schochastics / edgebundle

R package implementing edge bundling algorithms

 Hi, I'm David! 👋 

I am the Team Lead for "Transparent Social Anlytics" in the Department for Computational Social Science at GESIS

My main dev work is writing R packages and extensions for Quarto

Quarto

Academicons	Shortcodes to use academicons with Quarto
Devicons	Shortcodes to use devicons with Quarto
Blackboard revealjs theme	A blackboard style theme for revealjs
Sketchy HTML theme	A sketchy theme for HTML documents
Social Share Buttons	Add social share buttons to HTML documents
Nutshell	Include expandable explanations in documents
Quartocities	Geocities template for Quarto websites
Glossary with nutshell	Automatic glossary for websites using nutshell
CV templates	CV/Resume templates for Quarto

edgebundle's Issues

Error: datashader.bundling

Thanks for the package. The maps look awesome.

I am having a problem running the code provided on the website. Specifically I am struggling to run the following line:

hbundle <- edge_bundle_hammer(g, xy, bw = 0.7, decay = 0.5)

It returns this error:
Error: Python module datashader.bundling was not found.

I checked I have this datashader installed and it is:

(flowmap) MacBook-Pro-5:flow-map_edgebundle franciscorowe$ conda list
# packages in environment at /Users/franciscorowe/opt/anaconda3/envs/flowmap:
#
# Name                    Version                   Build  Channel
brotli                    1.0.9                    pypi_0    pypi
brotli-python             1.0.9           py311h814d153_9    conda-forge
bzip2                     1.0.8                h0d85af4_4    conda-forge
ca-certificates           2023.7.22            h8857fd0_0    conda-forge
certifi                   2023.7.22          pyhd8ed1ab_0    conda-forge
charset-normalizer        3.2.0              pyhd8ed1ab_0    conda-forge
click                     8.1.6           unix_pyh707e725_0    conda-forge
cloudpickle               2.2.1              pyhd8ed1ab_0    conda-forge
colorcet                  3.0.1              pyhd8ed1ab_0    conda-forge
dask-core                 2023.8.0           pyhd8ed1ab_0    conda-forge
**datashader**               0.15.2             pyhd8ed1ab_0    conda-forge
....

Yet, I get the same error when I run the line of code above.

Any advice on this issue would be appreciated, or general advice on how to get the package working.

My computational set up is:

R version 4.2.2 (2022-10-31) -- "Innocent and Trusting"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-apple-darwin17.0 (64-bit)

Switch to Visvalingam edge straightening

naive R code here:
https://coolbutuseless.github.io/2022/05/07/a-naive-implementation-of-visvalingams-line-simplification-in-r/

NOTE: this will reverse the meaning of epsilon

Kernel density based bundling

http://recherche.enac.fr/~hurter/KDEEB/KDEEB.html

Problems with hammer_bundle

Hi,

First of all, thanks for the great package - stunning viz.

I am trying to replicate the example here, and I am hitting an error running this line:
hbundle <- edge_bundle_hammer(g,xy,bw = 0.7,decay = 0.5)

The error message is:
Error in py_call_impl(callable, dots$args, dots$keywords) : ImportError: hammer_bundle operation requires scikit-image. Ensure you install the dependency before applying bundling.

I thought there was a compatibility issue with scikit-image so I installed a older version scikit-image-0.17.2. However, I keep getting the same issue.

Any ideas on how to resolve the issue?

I am using:

R version 4.0.2 (2020-06-22)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS  10.16

Edge bundling for parallel coordinates

This is a great package. Have you considered adding something for parallel coordinates such as the algorithm described in the attached paper? I'm currently trying to figure out a way to approximate this result with your package. It would be great to have a function designed specifically for parallel coordinates though.

Computer Graphics Forum - 2008 - Zhou - Visual Clustering in Parallel Coordinates.pdf

add some flow map algorithm(s)

some examples: https://geonetworks.github.io/

Can edgebundle be exported as a shapefile?

First of all, I would like to thank you so much for creating this package. I have been waiting (and trying solutions on my own) for edge bundling in R for a while and this really saves my life!

