Heat map issue

[LKguan]

Hi Boxuan,

Thank you for developing this helpful package; I’ve really enjoyed using it for exploratory data analysis.

Recently, I encountered an issue when generating a heatmap with DataExplorer. For a dataset with no missing data points, it worked perfectly. However, after deleting some outliers, the Pearson correlation coefficient (PCC) no longer seemed accurate. I calculated the PCC using other packages and Excel, and their results were consistent, but different from those generated by DataExplorer.

Could you kindly look into this issue? Thank you very much!

[boxuancui]

Hi @LKguan, thanks for raising the issue. If you take a look at the source code, it does nothing else but calling cor with supplies arguments (if any). Could you try passing your data directly to cor to validate again?