Repository navigation
Major tech hardening, closing unused loops, improved documentation.
- change memory model to prevent mem leackages
- autocorrelations could sometimes remove the wrong genes or lead downstream to wrong centroid calculations
- canopy cluster calculations are more deterministic, --seed works as intended now
- correct tie-breaks for spearman correlations
- strict input validations
- zero variance calculations revised