Skip to contents

Classing specification as a long table, with its file round trip

Source: R/classing.R
scr_classing_spec.Rd

One row per bin of every variable in the lab, optimal and manual, with the authoritative columns a reviewer may edit (lower/upper for numerics, categories/is_other for categoricals, reason) and context columns that are regenerated on read. Open ends are written as NA. scr_classing_read() validates a file back into a spec and scr_classing_import() turns every variable whose bins differ from the lab's current ones into a proposal, so a spreadsheet edit never enters silently.

Usage

scr_classing_spec(lab, file = NULL)

scr_classing_read(file, sep = "%;%")

scr_classing_import(lab, file)

Arguments

lab

An object from scr_coarse_classing(), or an scr_result returned by scr_classing_apply().

file

For scr_classing_spec(), an optional .csv or .xlsx path to write the table to. For scr_classing_read(), the path to read. For scr_classing_import(), a path or an scr_classing_spec object.

sep

Bin separator of the categories column (the configuration's bin_separator). It is validated (no empty category, no category in two bins) and recorded on the spec, so that scr_classing_import() refuses a spec read with a different separator.

Value

A data.frame of class scr_classing_spec.

scr_classing_import() returns a named list of proposals (one per variable whose bins differ from the lab's current ones), each to be accepted or discarded.

See also

scr_coarse_classing() for a complete session, from lab to scorecard.

Other classing: scr_classing_accept(), scr_classing_apply(), scr_classing_choose(), scr_classing_propose(), scr_classing_view(), scr_coarse_classing(), scr_decisions()

Examples

cfg <- scr_config(verbose = FALSE, nthread = 1, use_ranger = FALSE,
                  use_lightgbm = FALSE, xgb_rounds = 40, n_boot = 10)
d <- scr_demo[, c("default", "ref_date", "ds_region", "ds_band", "vl_score_01",
                  "vl_score_02", "vl_score_05", "vl_score_10", "vl_hist_01")]
res <- scr_select(d, "default", config = cfg, date_col = "ref_date")
lab <- scr_coarse_classing(res)
p <- scr_classing_propose(lab, "ds_region",
                          groups = list(edge = c("NORTH", "SOUTH"),
                                        core = c("EAST", "WEST", "CENTRE")))
lab <- scr_classing_accept(lab, p, reason = "edge/core is what pricing uses")
sp <- scr_classing_spec(lab)
sp
#> <scr_classing_spec> 33 bins | 8 variables (1 manual)
#>     variable        type bin_id             bin_label lower upper
#>    ds_region categorical      1         NORTH%;%SOUTH    NA    NA
#>    ds_region categorical      2  EAST%;%WEST%;%CENTRE    NA    NA
#>      ds_band categorical      1                     D    NA    NA
#>      ds_band categorical      2                     C    NA    NA
#>      ds_band categorical      3                     B    NA    NA
#>      ds_band categorical      4                     A    NA    NA
#>  vl_score_01     numeric      1      (-Inf;33.360000]    NA 33.36
#>  vl_score_01     numeric      2 (33.360000;38.150000] 33.36 38.15
#>  vl_score_01     numeric      3 (38.150000;44.240000] 38.15 44.24
#>  vl_score_01     numeric      4 (44.240000;48.060000] 44.24 48.06
#>  vl_score_01     numeric      5 (48.060000;63.940000] 48.06 63.94
#>  vl_score_01     numeric      6 (63.940000;72.610000] 63.94 72.61
#>            categories is_other  source                         reason
#>         NORTH%;%SOUTH    FALSE  manual edge/core is what pricing uses
#>  EAST%;%WEST%;%CENTRE    FALSE  manual edge/core is what pricing uses
#>                     D    FALSE optimal                           <NA>
#>                     C    FALSE optimal                           <NA>
#>                     B    FALSE optimal                           <NA>
#>                     A    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>                  <NA>    FALSE optimal                           <NA>
#>   ... (+21 rows)
# round trip through a file: a fresh lab receives the manual bins as proposals
f <- tempfile(fileext = ".csv")
scr_classing_spec(lab, file = f)
props <- scr_classing_import(scr_coarse_classing(res), scr_classing_read(f))
names(props)
#> [1] "ds_region"
unlink(f)