Extract the per-target diagnostics frame from a fitted RdistNN object

A tidy wrapper around diagnostics(): the same per-target frame (n_eff, weight concentration, distance radii, tuning provenance, and estimator-specific columns), returned as a plain data.frame with an integer target_id consistent with the other plot_data_*() extractors.

Usage

plot_data_diagnostics(object)

Arguments

object

A fitted result object with a diagnostics() method.

Value

A data.frame with one row per target point; the original diagnostics label is kept in target_label.

Examples

set.seed(1)
n <- 60
x_mat <- matrix(rnorm(n * 2), n, 2)
y_vec <- x_mat[, 1] + rnorm(n, sd = 0.3)
fit <- np_dnn(
  X = x_mat, x0 = c(0, 0), Y = y_vec,
  k = 1L, s = 10L, se = FALSE
)
plot_data_diagnostics(fit)
#>          target_id  n p k  s method distance    n_eff max_weight
#> target_1         1 60 2 1 10  exact       l2 10.54092  0.1666667
#>          top1_weight_share top5_weight_share top10_weight_share
#> target_1         0.1666667         0.6120561          0.8637521
#>          n_nonzero_weight distance_diagnostics_available radius_rank_k
#> target_1               46                          FALSE            NA
#>          radius_rank_s weighted_mean_distance weighted_median_distance
#> target_1            NA                     NA                       NA
#>          tuning_source tuning_name tuning_default_s tuning_grid_available
#> target_1          <NA>        <NA>               NA                    NA
#>          tuning_grid_n_candidates tuning_lambda_default outcome_available
#> target_1                       NA                    NA              TRUE
#>          weighted_mean_y weighted_sd_y weighted_y_q10 weighted_y_q50
#> target_1     0.002292107     0.4238332     -0.5058686     -0.1309296
#>          weighted_y_q90 target_label
#> target_1       0.586555     target_1