feature request: setting up another method for correlation analysis #66
Description
Related to #65
I found information that Spearman correlation coefficient is used for numerical variables and Cramer's V for categorical ones. It would be nice to be able to set different methods for correlation analysis (as function defined by user or at least the name of the implemented method).
PS You have a typo in documentation for check_cor. I believe it should be Cramer's V instead of Crammer V there.