I would like to know if edgebundle can be exported as line shapefiles, or converted to line shapefiles in some ways. I suppose that the group column in fbundle file indicates which line the points should sit in. If so, if I convert all the points under the same group as lines (probably through sf package), can the result be an exported shapefile? What does the index column mean in fbundle dataframe? The reason why I want a shapefile is because I have been using tmap for base map mapping and it is not compatible with ggplot syntax. I would like to use this package to process the data into edgebundles and then plot it using tmap.

Reproduce maps from your code using fbundle

Here is my first trial of trying to convert fbundle to dataframe and plot with tmap that can directly plot shapefile (see codes below). The output seems decently similar to the fbundle in ggplot, but still have some differences that I do not know what went wrong.

library(tmap)
library(ggraph)
library(tidyverse)
library(igraph)
library(sf)
library(edgebundle)

g <- us_flights
xy <- cbind(V(g)$longitude,V(g)$latitude)
verts <- data.frame(x=V(g)$longitude,y=V(g)$latitude)

fbundle <- edge_bundle_force(g,xy,compatibility_threshold = 0.6)

#trying to convert fbundle dataframe to a shapefile#
test = fbundle %>% st_as_sf(coords=c('x', 'y'), crs=4326) %>%
  group_by(group) %>% summarise(index=sum(index)) %>% st_cast("LINESTRING")

tm_shape(test) +
  tm_lines(col='black', lwd=0.1)

Many thanks.
Xiaofan

add edge-path bundling

See example and arxiv paper.
paper also links to an interesting dataset

Error during the installation

Hello mr. David, I'm not able to install your package because this error appears:

library(remotes) 
remotes::install_github("schochastics/edgebundle")
Downloading GitHub repo schochastics/edgebundle@HEAD
✓  checking for file ‘/private/var/folders/nz/cn8n1fgn7y3_t96jpvxplh6c0000gn/T/RtmplmTYGd/remotesa5d59b6e4ef/schochastics-edgebundle-5ee713a/DESCRIPTION’ (404ms)
─  preparing ‘edgebundle’:
✓  checking DESCRIPTION meta-information ...
─  cleaning src
─  checking for LF line-endings in source and make files and shell scripts
─  checking for empty or unneeded directories
   Removed empty directory ‘edgebundle/man/figures’
─  building ‘edgebundle_0.1.0.9000.tar.gz’
   
* installing *source* package ‘edgebundle’ ...
** using staged installation
** libs
clang++ -I"/Library/Frameworks/R.framework/Resources/include" -DNDEBUG  -I'/Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include' -I/usr/local/include   -fPIC  -Wall -g -O2  -c RcppExports.cpp -o RcppExports.o
In file included from RcppExports.cpp:4:
In file included from /Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include/Rcpp.h:57:
/Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include/Rcpp/DataFrame.h:136:18: warning: unused variable 'data' [-Wunused-variable]
            SEXP data = Parent::get__();
                 ^
1 warning generated.
clang++ -I"/Library/Frameworks/R.framework/Resources/include" -DNDEBUG  -I'/Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include' -I/usr/local/include   -fPIC  -Wall -g -O2  -c forceBundle.cpp -o forceBundle.o
In file included from forceBundle.cpp:4:
In file included from /Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include/Rcpp.h:57:
/Library/Frameworks/R.framework/Versions/4.0/Resources/library/Rcpp/include/Rcpp/DataFrame.h:136:18: warning: unused variable 'data' [-Wunused-variable]
            SEXP data = Parent::get__();
                 ^
forceBundle.cpp:28:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector vec = {P[2] - P[0],P[3] - P[1]};
                ^     ~~~~~~~~~~~~~~~~~~~~~~~~~
forceBundle.cpp:44:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector vec = {(Q_source_x + r * (Q_target_x - Q_source_x)),
                ^     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
forceBundle.cpp:54:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector qs = {Q[0],Q[1]};//Q[inds];
                ^    ~~~~~~~~~~~
forceBundle.cpp:55:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector qt = {Q[2],Q[3]};//Q[indt];
                ^    ~~~~~~~~~~~
forceBundle.cpp:60:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector midI = {(I0[0] + I1[0]) / 2.0,
                ^      ~~~~~~~~~~~~~~~~~~~~~~~
forceBundle.cpp:62:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector midP = {(P[0] + P[2]) / 2.0,
                ^      ~~~~~~~~~~~~~~~~~~~~~
forceBundle.cpp:135:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_source =  {P[0],P[1]};//P[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:136:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_target =  {P[2],P[3]};//P[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:137:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_source =  {Q[0],Q[1]};//Q[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:138:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_target =  {Q[2],Q[3]};//Q[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:151:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_source =  {P[0],P[1]};//P[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:152:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_target =  {P[2],P[3]};//P[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:153:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_source =  {Q[0],Q[1]};//Q[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:154:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_target =  {Q[2],Q[3]};//Q[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:169:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_source =  {P[0],P[1]};//P[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:170:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector P_target =  {P[2],P[3]};//P[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:171:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_source =  {Q[0],Q[1]};//Q[inds];
                ^           ~~~~~~~~~~~
forceBundle.cpp:172:17: error: non-aggregate type 'Rcpp::NumericVector' (aka 'Vector<14>') cannot be initialized with an initializer list
  NumericVector Q_target =  {Q[2],Q[3]};//Q[indt];
                ^           ~~~~~~~~~~~
forceBundle.cpp:215:25: error: non-aggregate type 'Rcpp::IntegerVector' (aka 'Vector<13>') cannot be initialized with an initializer list
          IntegerVector ecomp = {oe};
                        ^       ~~~~
fatal error: too many errors emitted, stopping now [-ferror-limit=]
1 warning and 20 errors generated.
make: *** [forceBundle.o] Error 1
ERROR: compilation failed for package ‘edgebundle’
* removing ‘/Library/Frameworks/R.framework/Versions/4.0/Resources/library/edgebundle’
Warning message:
In i.p(...) :
  installation of package ‘/var/folders/nz/cn8n1fgn7y3_t96jpvxplh6c0000gn/T//RtmplmTYGd/filea5d767034ec/edgebundle_0.1.0.9000.tar.gz’ had non-zero exit status

