Here, we’ll walk through the process of preprocessing 2D embedding data to obtain regular hexagons.
First, you’ll need 2D embedding data generated for your training data. For our example, we’ll use a 3-\(d\) S-curve dataset with four additional noise dimensions. We’ve used UMAP as our non-linear dimension reduction technique to generate embeddings for the S-curve data.
scaled_umap <- gen_scaled_data(data = s_curve_noise_umap)
glimpse(scaled_umap)
#> List of 3
#> $ scaled_nldr: tibble [75 × 3] (S3: tbl_df/tbl/data.frame)
#> ..$ UMAP1: num [1:75] 0.0804 0.7386 0.8399 0.1672 0.2629 ...
#> ..$ UMAP2: num [1:75] 0.32 1.0024 1.0836 0.0432 0.3984 ...
#> ..$ ID : int [1:75] 1 2 3 4 6 7 8 9 11 12 ...
#> $ lim1 : num [1:2] -3.27 2.46
#> $ lim2 : num [1:2] -5.74 5.82gen_scaled_data function preprocesses the 2D embedding
data to obtain regular hexagons.