Non uniformity in benchmarking scripts of Shogun about benchmarks HOT 6 CLOSED

Yes, I had a look. I fetched the pull request. I will make the necessary changes as a new branch and send a pull request from my side.

from benchmarks.

You are right they should behave the same, the decision tree script should work on a single dataset as well, without the prediction step. You are right about if (len(self.dataset) > 1), actually this is already fixed in my fork; will open a PR in the next days unless you like to do it?

from benchmarks.

I would love to change the benchmarking scripts so that they work on single dataset. I think there are more like that, for example knc.py. It's a comparatively easy fix and will help me get an idea of writing benchmarking scripts. Thanks alot!

And about the if (len(self.dataset) > 1), I think I will leave it to you if you have already fixed it.

from benchmarks.

If you like to take a look at this, it probably makes sense to take a look at the update I'm working on. I'll try to open a PR either today or tomorrow.

from benchmarks.

Okay sure. Thanks alot!

from benchmarks.

@braceletboy I've opened #137 let me know what you think.

from benchmarks.