Can you help me in solving this problem? I apologize in advance if this is a trivial problem. I'm a beginner in R but can't wait to use this library. Thank you for your work.

Divided Edge Bundling

http://vis.stanford.edu/papers/divided-edge-bundling

Add weight to edgebundle results

Hi David!

I wonder if there is a way to add weights to the edge bundle results (e.g., fbundle, bundle etc.). The weights represent the volumes in the flow OD network, which is a common characteristic in network visualization. Ideally, I should be able to add weights to igraph object g (e.g., E(g)$weight = c(1,2,3)) and the return data frame from edgebundle will include an additional column that indicates weights by group. I assume that each group represents a line because the total number of groups and the total number of edges are consistent in your US flight example. Currently, I cannot figure out a way to match the bundled edges (or the group) to the original igraph object g, and thus cannot associate weights with the bundled edges.

Can you help with a solution?

Many thanks!
Xiaofan

migrate hammer bundle from python to c++

datashader is too big of a library to depend on. A re-implementation of the algorithm may be a good idea
code: https://datashader.org/_modules/datashader/bundling.html
original: https://gitlab.com/ianjcalvert/edgehammer

Cannot install the package

Hello developer,

Excelent work! However, when I was tying to install the package, it poped out an error message, please check the error messgae below:

> remotes::install_github("schochastics/edgebundle")
Error: Failed to install 'unknown package' from GitHub:
  HTTP error 404.
  No commit found for the ref master

  Did you spell the repo owner (`schochastics`) and repo name (`edgebundle`) correctly?
  - If spelling is correct, check that you have the required permissions to access the repo.

Can you help me with this? Thank you very much.

Handling duplicates in triangulation

I have been trying to plot international trade flows from Aotearoa, New Zealand. I have ensured that there are no duplicates in nodes or edges but I still get the following error.

 Error in interp::tri.mesh(xymesh[, 1], xymesh[, 2]) : 
  duplicate data points 

In the end, I had to make the following change to tnss_tree() for the error to go away.

  #triangulate points
  tria <- interp::tri.mesh(xymesh[,1],xymesh[,2])
  tria <- interp::tri.mesh(xymesh[,1],xymesh[,2], duplicate = "remove")

I will post the code and link to the data in a blog post later this week as it is a little too long for an issue! I would love to see this change in the main package so that I can now plot flows that span the globe from little old New Zealand! Thanks for writing such a wonderful package! :-) I had been looking for something like for a long time!