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
Normalization currently only cycles through each reaction once. However, as implemented in RDKit it should ideally go back to the start and start again if the molecule changes, in case a later reaction has opened up an earlier one.
This would be a breaking change for our current models so must go into the next major release of NAGL. In addition, a condition should be included in GNNModel that checks the model version, compares it to the NAGL version, and warns or errors if it's being loaded with a version that has this change.
Current environment
Which version are you using? (run python -c "import openff.nagl; print(openff.nagl.__version__)")
Which version of Python (python -V)?
Which operating system?
What is the output of pip list?
If you use conda, what is the output of conda list?
The text was updated successfully, but these errors were encountered:
Normalization currently only cycles through each reaction once. However, as implemented in RDKit it should ideally go back to the start and start again if the molecule changes, in case a later reaction has opened up an earlier one.
This would be a breaking change for our current models so must go into the next major release of NAGL. In addition, a condition should be included in GNNModel that checks the model version, compares it to the NAGL version, and warns or errors if it's being loaded with a version that has this change.
Current environment
python -c "import openff.nagl; print(openff.nagl.__version__)"
)python -V
)?pip list
?conda list
?The text was updated successfully, but these errors were encountered: