[scikit-learn] Continues monitoring of benchmark performances

[Jeremie du Boisberranger]

 > Isn't Jérémie's project at 
https://github.com/jeremiedbb/scikit-learn_benchmarks meant to be doing 
this? What's its status? How does it relate to Tom's work?

Yes it's the same project. Tom kindly accepted to run it alongside other 
projects from the pydata ecosystem.


 > I can't find it back but I think Jérémie has submitted an issue about 
this?

I didn't submit an issue but briefly mentioned it at the Paris sprint in 
february.


 > but I had to disable them since they were taking too long (longer 
than a day)

Something wrong is definitely going on. It should take 1-2 hours (it 
does on my laptop).

Before it was running on Tom's machine, we were considering running it 
on a dedicated machine at INRIA.

Maybe it will be better to do that after all.


 > What's its status?

After seeing run for a few weeks, I think it still needs some more work. 
More readable presentation of the results. Some benchmarks show large 
fluctuations. It might be the hardware or maybe my settings are not good.



On 23/07/2019 08:25, Alexandre Gramfort wrote:
> it's the same projects.
>
> Alex
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn