Large data set giving have heap space error

Hi, I am trying to apply Decision Tree learner on a large data set about 20000X30000- (rows by columns), however it errors out with - “Execute failed: java heap space” error.

Current configuration of machine- 8 GB Ram
Knime.INI file- -Xmx4032m

Kindly let me know how to resolve this.

@thampy123 welcome to the knime forum.

4 GB RAM is not that much. You could try to give 5 or 6 leaving 2 for the system. Then you might want to explore these collection about performance issues and what go to about them:

Hoe large is your dataset when written to disk?

1 Like

Hi @thampy123 and welcome to the KNIME community forum

@mlauber71 is right about the need of extra memory but I’m also wondering about the underlying nature of the problem you need to solve.
Could you please tell us a bit about it ? What kind of problem do you need to solve ? Where do the 30,000 variables come from and why so many ?

Knowing a bit more about your data should help us to guide you further.


1 Like

Indeed a good point. One thing could be to reduce some dimensions - but it would be best to know more about the nature of the task.

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.