Running Time

[onionpork]

Dear Michael,

Thank you very much for this awesome clustering method.
I am wondering how long does it take for a large dataset, for example 3000+ samples with 2000 variables.
Do you know the the scalability? I am using on Mac M1 chip RAM 8 GB, it took me 10 hrs on GeneratePswarmVisualization, and it is still running.
Could you give suggestion on hardware requirement?

Many thanks.

[Mthrun]

Dear User,
we are aware of the running time issue if the number of cases N rises, and we are working currently on a faster version based on RcppParellel.
The number of dimension d does not influence computation time of the swarm significantly.
On our testing environment N=4000 cases take around 24hours.
If it takes significantly longer in your case, we could compute the results as an excepition if you provide script and data.
Best Regards

[onionpork]

Hi, I realized my data size is too large to upload to github. Could you provide my personal email? I guess I need to put in google drive to share with you. Again many thanks! Best regards, Yirui

…

On Mon, Nov 22, 2021 at 10:11 AM Yirui Chen ***@***.***> wrote: Hey there, Sorry for the late reply. I am testing the possibility of using your amazing clustering method on single-cell RNAseq clustering. I tested on a small published dataset having 3k cells (3k samples). I've attached two .rds, one is non-normalized (2638cells*13714genes), the other is normalized (2638cells*50pcs). If you don't mind, could you also let me know your computational resource? Thank you so much! Best regards, Yirui On Mon, Nov 15, 2021 at 2:00 AM Mthrun ***@***.***> wrote: > Dear User, > we are aware of the running time issue if the number of cases N rises, > and we are working currently on a faster version based on RcppParellel. > The number of dimension d does not influence computation time of the > swarm significantly. > On our testing environment N=4000 cases take around 24hours. > If it takes significantly longer in your case, we could compute the > results as an excepition if you provide script and data. > Best Regards > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#1 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AEAZ4Z52MNGUXHVF2IGQZ2TUMDKZ7ANCNFSM5H5KWZBQ> > . > 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>. > > -- Yirui Chen Graduate Student | Computational Biology Chen's Lab | 5100 Derring Hall Virginia Tech

-- Yirui Chen Graduate Student | Computational Biology Chen's Lab | 5100 Derring Hall Virginia Tech

[parisashahnazari]

Dear Dr Thrun,
Thank you very much for clustering methods. In DBS plots I don't know how can magnitude color of clustering points as well as class labels commands. I would be so appreciate. If you kindly help me in this regards
Kind regards
parisa