glmfit: Iteration limit reached
5 views (last 30 days)
Show older comments
Hello, I built a logistic model using glmfit:
[b,dev,stats]=glmfit(X,Y,'binomial','link','logit');
but I have obtained the warning: iteration limit reached. And some variables have large coefficients, so the produced model is not correct. I have 41 features and 2904 observations. The 2 classes (0 and 1) are already balanced. I have attached the matrix X and Y.
Can someone help me? Thanks.
0 Comments
Answers (1)
the cyclist
on 6 Oct 2023
Edited: the cyclist
on 6 Oct 2023
The first thing I notice is that the X data has missing values (NaN).
MATLAB automatically removes observations (i.e. rows) with missing values. That does not seem to yield a good model.
Removing variables (i.e. columns) with missing data does bring all the coefficients down into roughly the same range.
It's unclear without more context what is the best way to handle your missing data, but I would focus on that first.
0 Comments
See Also
Categories
Find more on Classification in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!