rsquaredacademy/rbin

Tools for binning data

rbin's Issues

Display IV

All print methods should display the information value below the table.

Winsorized

rbin_winsorized() should create bins using winsorized binning.

Custom binning

rb_bin_factor() should bin factor variables.

Flexible intervals

Users should be able to choose between the below intervla:

• left closed and right open
• left open and right closed

Quantiles

rbin_quantiles() should create bins using quantiles.

Visualization

rb_bin_visualize() should create visualization for binning.

Import shiny app from xplorerr

Import the shiny app from the xplorerr package.

Option to return plot object

Return plot objects instead of printing. Use the argument print_plot with the default value TRUE.

Equal length

rbin_equal_length should create bins of equal length.

Addins

Vignette

Error: Argument 2 must be length 3, not 2

when I use rbin_quantiles(data, response, predictors[i], bins=3) on my dataset, there is an error "Error: Argument 2 must be length 3, not 2"
rbin_quantiles(data, response, predictors[i], bins=2) works just fine

summary(Data$CONVERSION) #response
0 1
248996 24912

summary(Data$COUNT_VISITS_6M) #predictors[i]
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.00 2.00 10.64 11.00 707.00

I hope this helps!

Do not write files

CRAN feedback: In rbinAddin() and rbinFactorAddin(), remove options to download data and plots.

0.2.0 Checklist

Prepare for release:

• devtools::check_win_devel()
• rhub::check_for_cran()
• Polish NEWS

Perform release:

• Bump version (in DESCRIPTION and NEWS)
• devtools::check_win_devel() (again!)
• devtools::submit_cran()
• pkgdown::build_site()
• Approve email

Wait for CRAN...

• Tag release
• Bump dev version

Template from r-lib/usethis#338

Error in equal frequency binning

There are two errors in the way equal frequency binning is computed:

• bin count is used in place of bin number (453 instead of 10 in the below example)
• mismatch between the bin number and the metrics (see the bin column in the last table)

library(rbin)

# equal frequency binning
bins <- rbin_equal_freq(mbank, y, age, 10)
bins
#> Binning Summary
#> ------------------------------------
#> Method               Equal Frequency 
#> Response             y 
#> Predictor            age 
#> Bins                 10 
#> Count                4521 
#> Goods                517 
#> Bads                 4004 
#> Entropy              0.51 
#> Information Value    0.01 
#> 
#> 
#>    lower_cut upper_cut bin_count good bad  good_rate         woe           iv
#> 1         18        29       452   55 397 0.12168142 -0.07040317 5.091649e-04
#> 2         29        31       452   57 395 0.12610619 -0.11117177 1.289604e-03
#> 3         31        34       452   46 406 0.10176991  0.13070550 1.623852e-03
#> 4         34        36       452   44 408 0.09734513  0.18007127 3.023706e-03
#> 5         36        39       452   58 394 0.12831858 -0.13109837 1.807071e-03
#> 6         39        42       452   51 401 0.11283186  0.01512953 2.275202e-05
#> 7         42        46       452   45 407 0.09955752  0.15514443 2.266308e-03
#> 8         46        51       452   60 392 0.13274336 -0.17008899 3.087466e-03
#> 9         51        56       452   53 399 0.11725664 -0.02833676 8.116094e-05
#> 10        56        84       453   48 405 0.10596026  0.08567979 7.116156e-04
#>      entropy
#> 1  0.5341748
#> 2  0.5466619
#> 3  0.4745811
#> 4  0.4605229
#> 5  0.5528088
#> 6  0.5083990
#> 7  0.4675914
#> 8  0.5649142
#> 9  0.5214234
#> 10 0.4876093

# plot
plot(bins)

# bins 
bins$bins
#>    lower_cut upper_cut bin bin_count good bad bin_cum_count good_cum_count
#> 1         18        29   2       452   55 397           452             55
#> 2         29        31   7       452   57 395           904            112
#> 3         31        34   4       452   46 406          1356            158
#> 4         34        36   6       452   44 408          1808            202
#> 5         36        39   9       452   58 394          2260            260
#> 6         39        42   1       452   51 401          2712            311
#> 7         42        46   5       452   45 407          3164            356
#> 8         46        51   3       452   60 392          3616            416
#> 9         51        56   8       452   53 399          4068            469
#> 10        56        84 453       453   48 405          4521            517
#>    bad_cum_count   bin_prop  good_rate  bad_rate  good_dist   bad_dist
#> 1            397 0.09997788 0.12168142 0.8783186 0.10638298 0.09915085
#> 2            792 0.09997788 0.12610619 0.8738938 0.11025145 0.09865135
#> 3           1198 0.09997788 0.10176991 0.8982301 0.08897485 0.10139860
#> 4           1606 0.09997788 0.09734513 0.9026549 0.08510638 0.10189810
#> 5           2000 0.09997788 0.12831858 0.8716814 0.11218569 0.09840160
#> 6           2401 0.09997788 0.11283186 0.8871681 0.09864603 0.10014985
#> 7           2808 0.09997788 0.09955752 0.9004425 0.08704062 0.10164835
#> 8           3200 0.09997788 0.13274336 0.8672566 0.11605416 0.09790210
#> 9           3599 0.09997788 0.11725664 0.8827434 0.10251451 0.09965035
#> 10          4004 0.10019907 0.10596026 0.8940397 0.09284333 0.10114885
#>            woe    dist_diff           iv   entropy prop_entropy
#> 1  -0.07040317 -0.007232130 5.091649e-04 0.5341748   0.05340567
#> 2  -0.11117177 -0.011600102 1.289604e-03 0.5466619   0.05465410
#> 3   0.13070550  0.012423746 1.623852e-03 0.4745811   0.04744761
#> 4   0.18007127  0.016791719 3.023706e-03 0.4605229   0.04604211
#> 5  -0.13109837 -0.013784088 1.807071e-03 0.5528088   0.05526865
#> 6   0.01512953  0.001503815 2.275202e-05 0.5083990   0.05082866
#> 7   0.15514443  0.014607733 2.266308e-03 0.4675914   0.04674880
#> 8  -0.17008899 -0.018152061 3.087466e-03 0.5649142   0.05647892
#> 9  -0.02833676 -0.002864157 8.116094e-05 0.5214234   0.05213081
#> 10  0.08567979  0.008305524 7.116156e-04 0.4876093   0.04885800

