Closed b-mazur closed 1 month ago
Hi Bartosz, which version of psiflow did you install in your local micromamba environment, and which container are you using? We had to update the CP2K interface at some point because Pymatgen's parser couldn't deal with fluorine atoms (F -> FALSE), and I'm thinking that there was a small fix for an error like this right after we released 3.0.4.
Can you try installing psiflow directly from the main branch ( pip install git+https://github.com/molmod/psiflow
) while using the 3.0.4 container (at oras://ghcr.io/molmod/psiflow:3.0.4_python3.10_cuda
)?
That being said, we are planning a major overhaul of psiflow in the next few weeks, which will introduce a large number of new features but more importantly streamline the API and reduce possibility of errors like this one. We're currently finalizing this on the ipi
branch. While the API will be very similar, it should support many more use cases (path integral MD, replica exchange, thermodynamic integration, ... ). See here for some examples.
Hi Sander, thanks for the quick response! I was using the latest version of psiflow, while I had to copy the link to the old version of the container (3.0.0). I have corrected this and it appears that everything is now working correctly!
Happy to hear that psiflow is developing so fast, it looks like a very promising tool. I will update if I encounter any problems.
Hi! I was trying to simply run mof_phase_transition.py example, however I get an error:
I've tried to get to where the source of this problem is, but despite the beautiful documentation, I've failed... ☹️ Is the fault on the psiflow side or is it a cp2k issue? Any idea how to fix it?