v0.9.2

What's Changed

• Fix new Parameter crash in R by @Andrea-Havron-NOAA in #1364

Full Changelog: v0.9.1...v0.9.2

v0.9.1

What's Changed

• Adds vignette on FIMSFrame and data_big
• Adds logging information to fims-demo and updates the logging vignette to be
  dynamic instead of relying on a saved image
• Adds a vignette on how to add new C++ modules to FIMS
• Adds WHAM-inspired convergence checks to FIMS models
• Relies on SparseNUTS for Bayesian test
• Refactors FIMSFrame plot method to provide a summary of the data

Full Changelog: v0.8.1...v0.9.1

v0.9.0

What's Changed

• Adds time-varying weight-at-age
• Adds time-varying selectivity
• Makes recruitment deviations random effects by default when using the wrapper functions
• Adds a glossary of mathematical symbols to the doxygen documentation
• Brings doxygen to the pkgdown site
• Fixes hidden bug in dimension_folded_to_tibble() in while loop
• Removes get_log_module
• Removes calculation of uncertainty within the C++ code
• Changed package data name to data_big and added helper data
• Changes m_() functions to model_()

Full Changelog: v0.8.1...v0.9.0

Thank You

We would like to thank the new and continuing contributors that helped with this release, which includes changes from 25 Pull Requests. Way to go team 💪🏅.

v0.8.1

What's Changed

• Two 🐛 were fixed relating to getting the names of the random-effect parameters and not turning off sd reporting.
• The uncertainty of the length-to-age-conversion matrix is no longer being calculated because it is a fixed input as of right now
• Lots of documentation was improved and compiler warnings were decreased by making the code more robust

Full Changelog: v0.8.0...v0.8.1

v0.8.0

What's Changed

• Model time series can now include projections
• Vignette on sparsity with random effects given the parameterization of the model

Full Changelog: v0.7.1...v0.8.0

v0.7.1

What's Changed

• fix(doxygen): adds the correct header to doxygen website in #1062
• fix(doxygen): locations in #1064
• fix(json): allow for mixed types with NA by @msupernaw in #1070
• fix: tibble print formatting issue in vignettes #1012 by @awilnoaa in #1087
• fix(do_sd_report): Skips the sdreport calculation when optimize=FALSE to prevent NaN warning in #1102

Full Changelog: v0.7.0...v0.7.1

v0.7.0

What's Changed

• Turns on the code for model families that allows for more model types in the future than just the catch-at-age model that is currently implemented through the use of FisheryModelBase, which is a unified parent class for all model families. Moved derived quantities (e.g., numbers at age, spawning biomass, catch at age) from Fleet and Population objects into FisheryModelBase. Refactored Population and Fleet classes to be trait only objects.
• Converts nested list of lists used for model control to a tibble with one line per parameter that can be un-nested and nested.
• Increased the documentation of FIMS by using formulas in doxygen, building the doxygen on the NOAA-FIMS github.io site instead of its own web page, writing more information in the testing README, etc.
• Removes seasonality logic given it was not being used.
• Removes date columns in input data in exchange for timing.
• Uses {stockplotr} for the figures in the "fims-demo" vignette.

Full Changelog: v0.6.3...v0.7.0

v0.6.3

• Fix bugs in fims_math.hpp by @Andrea-Havron-NOAA in #946

v0.6.2

• Fix TMBad windows make file by @Andrea-Havron-NOAA in #935
• Fix bug with uninitialized unfished_biomass by @Andrea-Havron-NOAA in #935

Full Changelog: v0.6.1...v0.6.2

v0.6.1

Thanks to work from @Andrea-Havron-NOAA, @msupernaw, and @Bai-Li-NOAA for investigating the bug that has been persistent since v0.3.0 leading to failed builds on R-universe. This fix, implemented by @Andrea-Havron-NOAA, allows for user and operating system information to be found on more operating systems then we had previously implemented.