Regression with TPOT?

[dbikard]

This might be more of a feature request, but I'm wondering if it's possible to use TPOT to perform a regression task instead of a classification task. Maybe this is already possible and I'm missing something? Could someone point me in the right direction?

Thanks for your help.
David

[rhiever]

TPOT doesn't currently support regression, but we have plans to do so in the near future once we have the basic algorithm worked out in the supervised classification case.

[rhiever]

@teaearlgraycold / @tonyfast, it should be pretty easy to swap out the classifiers for regressors, the scoring function with MSE, and have TPOT work the same otherwise, yes? Am I missing anything? Maybe we should prioritize this addition.

[tonyfast]

This should be a constraint in #91 .

ClassifierMixins and RegessorMixins define their own scoring functions. I have been experimenting in this notebook.

[rhiever]

Hi @dbikard, we just released TPOT v0.6 today with regression support. Try upgrading TPOT via pip and using it on your regression data set. Usage docs: link

[dbikard]

Awesome! Thanks.
On Fri, 2 Sep 2016 at 22:29, Randy Olson notifications@github.com wrote:

Hi @dbikard https://github.com/dbikard, we just released TPOT v0.6
today with regression support. Try upgrading TPOT via pip and using it on
your regression data set. Usage docs: link
http://rhiever.github.io/tpot/using/

—
You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub
#186 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAl3Er6skb_L4T7hfee964eB4QkleecVks5qmIdBgaJpZM4I9i1e
.