You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'd like to be able to explore the data in a TTree via the Web UI.
Here's what I mean:
Column (=branch) selection (show me colums A, C, Q, but not the other 50 columns)
Row (=entry) selection (apply simple cuts like A > sin(B)
Diplay table nicely using one of the many Javascript table packages
Stats view summarizing the min, max, mean, std, quantiles of the data like here
Export current column and row selection as CSV and ROOT (and HDF5 or FITS or SQL database if appropriate package is there)
I want this to work for tables that easily fit into memory (i.e. a few GB on my machine), probably by reading the full tree into a recarray in memory once and then keeping it there.
There should be input fields or other web UI elements for the column and row selection, I don't think this will work well with the URL only.
I don't have time to do this now, but wanted to mention it already in case someone else has thoughts or time to work on this.
The text was updated successfully, but these errors were encountered:
Okay. There is a lot of stuff here. You should split it into separate digestible issues as we start attacking them (which we can turn into PRs). I suggest you create only one separate issue at a time (in priority order) rather than spamming issues and just overwhelming. I'm rather paranoid about weboot's (and rootpy's, for that matter) issues becoming an infinite list of things people want and none of them getting done, which is why I'm rather close-happy on them, even if they're not fixed.
Row selection already works by saying !select, though !csv might need to be made aware of it.
Column selection already works since you can specify expressions to be drawn in !draw and !scan. Maybe you mean you want to add it to !csv?
I'm fine with the idea of doing a summary. Implementation welcome
Exporting is fine, too.
Reading the full tree into memory as a recarray is an interesting idea. This collides with #44. WebOOT at some points needs a clearer notion of what is being cached and where all the memory is going, and possibly allowing the user to specify "that this expensive thing" should be cached.
I'd like to be able to explore the data in a TTree via the Web UI.
Here's what I mean:
A > sin(B)
I want this to work for tables that easily fit into memory (i.e. a few GB on my machine), probably by reading the full tree into a recarray in memory once and then keeping it there.
There should be input fields or other web UI elements for the column and row selection, I don't think this will work well with the URL only.
I don't have time to do this now, but wanted to mention it already in case someone else has thoughts or time to work on this.
The text was updated successfully, but these errors were encountered: