Cross Validation brings the alpha as 0.0 , although my model has variance #19359
Unanswered
berkaykepekci
asked this question in
Q&A
Replies: 1 comment 2 replies
-
This seems weird:
I would say the best way to get help is to put together a stand-alone snippet that reproduces the problem. A few other remarks:
|
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I built a model with ElasticNet and tuned the model with ElasticNetCV:
Before the tuning
There ise a huge gap between train and test scores,r2_scores and MSEs
So I asssumed that my model is overfitted. However, after the tuning, ElasticNetCV gave the alpha_ as "0.0".As I know L1 and L2 methods are against "variance and overfitting". Why I got the alpha_ as "0.0" although my model is overfitted? The most strange thing is that r2_scores,MSEs and model_scores increased after tuning:
Note: alpha was equal to 1.0 before the tuning.
There was 120k samples in train , 30k samples in test set and 11 features (73 features after One-Hot)
cv-scores (tuning): [0.47673268, 0.47434977, 0.47877276, 0.4823315 , 0.46908168, 0.47770449, 0.47593381, 0.48605855, 0.46665624, 0.45359717]
Beta Was this translation helpful? Give feedback.
All reactions