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

Add monte-carlo solver #17

Closed AlexLipp closed 1 year ago

AlexLipp commented 1 year ago

Add method to SampleNetwork class that performs a monte carlo estimation of uncertainty on predictions (upstream and downstream) by propagating a provided relative observational error through the unmixing routine.

AlexLipp commented 1 year ago

I've checked that this still works as expected after the squash+rebase. An import declaration had been removed but has been readded.

r-barnes commented 1 year ago

@AlexLipp : I'm still seeing a conflict with unmix_mwe.py, though.

AlexLipp commented 1 year ago

Oh yes sorry if I wasn’t clear. I’ve inspected this conflict directly and it’s trivial. I was going to do it myself but left it for you just in case you had some protocol on conflict resolution and merging. It’s simply the ordering of two commands which is throwing everything out of whack but has not got major consequences/conflicts.

On 18 Nov 2022, at 21:57, Richard Barnes @.***> wrote:



@AlexLipphttps://github.com/AlexLipp : I'm still seeing a conflict with unmix_mwe.py, though.

— Reply to this email directly, view it on GitHubhttps://github.com/r-barnes/faster-unmixer/pull/17#issuecomment-1320571573, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACNXQX6MCNWAJBMSOXNQPUTWI73VBANCNFSM6AAAAAARTJPHDQ. You are receiving this because you were mentioned.Message ID: @.***>