amoraglio
commented
3 years ago Hello William,
Thank you for your message and for creating a GP benchmark compatible with sklearn, which sounds like a very nice way to make GP available to the masses! :)
In principle I would be interested to provide a GSGP implementation for this project. However I am afraid I am not going to have time until April after the frenzy teaching term-time is over. So, hopefully this is fine. Also, perhaps I should mention that my GSGP implementation is an early minimal prototype to illustrate how the system works, and how we could address the problem of exponential growth using a neat functional programming approach, rather than an optimised robust implementation suitable for benchmarking, but perhaps being an open source project that everybody can contribute to it I could add my implementation and hope for somebody else to optimise it? :)
Thanks, Alberto
On Wed, 24 Feb 2021 at 15:07, William La Cava notifications@github.com wrote:
hello, thanks for providing a python implementation of GSGP! If you are interested, it would be great to get a sklearn-compatible interface for GSGP so we can benchmark it easily with other methods as part of our benchmark repo https://github.com/EpistasisLab/regression-benchmark. Let me know if you are interested in collaborating on this!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/amoraglio/GSGP/issues/3, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABCZ473YQZ4RAGH2QK3CEDDTAUIZRANCNFSM4YEVVQXQ .
hello, thanks for providing a python implementation of GSGP! If you are interested, it would be great to get a sklearn-compatible interface for GSGP so we can benchmark it easily with other methods as part of our benchmark repo. Let me know if you are interested in collaborating on this!