This function computes the weighted mean of the specified columns in the training data based on the distances from the average points in the non-linear dimensionality reduction (NLDR) space.
Arguments
- training_data
A data frame containing the training data with an ID column.
- nldr_df_with_id
A data frame containing 2D embeddings with an unique identifier.
- hb_object
An object containing information about hexbin IDs.
- column_start_text
The starting text of the column names in the training_data that should be considered for the weighted mean. Default is "x".
Examples
num_bins_x <- 4
shape_value <- 1.833091
hexbin_data_object <- extract_hexbin_centroids(nldr_df = s_curve_noise_umap,
num_bins = num_bins_x, shape_val = shape_value)
hexdf_data <- hexbin_data_object$hexdf_data
hb_object <- hexbin_data_object$hb_data
weighted_highD_data(training_data = s_curve_noise_training,
nldr_df_with_id = s_curve_noise_umap, hb_object = hb_object, column_start_text = "x")
#> # A tibble: 16 × 8
#> hb_id x1 x2 x3 x4 x5 x6 x7
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 -0.987 1.38 -1.16 0.00156 0.0112 -0.0293 0.00135
#> 2 2 -0.518 1.74 -1.82 0.00856 -0.00304 -0.00211 -0.00253
#> 3 6 -0.412 0.389 -1.86 -0.000733 -0.000295 -0.0369 0.000191
#> 4 7 0.584 1.26 -1.77 0.0123 -0.00998 -0.00123 -0.000406
#> 5 11 -0.175 0.0562 -1.98 0.00204 -0.00179 0.0112 -0.00111
#> 6 12 0.572 0.0605 -1.76 0.00273 0.0107 0.0308 -0.00430
#> 7 13 0.980 0.552 -1.04 0.00143 -0.00296 0.0207 0.00641
#> 8 17 0.713 0.406 -0.298 -0.0133 0.0127 -0.0697 -0.00924
#> 9 18 0.763 1.47 -0.394 -0.00223 0.000814 0.0307 -0.000439
#> 10 24 -0.0511 0.999 0.00835 0.0118 0.00227 0.00598 0.00329
#> 11 28 -0.506 1.92 0.138 0.0159 -0.0149 -0.0841 0.00453
#> 12 29 -0.900 1.13 0.620 -0.00675 -0.00196 0.0394 0.000203
#> 13 38 -0.613 1.69 1.79 0.0119 -0.00418 -0.000519 -0.00251
#> 14 39 -0.223 0.498 1.89 -0.00291 0.00145 -0.00323 -0.000516
#> 15 44 0.590 1.70 1.73 0.00468 -0.00115 -0.0293 0.00182
#> 16 45 0.806 0.604 1.50 -0.00503 0.00522 -0.00644 -0.00350