{"id":30526606,"url":"https://github.com/pachadotdev/armadillo","last_synced_at":"2025-09-01T13:01:56.521Z","repository":{"id":311764883,"uuid":"1044515453","full_name":"pachadotdev/armadillo","owner":"pachadotdev","description":"UNOFFICIAL fork of https://gitlab.com/conradsnicta/armadillo-code","archived":false,"fork":false,"pushed_at":"2025-08-25T21:58:24.000Z","size":15569,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-08-26T15:32:14.511Z","etag":null,"topics":["armadillo"],"latest_commit_sha":null,"homepage":"https://gitlab.com/conradsnicta/armadillo-code","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/pachadotdev.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.html","contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-08-25T19:58:34.000Z","updated_at":"2025-08-26T01:20:50.000Z","dependencies_parsed_at":"2025-08-26T15:57:03.115Z","dependency_job_id":null,"html_url":"https://github.com/pachadotdev/armadillo","commit_stats":null,"previous_names":["pachadotdev/armadillo"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/pachadotdev/armadillo","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pachadotdev%2Farmadillo","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pachadotdev%2Farmadillo/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pachadotdev%2Farmadillo/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pachadotdev%2Farmadillo/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pachadotdev","download_url":"https://codeload.github.com/pachadotdev/armadillo/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pachadotdev%2Farmadillo/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":272245498,"owners_count":24899146,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-26T02:00:07.904Z","response_time":60,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["armadillo"],"created_at":"2025-08-27T00:07:10.529Z","updated_at":"2025-08-30T04:02:18.813Z","avatar_url":"https://github.com/pachadotdev.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"### Armadillo: C++ Library for Linear Algebra \u0026 Scientific Computing  \r\nhttps://arma.sourceforge.net\r\n\r\nCopyright 2008-2025 Conrad Sanderson (https://conradsanderson.id.au)  \r\nCopyright 2008-2016 National ICT Australia (NICTA)  \r\nCopyright 2017-2025 Data61 / CSIRO  \r\n\r\n---\r\n\r\n### Quick Links  \r\n\r\n- [download latest stable release](https://arma.sourceforge.net/download.html)\r\n- [documentation for functions and classes](https://arma.sourceforge.net/docs.html)\r\n- [bug reports \u0026 questions](https://arma.sourceforge.net/faq.html)  \r\n\r\n---\r\n \r\n### Contents\r\n\r\n1.  [Introduction](#1-introduction)\r\n2.  [Citation Details](#2-citation-details)\r\n3.  [Distribution License](#3-distribution-license)\r\n\r\n4.  [Prerequisites and Dependencies](#4-prerequisites-and-dependencies)\r\n\r\n5.  [Linux and macOS: Installation](#5-linux-and-macos-installation)\r\n6.  [Linux and macOS: Compiling and Linking](#6-linux-and-macos-compiling-and-linking)\r\n\r\n7.  [Windows: Installation](#7-windows-installation)\r\n8.  [Windows: Compiling and Linking](#8-windows-compiling-and-linking)\r\n\r\n9.  [Support for OpenBLAS and Intel MKL](#9-support-for-openblas-and-intel-mkl)\r\n10. [Support for OpenMP](#10-support-for-openmp)\r\n\r\n11. [Documentation of Functions and Classes](#11-documentation-of-functions-and-classes)\r\n12. [Caveat on use of C++11 auto Keyword](#12-caveat-on-use-of-c11-auto-keyword)\r\n\r\n13. [API Stability and Version Policy](#13-api-stability-and-version-policy)\r\n14. [Bug Reports and Frequently Asked Questions](#14-bug-reports-and-frequently-asked-questions)\r\n\r\n15. [MEX Interface to Octave/Matlab](#15-mex-interface-to-octavematlab)\r\n16. [Related Software Using Armadillo](#16-related-software-using-armadillo)\r\n\r\n---\r\n\r\n### 1. Introduction\r\n\r\nArmadillo is a high quality C++ library for linear algebra and scientific computing,\r\naiming towards a good balance between speed and ease of use.\r\n\r\nIt's useful for algorithm development directly in C++,\r\nand/or quick conversion of research code into production environments.\r\nIt has high-level syntax and functionality which is deliberately similar to Matlab.\r\n\r\nThe library provides efficient classes for vectors, matrices and cubes,\r\nas well as 200+ associated functions covering essential and advanced functionality\r\nfor data processing and manipulation of matrices.\r\n\r\nVarious matrix decompositions (eigen, SVD, QR, etc) are provided through\r\nintegration with LAPACK, or one of its high performance drop-in replacements\r\n(eg. OpenBLAS, Intel MKL, Apple Accelerate framework, etc).\r\n\r\nA sophisticated expression evaluator (via C++ template meta-programming)\r\nautomatically combines several operations (at compile time) to increase speed\r\nand efficiency.\r\n\r\nThe library can be used for machine learning, pattern recognition, computer vision,\r\nsignal processing, bioinformatics, statistics, finance, etc.\r\n\r\nAuthors:\r\n  * Conrad Sanderson - https://conradsanderson.id.au\r\n  * Ryan Curtin      - https://ratml.org\r\n\r\n---\r\n\r\n### 2: Citation Details\r\n\r\nPlease cite the following papers if you use Armadillo in your research and/or software.  \r\nCitations are useful for the continued development and maintenance of the library.\r\n\r\n  * Conrad Sanderson and Ryan Curtin.  \r\n    Armadillo: An Efficient Framework for Numerical Linear Algebra.  \r\n    International Conference on Computer and Automation Engineering, 2025.  \r\n  \r\n  * Conrad Sanderson and Ryan Curtin.  \r\n    Practical Sparse Matrices in C++ with Hybrid Storage and Template-Based Expression Optimisation.  \r\n    Mathematical and Computational Applications, Vol. 24, No. 3, 2019.\r\n\r\n---\r\n\r\n### 3: Distribution License\r\n\r\nArmadillo can be used in both open-source and proprietary (closed-source) software.\r\n\r\nArmadillo is licensed under the Apache License, Version 2.0 (the \"License\").\r\nA copy of the License is included in the \"LICENSE.txt\" file.\r\n\r\nAny software that incorporates or distributes Armadillo in source or binary form\r\nmust include, in the documentation and/or other materials provided with the software,\r\na readable copy of the attribution notices present in the \"NOTICE.txt\" file.\r\nSee the License for details. The contents of the \"NOTICE.txt\" file are for\r\ninformational purposes only and do not modify the License.\r\n\r\n---\r\n\r\n### 4: Prerequisites and Dependencies\r\n\r\nThe functionality of Armadillo is partly dependent on other libraries:\r\n- OpenBLAS (or standard BLAS)\r\n- LAPACK\r\n- ARPACK\r\n- SuperLU\r\n\r\nUse of OpenBLAS (instead of standard BLAS) is strongly recommended on all systems.\r\nOn macOS, the Accelerate framework can be used for BLAS and LAPACK functions.\r\n\r\nIf sparse matrices are not needed, ARPACK and SuperLU are not required.\r\n\r\nArmadillo requires a C++ compiler that supports at least the C++14 standard.\r\n\r\nOn Linux-based systems, install the GCC C++ compiler, which is available as a pre-built package.\r\nThe package name might be `g++` or `gcc-c++` depending on your system.\r\n\r\nOn macOS systems, a C++ compiler can be obtained by first installing Xcode (at least version 8)\r\nand then running the following command in a terminal window:  \r\n\r\n    xcode-select --install\r\n\r\nOn Windows systems, the MinGW toolset or Visual Studio C++ 2019 (MSVC) can be used.\r\n\r\nCaveats on the use of SuperLU:\r\n- SuperLU must be available as a shared library\r\n- Only the following SuperLU versions are supported: 5.2.x, 5.3.x, 6.0.x, 7.0.x\r\n- SuperLU 6.0.x and 7.0.x must be compiled with default integer size (32 bits)\r\n\r\n---\r\n\r\n### 5: Linux and macOS: Installation\r\n\r\nArmadillo can be installed in several ways: either manually or via cmake, with or without root access.\r\nThe cmake based installation is preferred.\r\n\r\nThe cmake tool can be downloaded from https://www.cmake.org\r\nor (preferably) installed using the package manager on your system;\r\non macOS systems, cmake can be installed through MacPorts or Homebrew.\r\n\r\nBefore installing Armadillo, first install OpenBLAS and LAPACK, and optionally ARPACK and SuperLU.\r\nIt is also necessary to install the corresponding development files for each library.\r\nFor example, when installing the `libopenblas` package, also install the `libopenblas-dev` package.\r\n\r\n\r\n#### 5a: Installation via CMake\r\n\r\nThe cmake based installer detects which relevant libraries\r\nare installed on your system (eg. OpenBLAS, LAPACK, SuperLU, ARPACK, etc)\r\nand correspondingly modifies Armadillo's configuration.\r\nThe installer also generates the Armadillo runtime library,\r\nwhich is a wrapper for all the detected libraries.\r\n\r\nChange into the directory that was created by unpacking the armadillo archive\r\n(eg. `cd armadillo-10.6.1`) and then run cmake using:\r\n\r\n    cmake .\r\n\r\n**NOTE:** the full stop (.) separated from `cmake` by a space is important.\r\n\r\nOn macOS, to enable the detection of OpenBLAS, \r\nuse the additional `ALLOW_OPENBLAS_MACOS` option when running cmake:\r\n\r\n    cmake -DALLOW_OPENBLAS_MACOS=ON .\r\n\r\nDepending on your installation, OpenBLAS may masquerade as standard BLAS.\r\nTo detect standard BLAS and LAPACK, use the `ALLOW_BLAS_LAPACK_MACOS` option:\r\n\r\n    cmake -DALLOW_BLAS_LAPACK_MACOS=ON .\r\n\r\nBy default, cmake assumes that the Armadillo runtime library and the corresponding header files \r\nwill be installed in the default system directory (eg. in the `/usr` hierarchy in Linux-based systems).\r\nTo install the library and headers in an alternative directory,\r\nuse the additional option `CMAKE_INSTALL_PREFIX` in this form:\r\n\r\n    cmake . -DCMAKE_INSTALL_PREFIX:PATH=alternative_directory\r\n\r\nIf cmake needs to be re-run, it's a good idea to first delete the `CMakeCache.txt` file\r\n(not `CMakeLists.txt`).\r\n\r\n**Caveat:** if Armadillo is installed in a non-system directory,\r\nmake sure that the C++ compiler is configured to use the `lib` and `include`\r\nsub-directories present within this directory.\r\nNote that the `lib` directory might be named differently on your system.\r\nOn recent 64 bit Debian \u0026 Ubuntu systems it is `lib/x86_64-linux-gnu`.\r\nOn recent 64 bit Fedora \u0026 RHEL systems it is `lib64`.\r\n\r\nIf you have sudo access (ie. root/administrator/superuser privileges)\r\nand didn't use the `CMAKE_INSTALL_PREFIX` option, run the following command:\r\n\r\n    sudo make install\r\n\r\nIf you don't have sudo access, make sure to use the `CMAKE_INSTALL_PREFIX` option\r\nand run the following command:\r\n\r\n    make install\r\n\r\n\r\n#### 5b: Manual Installation\r\n\r\nManual installation involves simply copying the `include/armadillo` header\r\n**and** the associated `include/armadillo_bits` directory to a location\r\nsuch as `/usr/include/` which is searched by your C++ compiler.\r\nIf you don't have sudo access or don't have write access to `/usr/include/`,\r\nuse a directory within your own home directory (eg. `/home/user/include/`).\r\n\r\nIf required, modify `include/armadillo_bits/config.hpp`\r\nto indicate which libraries are currently available on your system.\r\nComment or uncomment the following lines:\r\n\r\n    #define ARMA_USE_LAPACK  \r\n    #define ARMA_USE_BLAS  \r\n    #define ARMA_USE_ARPACK  \r\n    #define ARMA_USE_SUPERLU  \r\n\r\nIf support for sparse matrices is not needed, ARPACK and SuperLU are not necessary.\r\n\r\nNote that the manual installation will not generate the Armadillo runtime library,\r\nand hence you will need to link your programs directly with OpenBLAS, LAPACK, etc.\r\n\r\n---\r\n\r\n### 6: Linux and macOS: Compiling and Linking\r\n\r\nIf you have installed Armadillo via the cmake installer,\r\nuse the following command to compile your programs:\r\n\r\n    g++ prog.cpp -o prog -O2 -std=c++14 -larmadillo\r\n\r\nIf you have installed Armadillo manually, link with OpenBLAS and LAPACK\r\ninstead of the Armadillo runtime library:\r\n\r\n    g++ prog.cpp -o prog -O2 -std=c++14 -lopenblas -llapack\r\n\r\nIf you have manually installed Armadillo in a non-standard location,\r\nsuch as `/home/user/include/`, you will need to make sure \r\nthat your C++ compiler searches `/home/user/include/` \r\nby explicitly specifying the directory as an argument/option. \r\nFor example, using the `-I` switch in GCC and Clang:\r\n\r\n    g++ prog.cpp -o prog -O2 -std=c++14 -I /home/user/include/ -lopenblas -llapack\r\n\r\nIf you're getting linking issues (unresolved symbols),\r\nenable the `ARMA_DONT_USE_WRAPPER` option:\r\n\r\n    g++ prog.cpp -o prog -O2 -std=c++14 -I /home/user/include/ -DARMA_DONT_USE_WRAPPER -lopenblas -llapack\r\n\r\nIf you don't have OpenBLAS, on Linux change `-lopenblas` to `-lblas`;\r\non macOS change `-lopenblas -llapack` to `-framework Accelerate`\r\n\r\nThe `examples` directory contains a short example program that uses Armadillo.\r\n\r\nWe recommend that compilation is done with optimisation enabled,\r\nin order to make best use of the extensive template meta-programming\r\ntechniques employed in Armadillo.\r\nFor GCC and Clang compilers use `-O2` or `-O3` to enable optimisation.\r\n\r\nFor more information on compiling and linking, see the Questions page: \r\nhttps://arma.sourceforge.net/faq.html\r\n\r\n---\r\n\r\n### 7: Windows: Installation\r\n\r\nThe installation is comprised of 3 steps:\r\n\r\n* Step 1:\r\n  Copy the entire `include` folder to a convenient location\r\n  and tell your compiler to use that location for header files\r\n  (in addition to the locations it uses already).\r\n  Alternatively, the `include` folder can be used directly.\r\n\r\n* Step 2:\r\n  If required, modify `include/armadillo_bits/config.hpp`\r\n  to indicate which libraries are currently available on your system:\r\n\r\n    #define ARMA_USE_LAPACK  \r\n    #define ARMA_USE_BLAS  \r\n    #define ARMA_USE_ARPACK  \r\n    #define ARMA_USE_SUPERLU  \r\n\r\n  If support for sparse matrices is not needed, ARPACK or SuperLU are not necessary.\r\n\r\n* Step 3:\r\n  Configure your compiler to link with LAPACK and BLAS\r\n  (and optionally ARPACK and SuperLU).\r\n  Note that OpenBLAS can be used as a high-performance substitute\r\n  for both LAPACK and BLAS.\r\n\r\n---\r\n\r\n### 8: Windows: Compiling and Linking\r\n\r\nWithin the `examples` folder, the MSVC project named `example1_win64`\r\ncan be used to compile `example1.cpp`.\r\nThe project needs to be compiled as a 64 bit program:\r\nthe active solution platform must be set to x64, instead of win32.\r\n\r\nThe MSVC project was tested on Windows 10 (64 bit) with Visual Studio C++ 2019.\r\nAdaptations may be required for 32 bit systems, later versions of Windows and/or the compiler.\r\nFor example, options such as `ARMA_BLAS_LONG_LONG` and `ARMA_BLAS_UNDERSCORE`,\r\ndefined in `include/armadillo_bits/config.hpp`, may need to be either enabled or disabled.\r\n\r\nThe folder `examples/lib_win64` contains a copy of lib and dll files\r\nobtained from a pre-compiled release of OpenBLAS:\r\nhttps://github.com/OpenMathLib/OpenBLAS/releases  \r\nThe compilation was done by a third party.  USE AT YOUR OWN RISK.\r\n\r\n**Caveat:** \r\nfor any high performance scientific/engineering workloads,\r\nwe strongly recommend using a Linux-based operating system, such as:\r\n  * Fedora       https://fedoraproject.org/\r\n  * Ubuntu       https://www.ubuntu.com/\r\n  * Alma Linux   https://almalinux.org/\r\n  * Rocky Linux  https://rockylinux.org/\r\n\r\n---\r\n\r\n### 9: Support for OpenBLAS and Intel MKL\r\n\r\nArmadillo can use OpenBLAS or Intel Math Kernel Library (MKL) as high-speed\r\nreplacements for BLAS and LAPACK. In essence this involves linking with the\r\nreplacement libraries instead of BLAS and LAPACK.\r\n\r\nMinor modifications to `include/armadillo_bits/config.hpp` may be required\r\nto ensure Armadillo uses the same integer sizes and style of function names\r\nas used by the replacement libraries. Specifically, the following defines\r\nmay need to be enabled or disabled:\r\n\r\n    ARMA_USE_WRAPPER  \r\n    ARMA_BLAS_LONG_LONG  \r\n    ARMA_DONT_USE_FORTRAN_HIDDEN_ARGS  \r\n    ARMA_BLAS_UNDERSCORE  \r\n    ARMA_BLAS_CAPITALS  \r\n\r\nSee the documentation for more information on the above defines.\r\n\r\nOn Linux-based systems, MKL might be installed in a non-standard location such as `/opt`\r\nwhich can cause problems during linking.\r\nExamples: `/opt/intel/oneapi/mkl/latest/lib`, `/opt/intel/mkl/lib/intel64/`.\r\n\r\nBefore installing Armadillo, the system must know where the MKL libraries are located.\r\nThis can be achieved via several ways:\r\n\r\n1. By setting the `LD_LIBRARY_PATH` environment variable.\r\n\r\n2. By adding the MKL library directory locations to the `/etc/ld.so.conf` text file,\r\n   followed by running `/sbin/ldconfig`.\r\n\r\n3. By creating a text file named `/etc/ld.so.conf.d/mkl.conf`\r\n   which contains the MKL library directory locations,\r\n   followed by running `/sbin/ldconfig`.\r\n\r\nIf MKL is installed and it is persistently giving problems during linking,\r\nSupport for MKL can be disabled by editing the CMakeLists.txt file,\r\ndeleting CMakeCache.txt and re-running the cmake based installation.\r\nComment out the line containing:\r\n\r\n    INCLUDE(ARMA_FindMKL)\r\n\r\n---\r\n\r\n### 10: Support for OpenMP\r\n\r\nArmadillo can use OpenMP to automatically speed up computationally\r\nexpensive element-wise functions such as exp(), log(), cos(), etc.\r\nThis requires a C++ compiler with OpenMP 4.0+ support.\r\n\r\nFor GCC and Clang compilers, use the following option to enable OpenMP:\r\n`-fopenmp`\r\n\r\n---\r\n\r\n### 11: Documentation of Functions and Classes\r\n\r\nThe documentation of Armadillo functions and classes is available at:  \r\nhttps://arma.sourceforge.net/docs.html\r\n\r\nThe documentation is also in the `docs.html` file distributed with Armadillo.\r\nUse a web browser to view it.\r\n\r\n---\r\n\r\n### 12: Caveat on use of C++11 auto Keyword\r\n\r\nUse of the C++11 `auto` keyword is not recommended with Armadillo objects and expressions.\r\n\r\nArmadillo has a template meta-programming framework which creates short-lived temporaries\r\nthat are not properly handled by `auto`.