Open sfd99 opened 6 years ago
Hi Biecek,
breakDown R Pkg is great and easy to use.
I followed your Wine quality example lm() model in: https://pbiecek.github.io/breakDown/
But my question is on how to interpret / understand the results, shown in plot(br) of the WINE example.
Q1) If the var "residual.sugar" is = 1.2 (highest positive value of all vars), does it mean, it is the most significant var contributing?.
Q2) Why do some vars have positive values, (ie: var "residual.sugar" is = 1.2) and other vars in the plot show negative values?.
Q3) What is the meaning of the plot bar colors?. Some vars in the plot(br) are yellow, some are green.
and finally, Q4) What is the meaning of "final_prognosis" = 5.6? (gray color bar).
Just trying to learn and understand what the results mean in practical terms. Thanks for your help Biecek!
Hi Biecek,
breakDown R Pkg is great and easy to use.
I followed your Wine quality example lm() model in: https://pbiecek.github.io/breakDown/
But my question is on how to interpret / understand the results, shown in plot(br) of the WINE example.
Q1) If the var "residual.sugar" is = 1.2 (highest positive value of all vars), does it mean, it is the most significant var contributing?.
Q2) Why do some vars have positive values, (ie: var "residual.sugar" is = 1.2) and other vars in the plot show negative values?.
Q3) What is the meaning of the plot bar colors?. Some vars in the plot(br) are yellow, some are green.
and finally, Q4) What is the meaning of "final_prognosis" = 5.6? (gray color bar).
Just trying to learn and understand what the results mean in practical terms. Thanks for your help Biecek!
RAY San Francisco using latest Rstudio, R and Ubuntu Linux