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
There seems to be some inconsistent between how the Eigen and Cholesky decompositions of correlation matrices are implement. Results are correct, but the method used is weird. Need to check the write.morpho and other functions that use the decomposition, and standardise. Perhaps we should just keep the Eigen decomposition. For example, for post hoc ancestral sequence reconstruction, it is important to use the same decomposition used when writing the morphological alignment, otherwise the reconstruction may be wrong (need to check this). Using a single method for the decomposition would avoid this problem.
The text was updated successfully, but these errors were encountered:
There seems to be some inconsistent between how the Eigen and Cholesky decompositions of correlation matrices are implement. Results are correct, but the method used is weird. Need to check the
write.morpho
and other functions that use the decomposition, and standardise. Perhaps we should just keep the Eigen decomposition. For example, for post hoc ancestral sequence reconstruction, it is important to use the same decomposition used when writing the morphological alignment, otherwise the reconstruction may be wrong (need to check this). Using a single method for the decomposition would avoid this problem.The text was updated successfully, but these errors were encountered: