I am trying it on large datasets and even sampling with 5000 row each and a relatively normal mapping (20 data properties, 5 classes, 5 entity linking), I am quickly coming to a point where it runs over 10 mins.
Is there something that I am missing in use of the yarrrml parser + rmlmapper?
If not, I would find it interesting to have a GPU accelerated version of the mapper or recommendationon CPU config sizing based on mapping size and sources dimension (even roughly) if someone has such a study.
Hello,
Thanks a lot for this great tool.
I am trying it on large datasets and even sampling with 5000 row each and a relatively normal mapping (20 data properties, 5 classes, 5 entity linking), I am quickly coming to a point where it runs over 10 mins.
Is there something that I am missing in use of the yarrrml parser + rmlmapper?
If not, I would find it interesting to have a GPU accelerated version of the mapper or recommendationon CPU config sizing based on mapping size and sources dimension (even roughly) if someone has such a study.
Thanks.