Skip to content
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

Smaller Example Data filesize #68

Closed
efletcherPIFSC opened this issue Jan 9, 2023 · 5 comments
Closed

Smaller Example Data filesize #68

efletcherPIFSC opened this issue Jan 9, 2023 · 5 comments
Assignees
Labels
enhancement New feature or request topic: Stock-Synthesis Specific for Stock-Synthesis Models or its functionality
Milestone

Comments

@efletcherPIFSC
Copy link
Collaborator

❯ checking installed package size ... NOTE
    installed size is 12.4Mb
    sub-directories of 1Mb or more:
      data  11.9Mb

CRAN maximum (installed) package file size limit is 5MB

*.rda Datafile fileslzes

mcmcsimple.rda 152.8 KB
simple.rda 1.1 MB
retroSimple.rda 5.7 MB
@efletcherPIFSC efletcherPIFSC added status: help wanted Extra attention is needed topic: Stock-Synthesis Specific for Stock-Synthesis Models or its functionality labels Jan 9, 2023
@MOshima-PIFSC MOshima-PIFSC self-assigned this Feb 16, 2023
@MOshima-PIFSC MOshima-PIFSC added enhancement New feature or request and removed status: help wanted Extra attention is needed labels Feb 16, 2023
@MOshima-PIFSC
Copy link
Contributor

I don't think we will be able to reduce the data down to a size small enough for CRAN, even if we just submit it separately as a data package. I found this option that allows you to create a GitHub repo for a data package and allows users to install that package regularly using install.packages("datapackage") and then load the data via data("dataset", package="datapackage").
@efletcherPIFSC I'm wondering if it would be helpful to collaborate with the SS/r4ss team on developing this package, that way these example SS runs could be used by their package too? And it could be hosted by either the nmfs-stock-synthesis or r4ss organizations.

@MOshima-PIFSC
Copy link
Contributor

MOshima-PIFSC commented Mar 13, 2023

We have decided to separate the data from the ss3diags package to keep functionality and allow us to publish on CRAN.
Next steps:

  • For testing purposes, replace rda files with the simple stock synthesis input model files into data folder and write code to run the model and retrospectives to create the output for unit tests.
  • Remove rda files in data folder
  • Ask Kathryn about options for hosting the he simple stock synthesis input model files and ss executable to allow people to run the model themselves (for reproducibility)

@MOshima-PIFSC
Copy link
Contributor

MOshima-PIFSC commented Jun 13, 2023

I created the FEAT-update-testing-fixture branch to work on this. Based on communication with the r4ss team, we are going to remove the simple SS model in our data folder and use the simple_small SS model. Since r4ss is a dependency of ss3diags we can call the simple_small Report.sso using system.file("extdata", "simple_small", package= "r4ss").
To run the retrospective and hindcast tests (SSmase_test.R, test-HCxval.R, and test-retro.R), we can add something like this to the test fixtures to run the retrospectives. The SS executable can be retrieved via a github action.

@MOshima-PIFSC
Copy link
Contributor

MOshima-PIFSC commented Jun 16, 2023

Currently tests are running and passing on unbuntu set up. Next step is to add in windows and mac OS to config matrix.
@efletcherPIFSC I created an inst/extdata directory and put the input files in there and am copying the ss exe to there. Then in the tests I am copying all those files to a temp dir and running retrospectives there. If we do this, we should be able to delete the simple.rda, retroSimple.rda, and mcmcSimple.rda objects later on. Before we remove them from the package we will need to update the documentation so that these objects are created at time of knitting the docs. What do you think about this?

@MOshima-PIFSC
Copy link
Contributor

Tests are running and passing on all OS configurations now (ubuntu, windows, and macOS). The next step would be to update the documentation and vignettes to use these model structures. Adding this issue to milestone "Documentation".

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request topic: Stock-Synthesis Specific for Stock-Synthesis Models or its functionality
Projects
Archived in project
Development

No branches or pull requests

2 participants