Skip to contents

Data

The example data sets.

s_curve_noise
S-curve dataset with noise dimensions
s_curve_noise_training
S-curve dataset with noise dimensions for training
s_curve_noise_test
S-curve dataset with noise dimensions for test
s_curve_noise_umap
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap_scaled
Scaled UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap_predict
Predicted UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap2
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap3
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap4
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap5
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_noise_umap6
UMAP embedding for S-curve dataset which with noise dimensions
s_curve_obj
Object for S-curve dataset

Data preprocessing

These are for data preprocessing.

gen_scaled_data()
Scaling the NLDR data

Hexagonal binning

These are for hexagonal binning.

calc_bins_y()
Calculate the effective number of bins along x-axis and y-axis
gen_centroids()
Generate centroid coordinate
gen_hex_coord()
Generate hexagonal polygon coordinates
assign_data()
Assign data to hexagons
compute_std_counts()
Compute standardize counts in hexagons
hex_binning()
Hexagonal binning

2D model construction

These are to construct the 2D model.

extract_hexbin_centroids()
Extract hexagonal bin centroids coordinates and the corresponding standardise counts.
extract_hexbin_mean()
Extract hexagonal bin mean coordinates and the corresponding standardize counts.
tri_bin_centroids()
Triangulate bin centroids
gen_edges()
Generate edge information

Parameters

These are to compute default parameter values.

compute_mean_density_hex()
Compute mean density of hexagonal bins
find_low_dens_hex()
Find low-density Hexagons

Lift to high dimensions

The function is to lift the constructed 2D model into high-dimensions.

avg_highd_data()
Create a dataframe with averaged high-dimensional data

Fit the model

The function is to fit the model.

fit_highd_model()
Construct the 2D model and lift into high-D

Model summaries

These are to obtain model summaries.

glance()
Generate evaluation metrics
predict_emb()
Predict 2D embeddings
augment()
Augment Data with Predictions and Error Metrics

2D visualisation

These are to visualise the model in 2D.

GeomHexgrid
GeomHexgrid: A Custom ggplot2 Geom for Hexagonal Grid
stat_hexgrid()
stat_hexgrid Custom Stat for hexagonal grid plot
geom_hexgrid()
Create a hexgrid plot
GeomTrimesh
GeomTrimesh: A Custom ggplot2 Geom for Triangular Meshes
stat_trimesh()
stat_trimesh Custom Stat for trimesh plot
geom_trimesh()
Create a trimesh plot
vis_rmlg_mesh()
Visualize triangular mesh after removing the long edges

High-dimensional visualisation

These are to visualise the model in high-dimensional space.

comb_data_model()
Create a dataframe with averaged high-dimensional data and high-dimensional data
show_langevitour()
Visualize the model overlaid on high-dimensional data

Projection

These are to generate 2D projections from tour.

gen_axes()
Generate Axes for Projection
get_projection()
Compute Projection for High-Dimensional Data
plot_proj()
Plot Projected Data with Axes and Circles

These are to diagnoise with interactivity.

comb_all_data_model()
Create a dataframe with averaged high-dimensional data and high-dimensional data, non-linear dimension reduction data
show_link_plots()
Visualize the model overlaid on high-dimensional data along with 2D wireframe model.
comb_all_data_model_error()
Create a dataframe with averaged high-dimensional data and high-dimensional data, non-linear dimension reduction data, model error data
show_error_link_plots()
Visualize the model overlaid on high-dimensional data along with 2D wireframe model and error.

Additional

These are additional functionalities.

find_pts()
Find points in hexagonal bins
find_non_empty_bins()
Find the number of bins required to achieve required number of non-empty bins.
get_min_indices()
Get indices of all minimum distances