Open carrieholt opened 2 years ago
Maybe we can connect later in the week to discuss? I think this should be a quick fix, and probably quicker if I trace through to figure out where the problem is. Also wanted to give you updates on other stuff anyway. If this is urgent, go ahead. I won't get to it until Thu at the earliest...
Thanks. I am hoping to work on this earlier rather than later. I'll probably look at it late Thursday or Friday if you don't have a chance before then. Yes let's discuss where we're at this early next week (if not before), and the possibility of moving the repo over to pacific-salmon-assess.
I think this is fixed now. You can try this testing script, then run it on your own data.
Let me know if it runs into any troubles...
I'm getting an error about subsetting columns when running multifit(), which I think is because of the change form perc.change.bm=-25 to perc.change.bm= c(-30, -50, -70). Do you want me to go ahead and trouble shoot in a branch, and then push to the repo? The base function calcPercChangeMCMC still works with the next benchmarks.