Skip to contents

Given a test dataset, the centroid coordinates of hexagonal bins in 2D and high-dimensional space, predict the 2D embeddings for each data point in the test dataset.

Usage

predict_emb(highd_data, model_2d, model_highd)

Arguments

highd_data

The test dataset containing high-dimensional coordinates and an unique identifier.

model_2d

Centroid coordinates of hexagonal bins in 2D space.

model_highd

Centroid coordinates of hexagonal bins in high dimensions.

Value

A tibble contains predicted 2D embeddings, ID in the test data, and predicted hexagonal IDs.

Examples

df_bin_centroids <- s_curve_obj$s_curve_umap_model_obj$df_bin_centroids
df_bin <- s_curve_obj$s_curve_umap_model_obj$df_bin
predict_emb(highd_data = s_curve_noise_training, model_2d = df_bin_centroids,
model_highd = df_bin)
#> # A tibble: 3,750 × 4
#>    pred_emb_1 pred_emb_2    ID pred_hb_id
#>         <dbl>      <dbl> <int>      <int>
#>  1      0.272     0.842      1        200
#>  2      0.809     0.484      2        132
#>  3      0.809     0.341      3        102
#>  4      0.272     0.556      5        140
#>  5      0.561     0.770      6        189
#>  6      0.189     0.269      7         79
#>  7      0.685     0.556      9        145
#>  8      0.726     0.341     10        101
#>  9      0.809     0.484     11        132
#> 10      0.231     0.0546    12         35
#> # ℹ 3,740 more rows