NVIDIA-BioNeMo / NVIDIA-BioNeMo/KERMT
make cuik-molmaker optional
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 99
- Forks
- 18
- Avg merge
- 5d 14h
- Merged PRs (30d)
- 1
Description
Hello,
cuik-molmaker is currently required for fine-tuning KERMT. However, cuik-molmaker may not be compatible with many supercomputing environments, since it requires >= GLIBC_2.32. For example, NERSC perlmutter is compiled with GLIBC_2.31, so KERMT cannot be used on NERSC. Since cuik-molmaker mol graphs are equivalent to RDKit mol graphs, using cuik-molmaker should be made optional in molgraph.py.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading molgraph.py and tracing where cuik-molmaker is imported or required during KERMT fine-tuning. Compare the existing cuik-molmaker and RDKit molecular graph paths described in the issue. Done means cuik-molmaker is optional and KERMT can be used in environments with GLIBC_2.31.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Refactor
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100