This repo is a starter layout for an EBS pollock assessment wrapper that uses SPoRC as a pinned dependency. The intent is to keep SPoRC as the modeling engine while this repo holds data conditioning, configs, and reporting.
Proposed layout:
- config/
- base.yml
- scenarios.yml
- R/
- config.R
- build_inputs.R
- data-raw/
- build_ebs_pollock_data.R
- data/
- bering_2area.rds
- analysis/
- run_scenarios.R
- reporting/
Quick start:
- Pin SPoRC (renv or remotes) and update config/base.yml
sporc.ref. - Build data with data-raw/build_ebs_pollock_data.R and save data/bering_2area.rds. The default script saves the SPoRC example dataset; replace it with your data conditioning workflow.
- Optional: add Region B catches to data-raw/region_b_catch.csv (columns: year, catch).
- Run analysis/run_scenarios.R to fit the scenario grid.
The minimal configuration schema lives in config/base.yml and config/scenarios.yml. The scenarios file mirrors the selectivity random effects settings from the vignette.
Note: base.yml stores steepness on the h scale by default. If you prefer to pass the logit-transformed parameter directly, set recruitment.steepness_is_h to false.