Skip to content

ageproR version 0.8.0

Choose a tag to compare

@efletcherPIFSC efletcherPIFSC released this 23 Jul 21:54
· 61 commits to main since this release
44f3c6e
  • Recruitment
    • Added empirical_ssb for Empirical Recruit Per Spawning Mass Distribution (Recruitment model 2) (#76)
    • Added parametric_autocorrelated_error as a base R6Class for Parametric Recruitment w/ Autocorrelated Lognormal Errors (#75)
      • Added berverton_holt_autocorrelated_error for Beverton-Holt w/ Autocorrelated Lognormal Error (Model 10)
      • Added shepherd_curve_autocorrelated_error for Shepherd Curve w/ Autocorrelated Lognormal Error (Model 11)
      • Added ricker_curve_autocorrelated_error for Ricker Curve w/ Autocorrelated Lognormal Error (Model 12)
    • Recruitment models 2, 10, 11, 12 added to list to valid recruitment numbers, and can be passed through set_recruit_model function.
    • Allow active binding recruitment fields (recruit_scaling_factor, ssb_scaling_factor, and max_recruit_obs) to be settable.
    • Empirical observation tables
      • Added obs_table parameter to set observation data at initialization. Default is NULL, generating default values, the original behavior.
      • Added method to set empirical recruitment observation data from data frames
    • Recruitment cli tweaks
      • cli tweaks improve readability with Recruitment Model and Name on Rconsole.
        • print now shows max_recruit_obs (#71)
      • Rename print_input_option_name -> print_process_error_fields
      • Minor read_inp_lines message tweak to clarify recruitment time period range.
    • Consolidated validate_numeric_substrings code into read_inp_numeric_lines function
    • Added option to toggle verbosity when setting new recruitment models via set_recruit_model
  • Options
    • Consistency cli tweaks between Option classes
    • Renamed options-related files with option prefix and simplified some R6Classes names:
      • renamed R script max_bounds.R -> options_max_bounds.R
      • renamed R script user_percentile_summary.R -> options_percentile_summary.R
        • renamed R6Class user_percentile_summary -> percentile_summary
      • renamed R script reference_points.R -> options_reference_points.R
      • renamed R script scaling_factors.R -> options_scaling_factors.R
      • renamed R script retrospective_adjustments.R -> options_retrospective_adjustments.R
      • renamed R script output_options.R -> options_output.R
        • renamed R6class output_options -> options_output
      • renamed R script optional_options_flags.R -> option_flags.R
        • renamed R6Class optional_optional_flags -> option_flags
    • "options w/ options_flags" (option_flag class reference for: max_bounds, percentile_summary, reference_points, scaling_factors, retrospective_adjustments)
      • Fixed an issue where option_flags reference class was improperly referenced, causing unintended consequences using multiple agepro_model instances (#85)
        • Make option_flag R6class a non-shared Reference, and instantiate reference per agepro_model class instance.
        • During agepro_model instantiation, each "options w/ options_flags" is instantiated with an option_flag class reference.
      • Use formals to detect parameter defaults to determine option_flag is enabled at initialization:
        • max_bounds, percentile_summary, reference_points, scaling_factors: Initialize with a non-default parameter: set its flag to TRUE; If all class parameters match default values, set the flag to FALSE.
        • retrospective_adjustments: removed default value for retro_adjust. If retrospective_adjust is missing in a new instances of this class, set enable_retrospective_adjustments to FALSE. Otherwise, its TRUE
    • Fixup option_flags inconsistencies, messages
  • cli consistency tweaks (R message format)
    • Reworded Rconsole to represent the keyword parameter class field names.
      • Console alert symbols ℹ️ (or i) focuses on keyword parameter classes fields; ◀️ (or <) for a interface event (or input line read)
    • Output from initializing agepro_model classes or "printing" AGEPRO keyword parameters classes, will print under the R message format.
      • This resolves an issue with console output changes with RStudio 2025.05.0+496.
      • Pass print_parameter_table output through capture_output_as_message
        • Replaces use of tibbles for output; Removes tibble dependency.
        • Replaces redundant function cli_print_process_error_table in process_error class.
    • Added verbosity toggle to specifically show General Parameters to R console during agepro_model initialization. TRUE by default.
  • Bootstrap
    • Clarified bootstrap_file warning message for relative path cases
    • cli fix to show Bootstrap File at agepro_model initialization
  • Replace test AGEPRO input File, Bootstrap File, and json input file with AGEPRO-GUI Hawaii Uku Example, based on 2020 Hawaii Uku (Aprion virescens) Assessment.
    • Hawaii Uku JSON input file imported from Hawaii Uku AGEPRO input file.
    • Added description file of example.
  • Added Vignette as "Article" format (#84)
    • Includes AGEPRO Input File Keyword Parameter Structure and stub for JSON experimental file
  • README update (#88)