search

kaizhang / SnapATAC2

Single-cell epigenomics analysis tools
https://kzhang.org/SnapATAC2/
235 stars 26 forks source link

Benchmark clustering methods #17

Closed kaizhang closed 2 years ago

kaizhang commented 2 years ago
beyondpie commented 2 years ago

Maybe we can test about Annoy. Below is my recent observation. https://github.com/beyondpie/SnapATAC/releases/tag/v2.3

kaizhang commented 2 years ago

Yes, I think it is good to have multiple backends for KNN construction. We can use the exact algorithm as the default and provide other algorithms as options.

kaizhang commented 2 years ago

Maybe we can test about Annoy. Below is my recent observation. https://github.com/beyondpie/SnapATAC/releases/tag/v2.3

According to this benchmark: https://github.com/erikbern/ann-benchmarks, pynndescent seems to be the best algorithm.

beyondpie commented 2 years ago

Great to know this!

On Jan 28, 2022, at 8:46 AM, Kai Zhang @.***> wrote:

Maybe we can test about Annoy. Below is my recent observation. https://github.com/beyondpie/SnapATAC/releases/tag/v2.3 https://github.com/beyondpie/SnapATAC/releases/tag/v2.3 According to this benchmark: https://github.com/erikbern/ann-benchmarks https://github.com/erikbern/ann-benchmarks, pynndescent seems to be the best algorithm.

— Reply to this email directly, view it on GitHub https://github.com/kaizhang/SnapATAC2/issues/17#issuecomment-1024406659, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRGSRUKC4ANOYXQ2KFIRSDUYLB5FANCNFSM5MSGKWSA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub. You are receiving this because you commented.