The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
R package umap provides an interface to uniform manifold approximation and projection (UMAP) algorithms. There are now several implementations, including versions of python package umap-learn. This vignette explains some aspects of interfacing with the python package.
(For general information on usage of package umap, see the introductory vignette.)
As prep, let’s load the package and prepare a small dataset.
library(umap)
iris.data <- iris[, grep("Sepal|Petal", colnames(iris))]The basic command to perform dimensional reduction is umap. By default, this function uses an implementation written in R. To use the python package umap-learn instead, that package and its dependencies must be installed separately (see python package index or the package source). The R package reticulate is also required (use install.packages('reticulate') and library('reticulate')).
After completing installations, the python implementation is activated by specifying method="umap-learn".
library(reticulate)
iris.umap_learn <- umap(iris.data, method="umap-learn")(This command is not actually executed in the vignette because umap-learn may not be available on the rendering system. If umap-learn is available, the command should execute quietly and create a new object iris.umap_learn that contains an embedding.)
As covered in the introductory vignette, tuning parameters can be set via a configuration object and via explicit arguments in the umap function call. The default configuration is accessible as object umap.defaults.
umap.defaults## umap configuration parameters
## n_neighbors: 15
## n_components: 2
## metric: euclidean
## n_epochs: 200
## input: data
## init: spectral
## min_dist: 0.1
## set_op_mix_ratio: 1
## local_connectivity: 1
## bandwidth: 1
## alpha: 1
## gamma: 1
## negative_sample_rate: 5
## a: NA
## b: NA
## spread: 1
## random_state: NA
## transform_state: NA
## knn: NA
## knn_repeats: 1
## verbose: FALSE
## umap_learn_args: NANote the entry umap_learn_args toward the end. It is set to NA by default. This indicates that appropriate arguments will be selected automatically and passed to umap-learn.
After executing dimensional reduction, the output object contains a copy of the configuration with the values actually used to produce the output. We can examine the effective configuration that was used for our embedding.
iris.umap_learn$config(Again, this command is not executed in the vignette because umap-learn may not be available on the rendering system. When umap-learn is available, this should produce a configuration printout.)
The entry for umap_learn_args should contain a vector of all the arguments passed from the configuration object to the python package. An entry in the configuration should also reveal the version of the python package used to perform the calculation.
A configuration object can contain many components, but not all may be used in a calculation. To verify that a setting is actually passed to umap-learn, ensure that it appears in umap_learn_args in the output.
As an example, consider setting foo and n_epochs during the function call.
## (not evaluated in vignette)
iris.foo <- umap(iris.data, method="umap-learn", foo=4, n_epochs=100)
iris.foo$configInspecting the output configuration will reveal that both foo and n_epochs are recorded (in the latter case, the default value is replaced by the new value). However, foo should not appear in umap_learn_args. This means that foo was not actually passed on to umap-learn.
Various version of umap-learn take different parameters as input. The R package is coded to work with umap-learn versions 0.2, 0.3, 0.4, and 0.5. It will adjust arguments automatically to suit those versions.
Note, however, that some arguments that are acceptable in new versions of umap-learn are not set in the default configuration object. To use those features (see python package documentation), set the appropriate arguments manually, either by preparing a custom configuration object or by specifying the arguments during the umap function call.
It is possible to set umap_learn_args manually while calling umap.
## (not evaluated in vignette)
iris.custom <- umap(iris.data, method="umap-learn",
umap_learn_args=c("n_neighbors", "n_epochs"))
iris.custom$configHere, only the two specified arguments have been passed on to the calculation.
Summary of R session:
sessionInfo()## R version 4.1.1 (2021-08-10)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 20.04.5 LTS
##
## Matrix products: default
## BLAS: /software/opt/R/R-4.1.1/lib/libRblas.so
## LAPACK: /software/opt/R/R-4.1.1/lib/libRlapack.so
##
## locale:
## [1] LC_CTYPE=en_GB.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_GB.UTF-8 LC_COLLATE=C
## [5] LC_MONETARY=en_GB.UTF-8 LC_MESSAGES=en_GB.UTF-8
## [7] LC_PAPER=en_GB.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] umap_0.2.10.0
##
## loaded via a namespace (and not attached):
## [1] Rcpp_1.0.7 knitr_1.34 magrittr_2.0.1 lattice_0.20-44
## [5] R6_2.5.1 rlang_1.0.2 fastmap_1.1.0 highr_0.9
## [9] stringr_1.4.0 tools_4.1.1 grid_4.1.1 data.table_1.14.0
## [13] xfun_0.26 png_0.1-7 cli_3.3.0 RSpectra_0.16-0
## [17] jquerylib_0.1.4 htmltools_0.5.2 askpass_1.1 openssl_2.0.5
## [21] yaml_2.2.1 digest_0.6.27 Matrix_1.5-3 sass_0.4.0
## [25] evaluate_0.14 rmarkdown_2.11 stringi_1.7.4 compiler_4.1.1
## [29] bslib_0.3.0 reticulate_1.21 jsonlite_1.7.2
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.