Package: quollr 0.1.8

quollr: Visualising How Nonlinear Dimension Reduction Warps Your Data

To construct a model in 2D space from 2D embedding data and then lift it to the high-dimensional space. Additionally, it provides tools to visualize the model in 2D space and to overlay the fitted model on data using the tour technique. Furthermore, it facilitates the generation of summaries of high-dimensional distributions.

Authors:Jayani P.G. Lakshika [aut, cre], Dianne Cook [aut], Paul Harrison [aut], Michael Lydeamore [aut], Thiyanga S. Talagala [aut]

quollr_0.1.8.tar.gz
quollr_0.1.8.zip(r-4.5)quollr_0.1.8.zip(r-4.4)quollr_0.1.8.zip(r-4.3)
quollr_0.1.8.tgz(r-4.4-any)quollr_0.1.8.tgz(r-4.3-any)
quollr_0.1.8.tar.gz(r-4.5-noble)quollr_0.1.8.tar.gz(r-4.4-noble)
quollr_0.1.8.tgz(r-4.4-emscripten)quollr_0.1.8.tgz(r-4.3-emscripten)
quollr.pdf |quollr.html✨
quollr/json (API)
NEWS

# Install 'quollr' in R:

install.packages('quollr', repos = c('https://jayanilakshika.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/jayanilakshika/quollr/issues

Datasets:

s_curve_noise - S-curve dataset with noise dimensions
s_curve_noise_test - S-curve dataset with noise dimensions for test
s_curve_noise_training - S-curve dataset with noise dimensions for training
s_curve_noise_umap - 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_umap_scaled - Scaled UMAP embedding for S-curve dataset which with noise dimensions

On CRAN:

31 exports 1 stars 2.46 score 77 dependencies 7 scripts 156 downloads

Last updated 29 days agofrom:14e3c59bca. Checks:OK: 7. Indexed: yes.

Target	Result	Date
Doc / Vignettes	OK	Aug 19 2024
R-4.5-win	OK	Aug 19 2024
R-4.5-linux	OK	Aug 19 2024
R-4.4-win	OK	Aug 19 2024
R-4.4-mac	OK	Aug 19 2024
R-4.3-win	OK	Aug 19 2024
R-4.3-mac	OK	Aug 19 2024

Exports:assign_data augment avg_highd_data cal_2d_dist calc_bins_y compute_mean_density_hex compute_std_counts extract_hexbin_centroids extract_hexbin_mean find_lg_benchmark find_low_dens_hex find_non_empty_bins find_pts fit_highd_model gen_centroids gen_edges gen_hex_coord gen_proj_langevitour gen_scaled_data geom_hexgrid geom_trimesh get_min_indices glance hex_binning predict_emb show_langevitour stat_hexgrid stat_trimesh tri_bin_centroids vis_lg_mesh vis_rmlg_mesh

Dependencies:assertthat base64enc bslib cachem cli codetools colorspace cpp11 crosstalk deldir digest dplyr evaluate fansi farver fastmap fontawesome fs furrr future generics ggplot2 globals glue gtable highr htmltools htmlwidgets interp isoband jquerylib jsonlite knitr labeling langevitour lattice lazyeval lifecycle listenv magrittr MASS Matrix memoise mgcv mime munsell nlme parallelly pillar pkgconfig proxy purrr R6 RANN rappdirs RColorBrewer Rcpp RcppEigen rlang rmarkdown rsample sass scales slider stringi stringr tibble tidyr tidyselect tinytex utf8 vctrs viridisLite warp withr xfun yaml

Data preprocessing

Rendered fromquollr1dataprocessing.Rmdusingknitr::rmarkdownon Aug 19 2024.

Last update: 2024-05-08
Started: 2024-02-29

Algorithm for visualising the model overlaid on high-dimensional data

Rendered fromquollr2algo.Rmdusingknitr::rmarkdownon Aug 19 2024.

Last update: 2024-05-17
Started: 2024-02-29

Algorithm for binning data

Rendered fromquollr3hexbin.Rmdusingknitr::rmarkdownon Aug 19 2024.

Last update: 2024-07-22
Started: 2024-02-29

Generating model summaries

Rendered fromquollr4summary.Rmdusingknitr::rmarkdownon Aug 19 2024.

Last update: 2024-05-13
Started: 2024-02-29

Quick start

Rendered fromquollr5quickstart.Rmdusingknitr::rmarkdownon Aug 19 2024.

Last update: 2024-05-16
Started: 2024-03-19

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Assign data to hexagons	assign_data
Augment Data with Predictions and Error Metrics	augment
Create a dataframe with averaged high-dimensional data	avg_highd_data
Calculate 2D Euclidean distances between vertices	cal_2d_dist
Calculate the effective number of bins along x-axis and y-axis	calc_bins_y
Compute mean density of hexagonal bins	compute_mean_density_hex
Compute standardize counts in hexagons	compute_std_counts
Extract hexagonal bin centroids coordinates and the corresponding standardise counts.	extract_hexbin_centroids
Extract hexagonal bin mean coordinates and the corresponding standardize counts.	extract_hexbin_mean
Compute a benchmark value to remove long edges	find_lg_benchmark
Find low-density Hexagons	find_low_dens_hex
Find the number of bins required to achieve required number of non-empty bins.	find_non_empty_bins
Find points in hexagonal bins	find_pts
Construct the 2D model and lift into high-D	fit_highd_model
Generate centroid coordinate	gen_centroids
Generate edge information	gen_edges
Generate hexagonal polygon coordinates	gen_hex_coord
Visualize a specific projection of langevitour	gen_proj_langevitour
Scaling the NLDR data	gen_scaled_data
Create a hexgrid plot	geom_hexgrid
Create a trimesh plot	geom_trimesh
GeomHexgrid: A Custom ggplot2 Geom for Hexagonal Grid	GeomHexgrid
GeomTrimesh: A Custom ggplot2 Geom for Triangular Meshes	GeomTrimesh
Get indices of all minimum distances	get_min_indices
Generate evaluation metrics	glance
Hexagonal binning	hex_binning
Predict 2D embeddings	predict_emb
S-curve dataset with noise dimensions	s_curve_noise
S-curve dataset with noise dimensions for test	s_curve_noise_test
S-curve dataset with noise dimensions for training	s_curve_noise_training
UMAP embedding for S-curve dataset which with noise dimensions	s_curve_noise_umap
Predicted UMAP embedding for S-curve dataset which with noise dimensions	s_curve_noise_umap_predict
Scaled UMAP embedding for S-curve dataset which with noise dimensions	s_curve_noise_umap_scaled
Visualize the model overlaid on high-dimensional data	show_langevitour
stat_hexgrid Custom Stat for hexagonal grid plot	stat_hexgrid
stat_trimesh Custom Stat for trimesh plot	stat_trimesh
Triangulate bin centroids	tri_bin_centroids
Visualize triangular mesh with coloured long edges	vis_lg_mesh
Visualize triangular mesh after removing the long edges	vis_rmlg_mesh