\r\n\r\n---\r\n\r\n### 13: API Stability and Version Policy\r\n\r\nEach release of Armadillo has its public API (functions, classes, constants)\r\ndescribed in the accompanying API documentation (docs.html) specific\r\nto that release.\r\n\r\nEach release of Armadillo has its full version specified as A.B.C,\r\nwhere A is a major version number, B is a minor version number, and C is a patch level.\r\nThe version specification has explicit meaning\r\n(similar to [Semantic Versioning](https://semver.org/)), as follows:\r\n\r\n* Within a major version (eg. 10), each minor version has a public API that\r\n  strongly strives to be backwards compatible (at the source level) with the\r\n  public API of preceding minor versions. For example, user code written for\r\n  version 10.0 should work with version 10.1, 10.2, etc.\r\n  However, subsequent minor versions may have more features (API additions and extensions)\r\n  than preceding minor versions. As such, user code _specifically_\r\n  written for version 10.2 may not work with 10.1.\r\n\r\n* An increase in the patch level, while the major and minor versions are retained,\r\n  indicates modifications to the code and/or documentation which aim to fix bugs\r\n  without altering the public API.\r\n\r\n* We don't like changes to existing public API and strongly prefer not to break\r\n  any user software. However, to allow evolution, the public API in future major versions\r\n  while remaining backwards compatible in as many cases as possible\r\n  (eg. major version 11 may have slightly different public API than major version 10).\r\n\r\n**CAVEAT:**\r\nthe above policy applies only to the public API described in the documentation.\r\nAny functionality within Armadillo which is _not explicitly_ described\r\nin the public API documentation is considered as internal implementation detail,\r\nand may be changed or removed without notice.\r\n\r\n---\r\n\r\n### 14: Bug Reports and Frequently Asked Questions\r\n\r\nArmadillo has gone through extensive testing and has been successfully\r\nused in production environments. However, as with almost all software,\r\nit's impossible to guarantee 100% correct functionality.\r\n\r\nIf you find a bug in the library or the documentation, we are interested\r\nin hearing about it. Please make a _small_ and _self-contained_ program\r\nwhich exposes the bug, and then send the program source and the bug description\r\nto the developers. The small program must have a main() function and use only\r\nfunctions/classes from Armadillo and the standard C++ library (no other libraries).\r\n\r\nThe contact details are at:  \r\nhttps://arma.sourceforge.net/contact.html\r\n\r\nFurther information about Armadillo is on the frequently asked questions page:  \r\nhttps://arma.sourceforge.net/faq.html\r\n\r\n---\r\n\r\n### 15: MEX Interface to Octave/Matlab\r\n\r\nThe `mex_interface` folder contains examples of how to interface\r\nOctave/Matlab with C++ code that uses Armadillo matrices.\r\n\r\n---\r\n\r\n### 16: Related Software Using Armadillo\r\n\r\n* MLPACK: extensive library of machine learning algorithms  \r\n  https://mlpack.org\r\n\r\n* ensmallen: C++ library for numerical optimisation (L-BFGS, SGD, CMA-ES, etc)  \r\n  https://ensmallen.org/\r\n\r\n* RcppArmadillo: integration of Armadillo with R  \r\n  https://dirk.eddelbuettel.com/code/rcpp.armadillo.html\r\n\r\n* CARMA: interface between Armadillo and Python / NumPy  \r\n  https://github.com/RUrlus/carma\r\n\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpachadotdev%2Farmadillo","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpachadotdev%2Farmadillo","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpachadotdev%2Farmadillo/lists"}