search

edickie / ciftify

The tools of the Human Connectome Project (HCP) adapted for working with non-HCP datasets
https://edickie.github.io/ciftify/
MIT License
117 stars 157 forks source link

ciftify_statclust_report's csv report is inaccurate when other resolution surfaces file served #116

Open YaoMeng94 opened 5 years ago

YaoMeng94 commented 5 years ago

I run the ciftify_statclust_report command with a self-downsampled surface(10k per hemisphere) and the overlap information with the atlases in the csv report looks inaccurate, is this situation about the mismatch between surface option file with the default atlases in ciftify?

edickie commented 5 years ago

Yes, sorry about that, the atlases it's drawing from are in 32k per hemisphere resolution, and there's no internal resampling step.. I'll add extra checks and warnings into the command.

YaoMeng94 commented 5 years ago

Thanks answering, would you consider to add a extra option that users can choose the specific atlas to compute the overlap information.

edickie commented 5 years ago

I always intended to add that..(I tried to build the code with the option of adding it) but I don't know if I figured out what that interface would look like for the user. Any suggestions?

On Mon, Jul 8, 2019 at 9:17 PM DearBraveJohn notifications@github.com wrote:

Thanks answering, would you consider to add a extra option that users can choose the specific atlas to compute the overlap information.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/edickie/ciftify/issues/116?email_source=notifications&email_token=ADEXT5SEZLNBUI2BX6QJQB3P6PRKNA5CNFSM4H24KMLKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZOZGFA#issuecomment-509448980, or mute the thread https://github.com/notifications/unsubscribe-auth/ADEXT5UY4E7F3YFBKH34CALP6PRKNANCNFSM4H24KMLA .