User qmeeus

37 votes

Accepted

Early stopping on validation loss or on accuracy?

answered Apr 19, 2019 at 14:01

4 votes

Machine learning - 'train_test_split' function in scikit-learn: should I repeat it several times?

answered Aug 22, 2018 at 11:48

2 votes

Seemingly good results with training a CNN but bad when testing

answered Aug 22, 2018 at 11:55

2 votes

Accepted

How do we implement a custom loss that backpropagates with PyTorch?

answered Jun 3, 2020 at 7:37

2 votes

ValueError: Error when checking input: expected conv2d_13_input to have shape (3, 150, 150) but got array with shape (150, 150, 3)

answered Apr 6, 2019 at 10:32

2 votes

Accepted

How do you retrain a model as new data comes in?

answered Feb 28, 2023 at 10:49

2 votes

Random Forest always predicting the majority class

answered Nov 7 at 16:58

1 vote

Scoring vectors by "distinctness"

answered Feb 28, 2023 at 11:05

1 vote

Accepted

Getting value of 2 in normalized confusion matrix

answered Jul 10, 2020 at 11:58

1 vote

How to change parameters in LSTM for multivariate binary classification time series?

answered Jul 17, 2020 at 10:05

1 vote

Which courses should I take in order to learn ML and AI?

answered Apr 6, 2019 at 10:27

1 vote

Accepted

Encoding categorical data with pre-determined dictionary

answered Sep 10, 2020 at 11:23

1 vote

Can time series foundation models be used for classification?

answered Oct 16 at 13:04

1 vote

Efficacy of model depends on scaling?

answered Oct 31, 2018 at 23:44

1 vote

ROC curve for deep learning

answered Oct 31, 2018 at 23:56

1 vote

Ways to create simulated numerical data based on small sample (using Python)?

answered Aug 22, 2018 at 12:00

1 vote

Accepted

Suspected Exploding Gradient in Character Generator LSTM

answered Aug 28, 2018 at 7:00

1 vote

Label Encoder encoding the same class as two integers

answered Sep 19, 2018 at 13:06

1 vote

Accepted

Batch data before feed into CNN network

answered Sep 26, 2018 at 8:16

0 votes

Binary classificaiton for weather data if its class 1 or class 0 alert

answered Oct 17, 2018 at 12:49

0 votes

How to recommend items after finding similar users in recommendation system

answered Aug 22, 2018 at 13:29

0 votes

Accepted

ML technique to predict next year output based on text quantities

answered Feb 2, 2019 at 9:15

0 votes

How can I detect patterns and/or keywords or phrases?

answered Apr 6, 2019 at 10:43

0 votes

Accepted

parsing dictionary to JSON

answered Jul 10, 2020 at 11:39

0 votes

A way to init sentence embedding for unsupervised text clustering, better than glove wordvec?

answered Jul 20, 2020 at 8:20

0 votes

Are hidden markov models limited to just bigrams?

answered Oct 16 at 13:00

0 votes

Why does tanh activation work better with Pytorch than with Keras?

answered Nov 7 at 17:14

-1 votes

Remove columns with a certain number of consecutive zeros

answered Apr 19, 2019 at 14:25

Stack Exchange Network

28 Answers

Early stopping on validation loss or on accuracy?

Machine learning - 'train_test_split' function in scikit-learn: should I repeat it several times?

Seemingly good results with training a CNN but bad when testing

How do we implement a custom loss that backpropagates with PyTorch?

ValueError: Error when checking input: expected conv2d_13_input to have shape (3, 150, 150) but got array with shape (150, 150, 3)

How do you retrain a model as new data comes in?

Random Forest always predicting the majority class

Scoring vectors by "distinctness"

Getting value of 2 in normalized confusion matrix

How to change parameters in LSTM for multivariate binary classification time series?

Which courses should I take in order to learn ML and AI?

Encoding categorical data with pre-determined dictionary

Can time series foundation models be used for classification?

Efficacy of model depends on scaling?

ROC curve for deep learning

Ways to create simulated numerical data based on small sample (using Python)?

Suspected Exploding Gradient in Character Generator LSTM

Label Encoder encoding the same class as two integers

Batch data before feed into CNN network

Binary classificaiton for weather data if its class 1 or class 0 alert

How to recommend items after finding similar users in recommendation system

ML technique to predict next year output based on text quantities

How can I detect patterns and/or keywords or phrases?

parsing dictionary to JSON

A way to init sentence embedding for unsupervised text clustering, better than glove wordvec?

Are hidden markov models limited to just bigrams?

Why does tanh activation work better with Pytorch than with Keras?

Remove columns with a certain number of consecutive zeros