based on contributor covenant

do you have any plan to support these options via your interface?

See the preprint by Kaji, Sudo, and Ahara. While the interface is in Python, the source code is in C++. Rather than several single-engine porting packages, it may be good to port several related C++ engines to R in one package, and this cubical Ripser looks like a suitable candidate!

Where are the infinite length persistence bars? I tried out ripserr on a simple uniformly random point cloud, and get:

> phom
# A tibble: 119 × 3
   dimension birth    death
       <int> <dbl>    <dbl>
 1         0     0 0.000583
 2         0     0 0.000980
 3         0     0 0.00102 
 4         0     0 0.00206 
 5         0     0 0.00224 
 6         0     0 0.00228 
 7         0     0 0.00249 
 8         0     0 0.00282 
 9         0     0 0.00295 
10         0     0 0.00315 

Notably,

> max(phom$death)
[1] 0.0377396

But unless you are deliberately computing reduced (co)homology, you should expect to see at least one infinite bar in dimension 0. Are you quietly dropping all the infinite bars?!

Another experiment, with a circle point cloud:

> circle.phom = vietoris_rips(circle.cloud)
> circle.phom
PHom object containing persistence data for 200 features.

Contains:
* 199 0-dim features
* 1 1-dim feature

Radius/diameter: min = 0; max = 1.7333.

but if I restrict max-radius:

> circle.phom = vietoris_rips(circle.cloud, threshold=1.5)
> circle.phom
PHom object containing persistence data for 199 features.

Contains:
* 199 0-dim features

Radius/diameter: min = 0; max = 0.15284.

I would like to strongly urge you to implement tracking and reporting of infinite bars.

I have installed ripserr 0.2.0 on two different computers, one Ubuntu x86_64 compute server and on a MacBook Pro (running arm64).

The call signature for vietoris_rips differs between the two installations, requiring dim= on the Mac and max_dim= on the server to set the upper homological dimension limit. As a result, I cannot move scripts between the two computers.

While experimenting with Jose Bouza's tda-tools package, some possible improvements suggest themselves:

• Rather than a single matrix or data frame, it would save space to store persistence data as a list of 2-column matrices. Dimension would be inferred from position in the list, and metadata (e.g. extended persistence) could be stored in a separate list with same-dimension members. This assumes that TDA uses only nonnegative integer dimensions, which is unlikely to change any time soon.
• With the above chance, a coercion method as.data.frame() would recover the current form of output (without the 'PHom' class).
• Parameters like the maximum dimension and the distance threshold should be retained in the output object, as list elements or (less likely) as attributes. These are important for interpretation and may be used in downstream steps like plotting or calculating landscapes.
• Helper functions could be written to access birth–death pairs for a single dimension. These would be useful in analysis pipelines.

see corresponding Advanced R section

likely error in converting C++ code to Rcpp

Would it be possible (and how much work would it be) to do computations with coefficients in Z/3?

TDAstats will eventually import ripserr, circular dependency will likely cause an issue

Related engine based on Ripser for calculation of directed flag complexes; preprint at https://arxiv.org/pdf/1906.10458.pdf

stats > 4.0 requires R > 4.0 - would lower version of stats work?

• vietoris-rips phom w/ Ripser
• cubical phom w/ Cubical Ripser

Found on c++ package branch here: https://github.com/Ripser/ripser/tree/representative-cycles

I'm trying to do it but its not my skillset

Fixes clang-ASAN and r-oldrel-macos issues

Hi! My name is Xinyi and I am an undergraduate student. We are trying to use the ripserr package to produce homology features for a regression model. When we try to load a dataset containing x and y coordinates, an error occurs when the data set has only one single point. Experimentally, we think the package should return an empty data frame in this case. Hope you find this is helpful and could fix the issue. Thank you!

library("ripserr")
#> Warning: package 'ripserr' was built under R version 4.0.4
singlePoint <- data.frame(x = c(1), y = c(1))
rips <- as.data.frame(vietoris_rips(singlePoint));rips
#> Error in vietoris_rips(singlePoint): Point cloud must have at least 2 points and at least 2 dimensions.
#> Error in eval(expr, envir, enclos): object 'rips' not found
doublePoint <- data.frame(x = c(1,2), y = c(1,2))
rips <- as.data.frame(vietoris_rips(doublePoint));rips
#>   dimension birth    death
#> 1         0     0 1.414214

^{Created on 2021-05-31 by the reprex package (v2.0.0)}

The devel parameter max_dim replaced the parameter dim in the current CRAN release. Before the next CRAN submission, this and possibly other parameters should include corrections and warnings for backward compatibility.

Allow user to quit in the middle of Rcpp/C++ code (Rcpp::checkUserInterrupt)

can be used by other TDA R packages to identify type of persistence output to more easily plot (or calculate inference, etc.); could be as simple as adding elements to class of variable? likely distinct for Ripser and Cubical Ripser, can keep adding as Ripser-based engines are added

For Ripser & Cubical Ripser, should test more than the 2 basic tests currently included (checking nrow/ncol/# features in each dimension for cubical ripser 2dim & 4dim)

See https://cran.r-project.org/web/checks/check_results_ripserr.html

?check more edge cases