User Monica Heddneck

14 votes

3 answers

13k views

Why convolute if Max Pooling is just going to downsample the image anyway?

asked Sep 21, 2016 at 7:55

8 votes

1 answer

686 views

Why does "Depth = Semantic representation" in convolutional neural networks?

asked Sep 8, 2016 at 23:14

8 votes

2 answers

19k views

Counting the number of layers in a neural network

asked Sep 15, 2016 at 22:48

7 votes

3 answers

5k views

What is the purpose of multiple neurons in a hidden layer?

asked Sep 16, 2016 at 0:51

4 votes

1 answer

810 views

What does a predicted probability really mean, without considering the accuracy of the underlying model?

asked Aug 8, 2017 at 8:06

4 votes

1 answer

175 views

What regressors are recommended with text modeling?

asked Jan 19, 2016 at 6:26

3 votes

2 answers

2k views

What are alternatives to MLP, when you have rectangular, structured data?

asked Aug 10, 2017 at 16:32

2 votes

1 answer

104 views

Does dropout require multiple passes of the same data set, as a sort of ensemble method?

asked Sep 5, 2016 at 7:17

1 vote

1 answer

103 views

Graphs demonstrating the structure of neural networks are very unclear

asked Sep 6, 2016 at 15:58

1 vote

1 answer

707 views

How do multiple linear neurons together allow for nonlinearity in a neural network?

asked Sep 16, 2016 at 3:37

1 vote

0 answers

3k views

Can tuning individual precision and recall classification thresholds improve deep learning models?

asked Aug 28, 2017 at 3:04

1 vote

0 answers

97 views

how could RFECV ever give the same score to all number of features selected?

asked May 21, 2021 at 0:45

1 vote

2 answers

232 views

If ROC is used to find a threshold, but AUC is threshold invariant, why use AUC?

asked Jun 7, 2021 at 20:01

0 votes

1 answer

398 views

XGBClassifier and RandomForestClassifier variable importance plots are very different

asked Aug 9, 2017 at 7:41

Stack Exchange Network

14 Questions

Why convolute if Max Pooling is just going to downsample the image anyway?

Why does "Depth = Semantic representation" in convolutional neural networks?

Counting the number of layers in a neural network

What is the purpose of multiple neurons in a hidden layer?

What does a predicted probability really mean, without considering the accuracy of the underlying model?

What regressors are recommended with text modeling?

What are alternatives to MLP, when you have rectangular, structured data?

Does dropout require multiple passes of the same data set, as a sort of ensemble method?

Graphs demonstrating the structure of neural networks are very unclear

How do multiple linear neurons together allow for nonlinearity in a neural network?

Can tuning individual precision and recall classification thresholds improve deep learning models?

how could RFECV ever give the same score to all number of features selected?

If ROC is used to find a threshold, but AUC is threshold invariant, why use AUC?

XGBClassifier and RandomForestClassifier variable importance plots are very different