DOI-USGS / streamMetabolizer

streamMetabolizer uses inverse modeling to estimate aquatic metabolism (photosynthesis and respiration) from time series data on dissolved oxygen, water temperature, depth, and light.
http://usgs-r.github.io/streamMetabolizer/
Other
36 stars 22 forks source link

what does "overall warnings" mean? #362

Closed Kelsman28 closed 6 years ago

Kelsman28 commented 6 years ago

I have fit my two day model with mle and bayes. For the bayes model when I call "get_params" the warnings column says "overall warnings". From plot_DO_preds the model appears to fit DO and DO_sat fine but the dDO/dt is kinda wonky during the day.

Is there somewhere on the specs page I can find out more specifics about this error? https://rdrr.io/github/USGS-R/streamMetabolizer/man/specs.html ?

This is what I get when I specifically call:

get_params(bayes_fit) %>% select(date, warnings, errors) date warnings errors

1 2016-07-16 overall warnings data don't start when expected 2 2016-07-17 overall warnings 3 2016-07-18 overall warnings 4 2016-07-19 overall warnings data don't end when expected predict_metab(bayes_fit) %>% select(date, warnings, errors) date warnings errors 1 2016-07-16 2 2016-07-17 3 2016-07-18 4 2016-07-19
aappling-usgs commented 6 years ago

Overall warnings and errors apply to the entire model rather than any specific date, which is why they get condensed to "overall warnings" in the data.frame returned by get_params. The easiest way to see overall warnings is by calling:

get_fit(bayes_fit)$warnings

and there's a comparable option for overall errors in case those ever come up:

get_fit(bayes_fit)$errors