User Tom

60 votes

7 answers

8k views

Industry vs Kaggle challenges. Is collecting more observations and having access to more variables more important than fancy modelling?

asked Jul 10, 2018 at 12:42

11 votes

1 answer

2k views

What is the intuition behind the positional cosine encoding in the transformer network?

asked May 22, 2019 at 16:54

8 votes

3 answers

1k views

how to make sense of the number of observations per parameters in deep learning models?

asked Nov 22, 2018 at 13:27

7 votes

1 answer

966 views

how to deal with correlated/colinear features when using Permutation feature importance?

asked Sep 26, 2018 at 13:59

5 votes

1 answer

1k views

Bootstrap intervals for predictions, how to interpret it?

asked Aug 27, 2018 at 15:39

4 votes

1 answer

94 views

How to explain local minima found between two trained Neural networks?

asked Nov 14, 2019 at 17:08

3 votes

1 answer

69 views

How to deal with partial/unshared features in a dataset?

asked Jun 16, 2018 at 0:20

3 votes

1 answer

819 views

Classification followed by regression?

asked Jun 16, 2018 at 12:22

2 votes

1 answer

780 views

Standardization on training only or also including testing data?

asked Aug 11, 2018 at 20:53

2 votes

2 answers

199 views

Do earlier hidden layers learn more concepts/features than later ones, in neural networks?

asked Jan 21, 2020 at 14:42

1 vote

0 answers

15 views

If you want to train a model on the predictions of another trained model, would you use the same training set for both model?

asked Jun 17, 2021 at 21:51

1 vote

0 answers

22 views

class-specific neural networks vs overall network

asked Oct 22, 2019 at 19:21

1 vote

1 answer

148 views

How to use/treat a hidden layer as the new target to predict in a neural network?

asked Mar 9, 2019 at 19:20

1 vote

0 answers

23 views

How to quantify the effect of the inputs and the effect of the labels on our regression model?

asked Apr 6, 2019 at 23:42

1 vote

1 answer

382 views

Random Forests with correlated features

asked Jun 21, 2018 at 10:49

1 vote

1 answer

19 views

How to take advantage of the information/structure we have in the labels in multiple output regression?

asked Jun 29, 2018 at 18:32

1 vote

0 answers

350 views

Why using Out-of-fold predictions as metafeatures in stacking?

asked Jul 2, 2018 at 15:55

0 votes

0 answers

584 views

how to find a test for detecting fluctuation in a time series?

asked Aug 16, 2019 at 11:22

Stack Exchange Network

18 Questions

Industry vs Kaggle challenges. Is collecting more observations and having access to more variables more important than fancy modelling?

What is the intuition behind the positional cosine encoding in the transformer network?

how to make sense of the number of observations per parameters in deep learning models?

how to deal with correlated/colinear features when using Permutation feature importance?

Bootstrap intervals for predictions, how to interpret it?

How to explain local minima found between two trained Neural networks?

How to deal with partial/unshared features in a dataset?

Classification followed by regression?

Standardization on training only or also including testing data?

Do earlier hidden layers learn more concepts/features than later ones, in neural networks?

If you want to train a model on the predictions of another trained model, would you use the same training set for both model?

class-specific neural networks vs overall network

How to use/treat a hidden layer as the new target to predict in a neural network?

How to quantify the effect of the inputs and the effect of the labels on our regression model?

Random Forests with correlated features

How to take advantage of the information/structure we have in the labels in multiple output regression?

Why using Out-of-fold predictions as metafeatures in stacking?

how to find a test for detecting fluctuation in a time series?