^{Created on 2023-06-02 by the reprex package (v0.3.0)}

Build

• Travis
• Appveyor
• Code coverage

Tests

Use rlang for errors and warnings

Use rlang equivalents for errors, warnings and messages.

0.1.0 Checklist

Prepare for release:

• Check that description is informative
• Check licensing of included files
• devtools::check_win_devel()
• rhub::check_for_cran()

Perform release:

• Bump version (in DESCRIPTION and NEWS)
• devtools::check_win_devel() (again!)
• devtools::submit_cran()
• pkgdown::build_site()
• Approve email

Wait for CRAN...

• Tag release
• Bump dev version
• Write blog post
• Tweet
• Add link to blog post in pkgdown news menu

Template from r-lib/usethis#338

Use \donttest

CRAN feedback: Use \donttest instead of \dontrun in case of examples with run time > 5s.

Force decreasing trend

rbin_trend_decreasing() will force the variable to follow a monotonically decreasing trend.

Documentation

• README
• NEWS
• Website

Variable binning

Bin continuous variables based on weight of evidence and information value. Users should be able to bin
the variables in the following ways:

• API
• RStudio Addin

README

Visual binning

Explore the features of SPSS visual binning and incorporate them in the RStudio Addin or shiny app.

Save binned data as RDS

User should be able to save the final binned data both as CSV and RDS.

Shiny App

The shiny app for rbin should do the following:

• read data
• allow the user to transform data
  • rename columns
  • modify data type
  • select columns
  • filter rows
• bin multiple variables
• create dummy variables
• download data
• apply binning criteria to test/validation data

Create bins

rb_create_bins() should create binned variables in a data set.

Data

Force increasing trend

rbin_trend_increasing() will force the variable to follow a monotonically increasing trend.

Return entropy

All binning functions should return the bin-wise and total entropy.

Forthcoming release of ggplot2 and rbin

We are contacting you because you are the maintainer of rbin, which imports ggplot2 and uses vdiffr to manage visual test cases. The upcoming release of ggplot2 includes several improvements to plot rendering, including the ability to specify lineend and linejoin in geom_rect() and geom_tile(), and improved rendering of text. These improvements will result in subtle changes to your vdiffr dopplegangers when the new version is released.

Because vdiffr test cases do not run on CRAN by default, your CRAN checks will still pass. However, we suggest updating your visual test cases with the new version of ggplot2 as soon as possible to avoid confusion. You can install the development version of ggplot2 using remotes::install_github("tidyverse/ggplot2").

If you have any questions, let me know!

Test plots

Incorrect package alias

From CRAN:

Dear maintainer,

You have file 'rbin/man/rbin.Rd' with \docType{package}, likely
intended as a package overview help file, but without the appropriate
PKGNAME-package \alias as per "Documenting packages" in R-exts.

This seems to be the consequence of the breaking change

  Using @docType package no longer automatically adds a -package alias.
  Instead document _PACKAGE to get all the defaults for package
  documentation.

in roxygen2 7.0.0 (2019-11-12) having gone unnoticed, see
<https://github.com/r-lib/roxygen2/issues/1491>.

As explained in the issue, to get the desired PKGNAME-package \alias
back, you should either change to the new approach and document the new
special sentinel

  "_PACKAGE"

or manually add

  @aliases rbin-package

if remaining with the old approach.

Please fix in your master sources as appropriate, and submit a fixed
version of your package within the next few months.

Best,
-k

0.1.2 Release Checklist

Prepare for release:

• Check that description is informative
• Check licensing of included files
• devtools::check_win_devel()
• rhub::check_for_cran()

Perform release:

• Bump version (in DESCRIPTION and NEWS)
• devtools::check_win_devel() (again!)
• devtools::submit_cran()
• pkgdown::build_site()
• Approve email

Wait for CRAN...

• Tag release
• Bump dev version
• Write blog post
• Tweet
• Add link to blog post in pkgdown news menu

Template from r-lib/usethis#338

Manual split

rb_bin_manual() should allow users to specify manual binning.

Equal frequency

rbin_equal_freq should create bins with equal frequency.

Bin multiple variables

User should be able to bin multiple variables without having to launch rdinAddin() or rbinFactorAddin() multiple times.

Select data from RStudio

User should be able to select data from RStudio instead of uploading:

• rbinAddin()
• rbinFactorAddin()