New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Gadget projections #99
Comments
when recruitment is scheduled in a timestep different from step 1 the following error is returned:
I think the issue is that
|
Actually the recruitment step is not changed, meaning that if the immature stock was recruited at step 3 it will continue to recruit at step 3. These functions, however, only assume one recruitment step. The error you have looks like you haven't yet called gadget on the new model variant, so directly after
ps: stock |
this is the code I'm using (borrowed from the example) and that returns the error
If I check the file
recruitment happens in step 2 in this model like correctly reported by these parameters but judging from the error the function is looking for |
The most recent version I've fixed this issue in |
The fix does the job, great!
I need to specify the
The code above run until the second last line. It prints the files under 'out' and at a quick check they look fine, but when I try to read them with
which is surprising given that I've just updated gadget using the installation via the R package gadget2 - very handy by the way ;) |
Yes, this is my bad. Note that the notation
it will otherwise think that the output from the preceding steps is the output directory and the output directory is supplied into the |
yes, it fix it, thanks! |
I've added a functionality to
You can then just use the
|
I've started the redevelopment of the forward projections functionality with the more flexible
gadget_project_*
functions. The logic behind these functions is that the user defines the projections, unlikegadget.forward
, in a stepwise manner thus giving greater control over individual processes:gadget_project_time
defines the projection horizon, and behind the scene extends the area and time filesgadget_project_stock
links together two stocks in a SSB - Rec relationships by defining a (TimeVariable) stochastic spawning process. Currently only hockey-stick implemented but should accommodate all spawn function quite easily.gadget_project_fleet
defines a projection fleet by lifting the suitability from a already defined fleet in the model. The type of the model can be any of the fleet types in Gadget.gadget_project_recruitment
projects stochastically the recruitment based on a AR(1) regression of the estimated recruitment seriesgadget_project_advice
projects the advice error as a log-normally distributed AR(1) process with a given mean, CV and rhogadget_project_ref_point
defines the reference points for the simulation (currently only Blim) but can in theory update other parameter values for the spawn function. Fleet based reference points (Btrigger) cannot be defined yet due to a shortcoming of GadgetAll these functions could be applied repeatedly if the number of stocks in the model is greater than 2. @vbartolino and @pamelajwoods would you mind giving these functions a go and report on any shortcomings.
The text was updated successfully, but these errors were encountered: