Want to accelerate the example

[SQDSDA]

When I run this example in device of kaggle, it need a lot of time to get the results. My data dimensions are 499*20, and I want to get some tricks to accelerate that. Can you give me some suggestions? thanks

[julianspaeth]

Unfortunately, this pure python implementation is not very performant. The bottleneck is the splitting rule - In the past, I used the fastlogranktest package, but this was not maintained anymore - I will have a look at it and try to speed it up a little.