search

r-barnes / faster-unmixer

A faster implementation of the sediment inverse/unmixing scheme proposed in Lipp et al (2021).
5 stars 1 forks source link

Nicer Input Format #9

Closed r-barnes closed 1 year ago

r-barnes commented 1 year ago

@AlexLipp - our input data is divided between fitted_samp_locs_no_dupes and samples.dat.

Is this a natural divide on your end? It feels unnatural to handle it in src/faster-unmixer.cpp since it could require the user to break up their data if it's in two separate files, so I'd prefer that we require merged data.

Does that seem reasonable to you?

AlexLipp commented 1 year ago

Sorry, missed this when first opened. No I think this is an unnatural divide and really just a relict from some (poor) file-management of mine during the early stages of my PhD. In general I think its reasonable to expect an input observation dataset to be in just one file of form:

Sample name | x-coordinate | y-coordinate | element1 | element2 | ... | elementN

r-barnes commented 1 year ago

@AlexLipp : I've address this in #25 - could you review?

AlexLipp commented 1 year ago

Closing this unless you have any objections @r-barnes