Open serbinsh opened 8 years ago
@dlebauer when you were working up BIOCRO did you notice anything like this?
I've seen the 'cannot create directory' errors in the last few weeks but if I recall I 'solved' it by changing permissions to allow write access. I didn't realize it was a bigger problem. On Mon, Apr 4, 2016 at 4:01 PM Shawn P. Serbin notifications@github.com wrote:
@dlebauer https://github.com/dlebauer when you were working up BIOCRO did you notice anything like this?
— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/PecanProject/pecan/issues/776#issuecomment-205495066
I suspect the code is balking to run a single parameter PFT because R is automatically changing the type of a data object somewhere from data.frame or matrix to vector.
@mdietze Yeah, I think you are correct. Will have to go back further in the code to see if we can suss this type change out. However, running with two priors works fine.
I'll be adding instructions in the "Adding an Ecosystem" page in gitbook to associate two priors with a pft for a given model before writing write.configs.
Also, leave this open for reference.
This issue is stale because it has been open 365 days with no activity.
Running a model with a single prior / param seems to result in a improperly formated trait.samples() object to write.config.MODEL.
Note that despite only running with Vcmax the object expects two params resulting in a blank value or the median being copied together with the quantile value. If I then add, say, SLA to the pft then the expected behavior is the result
There is no more NULL when trying to display print(trait.values[[settings$pfts$pft$name]]) ##HAACK OR traits <- convert.samples.MAAT(trait.samples = trait.values[[settings$pfts$pft$name]]) print(traits)
Which is this function: