Stack Exchange

Ben Reiniger

Chicago, IL, USA

http://bmreiniger.github.io

math PhD (research in combinatorics) turned data scientist

top accounts reputation activity bookmarks subscriptions
Top Questions
No questions with score of 5 or more
Top Answers
19
L1 & L2 Regularization in Light GBM
datascience.stackexchange.com

14
What would I prefer - an over-fitted model or a less accurate model?
datascience.stackexchange.com

12
Sklearn components in pipeline is not fitted even if the whole pipeline is?
stackoverflow.com

11
Using a random forest, would a RandomForest performance be less if I drop the first or the last tree?
datascience.stackexchange.com

10
PSE Advent Calendar 2021 (Day 3): Tentai Show
puzzling.stackexchange.com

9
Is `TweedieRegressor` a completely general GLM solution?
stats.stackexchange.com

9
Calibration curve of XGBoost for binary classification
stats.stackexchange.com

8
Is There a Way to Re-Calibrate Predicted Probabilities After Using Class Weights?
datascience.stackexchange.com

8
Decision tree with final decision being a linear regression
datascience.stackexchange.com

8
Why use fit when already have fit_transform?
datascience.stackexchange.com

8
Can Boosted Trees predict below the minimum value of the training label?
datascience.stackexchange.com

8
Does scikit-learn have a forward selection/stepwise regression algorithm?
datascience.stackexchange.com

7
Target encoding in test data and target leakage
stats.stackexchange.com

7
Default value of learning rate in adam optimizer - Keras
datascience.stackexchange.com

7
Does oversampling lead to more overfitting than classweights for really small classes?
stats.stackexchange.com

7
How do we standardize arrays with NaN?
datascience.stackexchange.com

7
Sorting 6 boxes by selecting 3 boxes each time
puzzling.stackexchange.com

7
How to extract trees in XGBoost?
datascience.stackexchange.com

7
ROC AUC score is much less than average cross validation score
datascience.stackexchange.com

7
What is the proper way to use early stopping with cross-validation?
datascience.stackexchange.com

7
What happens if at leaf node both classes have same number of samples?
datascience.stackexchange.com

7
Why GridSearchCV returns nan?
datascience.stackexchange.com

6
URL Feature representations
stats.stackexchange.com

6
Why is there a difference between predicting on Validation set and Test set?
datascience.stackexchange.com

6
Probability calibration is worsening my model performance
datascience.stackexchange.com

6
Why decision tree needs categorical variable to be encoded?
datascience.stackexchange.com

6
What makes a Tree-Structured Parzen Estimator "tree-structured?"
datascience.stackexchange.com

5
handling missing values for LightGBM model
datascience.stackexchange.com

5
Logistic regression does cannot converge without poor model performance
datascience.stackexchange.com

5
Several independent variables based on the same underlying data
datascience.stackexchange.com

1 2