User Carlos Mougan

11 votes

4 answers

2k views

Why SMOTE is not used in prize-winning Kaggle solutions?

asked Dec 27, 2021 at 16:50

8 votes

1 answer

1k views

Why gradient boosting uses sampling without replacement?

asked Feb 7, 2020 at 6:59

6 votes

1 answer

777 views

Is there any difference between a weak learner and a weak classifier?

asked Jan 5, 2020 at 22:15

5 votes

0 answers

2k views

Dealing with categorical variables in Isolation Forest

asked May 5, 2020 at 11:14

4 votes

1 answer

3k views

For outliers treatment: clipping, winsorizing or removing?

asked Jan 3, 2020 at 16:09

4 votes

1 answer

5k views

F1_score(average='micro') is equal to calculating accuracy for multiclasification

asked Dec 12, 2019 at 11:47

4 votes

1 answer

116 views

For a square matrix of data, I achieve $R^2=1$ for Linear Regression and $R^2=0$ for Lasso. What's the intuition behind?

asked Dec 28, 2019 at 9:52

4 votes

2 answers

706 views

Repeated features in Neural Networks with tabular data

asked Nov 8, 2020 at 21:45

3 votes

2 answers

257 views

What is a "surrogate model"?

asked Jul 14, 2020 at 7:21

3 votes

2 answers

939 views

Why Deep Learning / Neural Networs don't achieve state of the art results in tabular data problems?

asked May 15, 2022 at 18:42

3 votes

2 answers

5k views

Is it possible to make F1_Score differentiable and use it directly as a Loss function?

asked Jan 16, 2020 at 13:28

3 votes

2 answers

4k views

Does it make sense to do train test split when trainning GANS?

asked Jan 19, 2020 at 9:13

3 votes

1 answer

199 views

Make a random forest estimator the exact same of a decision tree

asked Jan 22, 2020 at 17:53

2 votes

1 answer

1k views

What is the difference between horizontal and vertical ensemble?

asked Jan 2, 2020 at 11:36

2 votes

1 answer

794 views

Interpretable xgboost - Calculate cover feature importance

asked Feb 27, 2020 at 8:31

2 votes

1 answer

444 views

Categories with the same mean in target encoding

asked Apr 5, 2020 at 17:57

2 votes

1 answer

147 views

A deployed model has epistemic or aleatoric uncertainty?

asked Jul 29, 2021 at 10:01

1 vote

2 answers

436 views

Distribution Shift vs Transfer Learning

asked Nov 3, 2021 at 9:20

1 vote

1 answer

130 views

Using GANs to generate synthetic tabular data to improve supervised learning

asked Dec 8, 2020 at 8:47

1 vote

0 answers

81 views

Constrastive vs Counterfactual Explanations

asked Jun 1, 2021 at 8:38

1 vote

1 answer

287 views

Alternative regression model algorithms for machine learning

asked Jul 23, 2020 at 12:24

1 vote

0 answers

383 views

Partial Dependence Plot and categorical variables

asked Aug 1, 2020 at 15:30

1 vote

1 answer

1k views

Can Shapley/Lime values be used for unsupervised learning?

asked Aug 10, 2020 at 21:38

1 vote

1 answer

81 views

predict_proba() on a continuos target made binary

asked Sep 19, 2020 at 10:15

1 vote

2 answers

636 views

What is the difference between conformal prediction and uncertainty estimation

asked Jan 17 at 14:17

1 vote

1 answer

112 views

Using ARIMA parameters when transforming time series to Supervised Learning

asked May 4, 2020 at 16:02

1 vote

3 answers

495 views

Why would it be bad to fit a regression model to a binary classification problem?

asked Jun 1, 2020 at 16:17

0 votes

1 answer

3k views

Value on Decision Tree plot

asked May 14, 2020 at 8:20

0 votes

1 answer

689 views

What is the best way to encode features when clustering data? [duplicate]

asked Dec 19, 2019 at 17:37

Stack Exchange Network

29 Questions

Why SMOTE is not used in prize-winning Kaggle solutions?

Why gradient boosting uses sampling without replacement?

Is there any difference between a weak learner and a weak classifier?

Dealing with categorical variables in Isolation Forest

For outliers treatment: clipping, winsorizing or removing?

F1_score(average='micro') is equal to calculating accuracy for multiclasification

For a square matrix of data, I achieve $R^2=1$ for Linear Regression and $R^2=0$ for Lasso. What's the intuition behind?

Repeated features in Neural Networks with tabular data

What is a "surrogate model"?

Why Deep Learning / Neural Networs don't achieve state of the art results in tabular data problems?

Is it possible to make F1_Score differentiable and use it directly as a Loss function?

Does it make sense to do train test split when trainning GANS?

Make a random forest estimator the exact same of a decision tree

What is the difference between horizontal and vertical ensemble?

Interpretable xgboost - Calculate cover feature importance

Categories with the same mean in target encoding

A deployed model has epistemic or aleatoric uncertainty?

Distribution Shift vs Transfer Learning

Using GANs to generate synthetic tabular data to improve supervised learning

Constrastive vs Counterfactual Explanations

Alternative regression model algorithms for machine learning

Partial Dependence Plot and categorical variables

Can Shapley/Lime values be used for unsupervised learning?

predict_proba() on a continuos target made binary

What is the difference between conformal prediction and uncertainty estimation

Using ARIMA parameters when transforming time series to Supervised Learning

Why would it be bad to fit a regression model to a binary classification problem?

Value on Decision Tree plot

What is the best way to encode features when clustering data? [duplicate]