dfm / george

Fast and flexible Gaussian Process regression in Python
http://george.readthedocs.io
MIT License
445 stars 128 forks source link

GP with binned data #103

Open temuller opened 5 years ago

temuller commented 5 years ago

I was wondering if there is an implementation or a way to use the existing tools to do GP with binned data for bins with irregular shapes (like a passband). I want to do something similar to what is discussed here https://arxiv.org/pdf/1809.02010.pdf, but for any kernel and non-rectangular bins.

dfm commented 5 years ago

There isn't anything like that implemented in George. In most cases, you'll probably need to integrate the kernel yourself and provide a custom kernel if it can't be written using the existing ones. The docs for implementing custom kernels are here: https://george.readthedocs.io/en/latest/tutorials/new-kernel/

Hope this helps!

On Tue, Nov 6, 2018 at 8:20 AM temuller notifications@github.com wrote:

I was wondering if there is an implementation or a way to use the existing tools to do GP with binned data for bins with irregular shapes (like a passband). I want to do something similar to what is discussed here https://arxiv.org/pdf/1809.02010.pdf https://urldefense.proofpoint.com/v2/url?u=https-3A__arxiv.org_pdf_1809.02010.pdf&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=AvKCBU-zyph1qdf6NaatMQ&m=U2JYYTVXCWwsPU_apBIugbF_UoAPE4HScQ-H6XybZ0A&s=19jJTu5zXnNl7mp-zKADmSJ1t8rxUYRiRcf7NC2yhJU&e=, but for any kernel and non-rectangular bins.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_dfm_george_issues_103&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=AvKCBU-zyph1qdf6NaatMQ&m=U2JYYTVXCWwsPU_apBIugbF_UoAPE4HScQ-H6XybZ0A&s=4TtVztv5YW3lga-N4ZtXNvBFAD36dG-WMcIa82A0EtY&e=, or mute the thread https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AAVYShC6wQHMbQtvkDP3pFmp9UakKt8Lks5usYx7gaJpZM4YQbeZ&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=AvKCBU-zyph1qdf6NaatMQ&m=U2JYYTVXCWwsPU_apBIugbF_UoAPE4HScQ-H6XybZ0A&s=XqA7nXUKZOboZEgHItuBs_A6sOXrQ3rigvKZThZhJuI&e= .

-- Dan Foreman-Mackey Associate Research Scientist Flatiron Institute http://dfm.io

temuller commented 5 years ago

That might help, thanks!