Optimize for calculating correlations

fairnessforensics / wiggum

simpson's paradox inspired fairness forensics

https://fairnessforensics.github.io/wiggum/

MIT License

5 stars 3 forks source link

Optimize for calculating correlations #203

Open Shine226 opened 3 years ago

Shine226 commented 3 years ago

Current code is using corr to calculate correlation matrix. This approach is not good for asymmetrical case (e.g., 1 dependent and 10 independent variable). corr_mat = data_df[corr_var_list].corr(method=self.corrtype)

hatrick36 commented 3 years ago

Hi Shine, have you already started working on this issue?

Shine226 commented 3 years ago

I have some simple alternative code for this issue. But I am not working thoroughly on it.