You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Trying to install Model-Angelo into a fairly barebones Ubuntu 22.04 environment (in a cloud Docker instance, if that makes a difference) leaves me in conda package resolution hell (see stderr output below). I suspect one way to avert this would be to put specific version numbers for each package on the conda install line in install_script.sh... could you offer any advice?
It does look like installation is correctly proceeding after I specify pytorch<2.0 and reduce the Python environment from 3.10 to 3.9. Not absolutely sure if the former is necessary - I tried that first and didn't change it before changing the Python version specification.
Ok this is weird. Definitely better practice to specify specific versions. Let me try a few different combinations and update this. The main thing being I want to keep pytorch-2 because the torch.compile makes the Ca segmentation much faster.
Trying to install Model-Angelo into a fairly barebones Ubuntu 22.04 environment (in a cloud Docker instance, if that makes a difference) leaves me in conda package resolution hell (see stderr output below). I suspect one way to avert this would be to put specific version numbers for each package on the
conda install
line ininstall_script.sh
... could you offer any advice?The text was updated successfully, but these errors were encountered: