Skip to content

Latest commit

 

History

History
161 lines (125 loc) · 5.04 KB

NEWS.md

File metadata and controls

161 lines (125 loc) · 5.04 KB
Raw

tidyAML (development version)

Breaking Changes

None

New Features

  1. Fix #242 - Add function quantile_normalize().
  2. Fix #243 - Add function check_duplicate_rows().

Minor Fixes and Improvements

  1. Fix #239 - Fix erroneou documentation for plot_regression_predictions().

tidyAML 0.0.5

Breaking Changes

None

New Features

  1. Fix #217 - Add function plot_regression_residuals() to plot the residuals.
  2. Fix #215 - Add function plot_regression_predictions() to plot the predictions from the function extract_wflw_predictions()

Minor Fixes and Improvements

  1. Fix #214 - Drop selection message from load_deps()
  2. Fix #222 - Update fast_regression() and fast_classification() to drop NULL predictions.

tidyAML 0.0.4

Breaking Changes

None

New Features

  1. Fix #187 and #198 - Add function extract_regression_residuals()
  2. Fix #199 - Add parameters .drop_na to fast_classification() and fast_regression().

Minor Fixes and Improvements

  1. Fix #186 - Add the following to the core_packages():
    • discrim
    • mda
    • sda
    • sparsediscrim
    • liquidSVM
    • kernlab
    • klaR
  2. Fix #190 - Update the internal_make_wflw_predictions() to include all data
    • The actual data
    • The training predictions
    • The testing predictions

tidyAML 0.0.3

Breaking Changes

  1. Fix #150 - Require R version >= 4.1.0 in order to use the native pipe.
  2. Fix #170 - Drop magrittr from Imports of DESCRIPTION file from #150.

New Features

  1. Fix #157 - internal_make_spec_tbl() now adds a class to each model_spec created by parsnip, for example, a gee engine setting using linear_reg() will return an extra class of gee_linear_reg
  2. Fix #175 - Add Getting Started Vignette.

Minor Fixes and Improvements

  1. Fix #142 - Add gee, glmnet, and rules to the core_packages() function.
  2. Fix #148 - Update create_model_spec()
  3. Fix #155 - Add class tidyaml_base_tbl to the output of create_model_spec()
  4. Fix #156 - Update internal_set_args_to_tune() to use dplyr::pick() instead of dplyr::cur_data() since it was deprecated.
  5. Fix #161 - Update internal_set_args_to_tune() to use !names(new_mod_args) instead of !names(.)
  6. Fix #163 - Add attribute to model spec and drop class from #155
  7. Fix #147 - Add packages to suggests, as they are not necessary for the pkg to work.
  8. Fix #139 - Add suggestion to startup message to run tidymodels::tidymodels_prefer()
  9. Fix #167 - Add function internal_make_wflw_gee_lin_reg()
  10. Fix #168 - Add function full_internal_make_wflw
  11. Fix #169 - Update fast_classification() and fast_regression() to use full_internal_make_wflw()

tidyAML 0.0.2

Breaking Changes

None

New Features

  1. Fix #129 - Add the ability to extract model spec from the create_model_spec() function.

Minor Fixes and Improvements

  1. Fix #130 - Add checks to create_model_spec()
  2. Fix #127 - Fix internal_make_wflw_predictions() to use recipes::testing() instead of recipes::training()

tidyAML 0.0.1

Breaking Changes

None

New Features

  1. Fix #73 - Add function make_regression_base_tbl()
  2. Fix #74 - Add function make_classification_base_tbl()
  3. Fix #77 - Add function internal_make_spec_tbl()
  4. Fix #78 - Add function internal_set_args_to_tune()
  5. Fix #16 - Add function create_workflow_set()
  6. Fix #101 - Add function get_model()
  7. Fix #102 - Add function extract_model_spec()
  8. Fix #103 - Add function extract_wflw()
  9. Fix #104 - Add function extract_wflw_fit()
  10. Fix #105 - Add function extract_wflw_pred()
  11. Fix #71 - Add function match_args()

Minor Fixes and Improvements

  1. Fix #72 - Update fast_classification_parsnip_spec_tbl() and fast_regression_parsnip_spec_tbl() to use the make_regression and make_classification functions.
  2. Fix #79 #80 - Update fast_classification_parsnip_spec_tbl() and fast_regression_parsnip_spec_tbl() to use the internal_make_spec_tbl() function.
  3. Fix #85 - This also addresses sub tasks #86 #87 and #88 to make the base table functions to have a class and to then use that class in internal_make_spec_tbl()
  4. Fix #99 - Update DESCRIPTION for R >= 3.4.0

tidyAML 0.0.0.9002

New Features

  1. Fix #62 - Add function fast_classification_parsnip_spec_tbl()
  2. Fix #65 - Add function fast_classification()

Breaking Changes

None

Minor Fixes and Improvements

None

tidyAML 0.0.0.9001

New Features

  1. Fix #48 - Add functions core_packages() install_deps(), and load_deps()

Breaking Changes

None

Minor Fixes and Improvements

None

tidyAML 0.0.0.9000

New Features

  1. Fix #5 - Add function fast_regression_parsnip_spec_tbl()
  2. Fix #6 - Add function create_splits()
  3. Fix #7 - Add function fast_regression()
  4. Fix #11 - Add function create_model_spec()
  5. Fix #30, #31, 32 - Add internal functions internal_make_wflw(), internal_make_fitted_wflw(), internal_make_wflw_predictions()

Breaking Changes

None

Minor Fixes and Improvements

None

  • Added a NEWS.md file to track changes to the package.