Hi all,
When using the Spline regression tool I sometimes see that a variable that is listed in the variable importance plot has the word "-unused" next to it. That variable is then not listed in the coefficients list. The variable importance it is given can be quite high however.
What does this mean? Why doesn't the tool use these variables, even if they have high importance?
It seems odd, because other unused variables are just left out of the variable importance plot entirely.
Many thanks!
Hi @CharlotteIH
You may want to take a look at the new LIME macro that was just released: https://community.alteryx.com/t5/Data-Science/Expand-Your-Predictive-Palette-V-No-More-Black-Box-wit...
It helps a lot with model explainability and works with the spline model.
User | Count |
---|---|
107 | |
82 | |
70 | |
54 | |
40 |