the solution there doesn’t work for me.
pip install pynq and pip3 install pynq give me the same results.

Also, my error message is No module named 'pynqutils'.
I don’t know if it’s equivalent to 'pynq.utils'.

Thanks again !

5

Was this ever solved, I’m having alot of trouble installing pynq with a conda env activated via pip

6

PYNQ doesn’t directly support installation through conda, we use a Python virtual environment instead.

What is your use case for conda? The PYNQ SD card images already come with the Python virtual environment installed and is used when running the Jupyter environment. If you pip install any packages within the Jupyter environment (e.g. from the terminal) they will automatically be added to the virtual environment.

If you want to install packages outside the Jupyter interface (e.g. ssh or serial) then you can source the virtual environment by running source /etc/profile.d/pynq_venv.sh (make sure you are running as root), then just pip install any packages you want.