When installing under Linux Ubuntu 24.04, the installation of psychopy requires wxpython which,in most situations, will result in a crash as the system tries to build it from source.
uv pip install -r requirements.txt
Resolved 104 packages in 966ms
× Failed to build `wxpython==4.3.1`
├─▶ The build backend returned an error
╰─▶ Call to `setuptools.build_meta.build_wheel` failed (exit status: 1)
[...]
hint: This usually indicates a problem with the package or the build environment.
help: `wxpython` (v4.3.1) was included because `psychopy` (v2026.1.3) depends on `wxpython`
Solution:
# Download https://extras.wxpython.org/wxPython4/extras/linux/gtk3/ubuntu-24.04/wxPython-4.2.2-cp310-cp310-linux_x86_64.whl
uv pip install ~/Downloads/wxPython-4.2.2-cp310-cp310-linux_x86_64.whl
Note: This only works for Python 3.10. Different wheels are needed for different versions of python.
PS1: Under Linux, the installations instructions for psychopy recommend to run the following command to get better timing performance (rt priority). This may be worth adding to your instructions.
sudo groupadd --force psychopy
sudo usermod -a -G psychopy $USER
sudo bash -c 'printf "@psychopy - nice -20\n@psychopy - rtprio 50\n@psychopy - memlock unlimited\n" >>/etc/security/limits.d/99-psychopylimits.conf'
PS2: the doc recommend pyenv. I used to use pyenv, but have switched to uv. If you want to add instructions for uv, here they are:
uv python install 3.10
cd MTB_FOLDER
uv venv --python 3.10
uv pip install -r requirements.txt
When installing under Linux Ubuntu 24.04, the installation of psychopy requires wxpython which,in most situations, will result in a crash as the system tries to build it from source.
Solution:
Note: This only works for Python 3.10. Different wheels are needed for different versions of python.
PS1: Under Linux, the installations instructions for psychopy recommend to run the following command to get better timing performance (rt priority). This may be worth adding to your instructions.
PS2: the doc recommend
pyenv. I used to use pyenv, but have switched to uv. If you want to add instructions for uv, here they are: