### Kalman Filter Example with RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Mentions the addition of a new example demonstrating a Kalman filter implementation using RcppArmadillo, including a C++ version and timing comparisons. This serves as a practical guide for implementing state-space models.

```cpp
Kalman filter
```

--------------------------------

### RcppArmadillo: New fastLm example

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A new example file, fastLm.r, has been added to demonstrate the functionality of fastLm. This example also explores the impact of casting SEXP to Armadillo types.

```R
* inst/examples/fastLm.r: New example fastLm
```

```R
* inst/examples/fastLm.r: Also examine the 'two versus one' casts
        impact when converting SEXP to Armadillo types
```

--------------------------------

### RcppArmadillo Examples with make_list

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Updates examples in src/RcppArmadillo.cpp to use make_list and demonstrates various ways to create named lists and access elements using Rcpp.

```cpp
        * src/RcppArmadillo.cpp: update examples to use make_list and showcase
        various ways to use them : Named(.)=., _[.] = , Named( ., . )
```

--------------------------------

### Add Rank-Deficient Example to fastLm

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The manual page for fastLm (man/fastLm/Rd) is updated with an example demonstrating possible problems when using fastLmPure() and fastLm() with rank-deficient matrices.

```Rd
        * man/fastLm/Rd: Added rank-deficient example to illustrate possible
        problems with fastLmPure() and fastLm()
```

--------------------------------

### RcppArmadillo: New introductory vignette

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A new introductory vignette, RcppArmadillo-intro.Rnw, has been added. This vignette is based on a submitted paper and serves to guide users on how to use the RcppArmadillo package.

```R
* inst/doc/RcppArmadillo-intro.Rnw: New introductory vignette based
        on just-submitted paper by Eddelbuettel and Sanderson
```

--------------------------------

### Install RcppArmadillo from CRAN

Source: https://cran.r-project.org/web/packages/RcppArmadillo/readme/README.html

Use this command to install the package from the nearest CRAN mirror.

```R
install.packages("RcppArmadillo")
```

--------------------------------

### RcppArmadillo.package.skeleton

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Automates the creation of a new R package that intends to use RcppArmadillo features. It extends the functionality of `package.skeleton` by adding necessary RcppArmadillo dependencies and example code.

```APIDOC
## RcppArmadillo.package.skeleton

### Description
`RcppArmadillo.package.skeleton` automates the creation of a new source package that intends to use features of RcppArmadilo. It is based on the package.skeleton function which it executes first.

### Usage
```R
RcppArmadillo.package.skeleton(name = "anRpackage", list = character(),
	environment = .GlobalEnv, path = ".", force = FALSE,
	code_files = character(), example_code = TRUE, author = "Your Name",
        maintainer = if (missing(author)) "Your Name" else author,
        email = "your@email.com", githubuser = NA_character_,
        license = "GPL (>= 2)")
```

### Arguments
`name` |  See package.skeleton  
`list` |  See package.skeleton  
`environment` |  See package.skeleton  
`path` |  See package.skeleton  
`force` |  See package.skeleton  
`code_files` |  See package.skeleton  
`example_code` |  If TRUE, example c++ code using RcppArmadillo is added to the package  
`author` |  Author of the package.  
`maintainer` |  Maintainer of the package.  
`email` |  Email of the package maintainer.  
`githubuser` |  GitHub username for URL and BugReports, if present.  
`license` |  License of the package.  

### Details
In addition to package.skeleton :
The ‘⁠DESCRIPTION⁠’ file gains a Depends line requesting that the package depends on Rcpp and RcppArmadillo and a LinkingTo line so that the package finds Rcpp and RcppArmadillo header files.
The ‘⁠NAMESPACE⁠’, if any, gains a `useDynLib` directive.
The ‘⁠src⁠’ directory is created if it does not exists and a ‘⁠Makevars⁠’ file is added setting the environment variable ‘⁠PKG_LIBS⁠’ to accomodate the necessary flags to link with the Rcpp library.
If the `example_code` argument is set to `TRUE`, example files ‘⁠rcpparma_hello_world.h⁠’ and ‘⁠rcpparma_hello_world.cpp⁠’ are also created in the ‘⁠src⁠’. An R file ‘⁠rcpparma_hello_world.R⁠’ is expanded in the ‘⁠R⁠’ directory, the `rcpparma_hello_world` function defined in this files makes use of the C++ function ‘⁠rcpparma_hello_world⁠’ defined in the C++ file. These files are given as an example and should eventually by removed from the generated package.

### Value
Nothing, used for its side effects 

### References
Read the _Writing R Extensions_ manual for more details.
Once you have created a _source_ package you need to install it: see the _R Installation and Administration_ manual, `INSTALL` and `install.packages`.

### See Also
package.skeleton

### Examples
```R
## Not run:
## RcppArmadillo.package.skeleton( "foobar" )
## End(Not run)
```
```

--------------------------------

### GET /armadillo_version

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Retrieves the version information of the underlying Armadillo library.

```APIDOC
## GET /armadillo_version

### Description
Report the version of the Armadillo library.

### Method
GET

### Endpoint
/armadillo_version

### Parameters
#### Query Parameters
- **single** (logical) - Required - A logical vector indicating whether a single return value is requested, or a named vector with three elements (major, minor, patch).
```

--------------------------------

### Report Armadillo Version

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Use `armadillo_version` to get the Armadillo library version. Specify `single = TRUE` for a single numeric value or `single = FALSE` for a named vector (major, minor, patch). `armadillo_version_typed` returns an S3 object for version comparison.

```R
armadillo_version(single)

armadillo_version_typed()
```

--------------------------------

### Initialize Matrix with Fill Value in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Demonstrates how to initialize a matrix with a specific scalar value using `fill::value()`. This is useful for creating matrices pre-filled with a constant.

```cpp
mat X(4,5,fill::value(123));
```

--------------------------------

### Batch Insertion Constructors for Sparse Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces batch insertion constructors for sparse matrices, improving the efficiency of populating sparse matrices with data.

```cpp
sparse matrices
```

--------------------------------

### Saving and Loading Cubes in HDF5 Format with RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces support for saving and loading cubes in HDF5 format, contributed by Szabolcs Horvat. This is useful for handling multi-dimensional data efficiently.

```cpp
HDF5 format
```

--------------------------------

### Create RcppArmadillo Package Skeleton

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Automates the creation of a new source package configured to use RcppArmadillo.

```R
RcppArmadillo.package.skeleton(name = "anRpackage", list = character(),
	environment = .GlobalEnv, path = ".", force = FALSE,
	code_files = character(), example_code = TRUE, author = "Your Name",
        maintainer = if (missing(author)) "Your Name" else author,
        email = "your@email.com", githubuser = NA_character_,
        license = "GPL (>= 2)")
```

```R
## Not run: 
RcppArmadillo.package.skeleton( "foobar" )

## End(Not run)
```

--------------------------------

### Disable Default Initialization in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Shows how to construct a matrix without initializing its elements to zero by using `fill::none`. This can be beneficial for performance when the initial values are not critical.

```cpp
mat X(4,5,fill::none);
```

--------------------------------

### Include Armadillo Headers and Modify config.hpp

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The Armadillo headers and the 'armadillo_bits' directory are added to 'inst/include/'. Only the 'armadillo_bits/config.hpp' file is modified from the original Armadillo sources.

```cpp
        * inst/include/armadillo: Added the armadillo headers and the
        armadillo_bits directory.  This is from the armadillo sources with
        only the armadillo_bits/config.hpp file modified.
```

--------------------------------

### Construct a matrix with zero fill in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Demonstrates how to specify a fill type during the construction of Armadillo matrix classes like Mat, Col, Row, and Cube, using `fill::zeros` for initialization.

```cpp
mat X(4, 5, fill::zeros)
```

--------------------------------

### Force Approximate Solver in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Explains how to use the `solve_opts::force_approx` option with the `solve()` function to ensure the approximate solver is used. This can improve performance for certain types of matrix problems.

```cpp
solve_opts::force_approx
```

--------------------------------

### Add fastLm Functionality

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Introduces the bare-bones 'fastLm' C++ function in src/RcppArmadillo.cpp, along with an R wrapper in R/fastLm.R, a manual page in man/fastLm.Rd, a unit test, and makes 'fastLm' visible in the NAMESPACE.

```cpp
        * src/RcppArmadillo.cpp: Added bare-bones 'fastLm' function
```

```R
        * R/fastLm.R: Added fastLm R wrapper for C++ fastLm
```

```Rd
        * man/fastLm.Rd: Added manual page
```

```R
        * inst/unitTests/runit.RcppArmadillo.R: Added unit test
```

```R
        * NAMESPACE: Made fastLm visible
```

--------------------------------

### Compute Hessenberg Decomposition at the R Prompt

Source: https://cran.r-project.org/web/packages/RcppArmadillo/readme/README.html

Demonstrates using cppFunction to expose the Armadillo hess() method directly from an R session.

```r
> library(Rcpp)   # for cppFunction()
> cppFunction("arma::mat hd(const arma::mat& m) { return hess(m); }",
              depends="RcppArmadillo")
> set.seed(20251122)
> m <- matrix(rnorm(49), 7, 7)
> hd(m)  # compute Hessenberg decom using the function we just created
        [,1]     [,2]      [,3]      [,4]       [,5]      [,6]      [,7]
[1,] 1.15745  1.64589 -0.809926  0.220417 -1.6050835 -0.947760 -0.972432
[2,] 2.80695 -2.65613  0.119320  0.484858 -0.0877464  0.913440 -0.488142
[3,] 0.00000  1.95335 -1.436409  1.879113 -0.1532293 -0.532077 -0.319914
[4,] 0.00000  0.00000 -1.253221 -1.694414  0.2497111  1.233231  1.055706
[5,] 0.00000  0.00000  0.000000 -3.333499 -1.4800146 -0.913523 -0.667435
[6,] 0.00000  0.00000  0.000000  0.000000 -1.2050704 -0.294369  1.254402
[7,] 0.00000  0.00000  0.000000  0.000000  0.0000000  0.324987 -1.504329
>
```

--------------------------------

### Faster pinv() for Symmetric Positive Definite Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Reports performance improvements for the `pinv()` function when used with symmetric positive definite matrices. This optimization speeds up the computation of the pseudoinverse for this common matrix type.

```cpp
pinv()
```

--------------------------------

### Handle Sparse Matrices with trimatu() and trimatl() in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Details the expanded forms of `trimatu()` and `trimatl()` that now handle sparse matrices and allow for diagonal specification. These functions are used to trim matrices above or below the main diagonal.

```cpp
trimatu()
```

```cpp
trimatl()
```

--------------------------------

### Abridged Printing with .brief_print() in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `.brief_print()` member function for matrices and cubes, which provides an abridged representation. This is useful for quickly inspecting large matrices without overwhelming the console.

```cpp
.brief_print()
```

--------------------------------

### New Sample Function in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Adds a new `sample()` function and associated tests, contributed by Christian Gunning. This provides enhanced capabilities for random sampling.

```cpp
sample()
```

--------------------------------

### Handle Diagonal Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Highlights performance improvements in handling diagonal matrices for functions like `inv_sympd()`, `pinv()`, and `rank()`. Also notes more robust detection of incorrect data formats by `.load()`.

```cpp
inv_sympd()
```

```cpp
pinv()
```

```cpp
rank()
```

```cpp
.load()
```

--------------------------------

### Bare-bones Linear Model Fitting (Pure Interface)

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

The `fastLmPure` function estimates linear models using Armadillo's `solve` function directly. It requires a model matrix `X` and a response vector `y`. Returns coefficients, standard errors, and residual degrees of freedom.

```R
flm <- fastLmPure( cbind(1, log(trees$Girth)), log(trees$Volume) )
```

```R
print(flm)
```

--------------------------------

### Expanded .save() / .load() for Dense Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Details the expansion of `.save()` and `.load()` for dense matrices to support the `coord_ascii` format. This allows for saving and loading matrices in a specific sparse text representation.

```cpp
.save()
```

```cpp
.load()
```

--------------------------------

### Faster Submatrix Views and Element-wise Min/Max in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Highlights performance improvements for accessing submatrix views using `X.cols(first_col, last_col)` and for element-wise `min()` and `max()` operations within compound expressions.

```cpp
X.cols(first_col,last_col)
```

```cpp
min()
```

```cpp
max()
```

--------------------------------

### Checking for Elements Satisfying Conditions in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Shows the usage of `all()` and `any()` functions to check for the presence of elements that satisfy a relational condition within matrices or vectors. Useful for conditional logic and data validation.

```cpp
all()
any()
```

--------------------------------

### POST /fastLmPure

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Estimates a linear model using the Armadillo solve function for high-performance computation.

```APIDOC
## POST /fastLmPure

### Description
Estimates the linear model using the solve function of the Armadillo linear algebra library.

### Method
POST

### Endpoint
/fastLmPure

### Parameters
#### Request Body
- **X** (matrix) - Required - A model matrix.
- **y** (vector) - Required - A vector containing the explained variable.

### Response
#### Success Response (200)
- **coefficients** (vector) - A vector of coefficients.
- **stderr** (vector) - A vector of the estimated standard errors of the coefficient estimates.
- **df.residual** (scalar) - The degrees of freedom in the model.
```

--------------------------------

### Workaround for Win64

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A specific workaround for the Win64 platform is implemented in inst/include/RcppArmadillo.h to address potential compilation or runtime issues.

```cpp
        * inst/include/RcppArmadillo.h: workaround for win64
```

--------------------------------

### Handling Rank Deficiency with fastLm

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Demonstrates a case where `fastLm` may produce an approximate solution for rank-deficient matrices, unlike `lm` which detects and reports it. Use with caution and consider checking matrix condition numbers.

```R
dd <- data.frame(f1 = gl(4, 6, labels = LETTERS[1:4]),
                   f2 = gl(3, 2, labels = letters[1:3]))[-(7:8), ]
```

```R
xtabs(~ f2 + f1, dd)     # one missing cell
```

```R
mm <- model.matrix(~ f1 * f2, dd)
```

```R
kappa(mm)                # large, indicating rank deficiency
```

```R
set.seed(1)
dd$y <- mm %*% seq_len(ncol(mm)) + rnorm(nrow(mm), sd = 0.1)
```

```R
summary(lm(y ~ f1 * f2, dd))     # detects rank deficiency
```

```R
summary(fastLm(y ~ f1 * f2, dd)) # fits all coefficients via approx solution
```

--------------------------------

### Add C++ Flags for SunCC

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A configuration header, inst/include/RcppArmadilloConfig.h, is added to provide C++ configuration flags that help with Sun Studio Compiler (suncc).

```cpp
        * inst/include/RcppArmadilloConfig.h: Addition cconfig helping suncc
```

--------------------------------

### Swapping Matrix Contents in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `.swap()` method for efficiently exchanging the contents of two matrices. This is useful for memory management and avoiding unnecessary copying.

```cpp
`.swap()`
```

--------------------------------

### Handling Non-Square Matrices with QR Decomposition in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Includes fixes for handling non-square matrices with `qr()` and `qr_econ()`. This ensures correct behavior for QR decomposition in various matrix shapes.

```cpp
qr()
qr_econ()
```

--------------------------------

### Faster Median Calculation in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes the performance improvement for the `median()` function, contributed by Ruslan Shestopalyuk. This speeds up statistical calculations on large datasets.

```cpp
median()
```

--------------------------------

### Faster Find Function in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Reports performance improvements for the `find()` function, making it faster for locating elements within matrices.

```cpp
find()
```

--------------------------------

### Faster log_det() for Triangular Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Highlights performance improvements for the `log_det()` function when applied to triangular matrices. This optimization speeds up calculations involving determinants of such matrices.

```cpp
log_det()
```

--------------------------------

### Implement Linear Regression with RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/readme/README.html

A lightweight implementation of linear regression using Armadillo's matrix operations. Use the Rcpp::sourceCpp() function to compile this code.

```cpp
#include <RcppArmadillo/Lighter>
// [[Rcpp::depends(RcppArmadillo)]]

// [[Rcpp::export]]
Rcpp::List fastLm(const arma::mat& X, const arma::colvec& y) {
    int n = X.n_rows, k = X.n_cols;

    arma::colvec coef = arma::solve(X, y);     // fit model y ~ X
    arma::colvec res  = y - X*coef;            // residuals
    double s2 = arma::dot(res, res) / (n - k); // std.errors of coefficients
    arma::colvec std_err = arma::sqrt(s2 * arma::diagvec(arma::pinv(arma::trans(X)*X)));

    return Rcpp::List::create(Rcpp::Named("coefficients") = coef,
                              Rcpp::Named("stderr")       = std_err,
                              Rcpp::Named("df.residual")  = n - k);
}
```

--------------------------------

### RcppArmadillo: Adapt to Armadillo 2.99.1

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This change adapts RcppArmadilloWrap.h to be compatible with Armadillo version 2.99.1. This ensures proper integration with the updated Armadillo library.

```C++
* inst/include/RcppArmadilloWrap.h: adapt to Armadillo 2.99.1
```

--------------------------------

### Sparse Matrix Conversion in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Details the added support for converting sparse matrices of type `dgCMatrix` from the Matrix package. This facilitates interoperability with sparse matrix representations.

```cpp
class `dgCMatrix`
```

--------------------------------

### SSE2 Vectorization of Elementary Expressions in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Highlights automatic SSE2 vectorization for elementary expressions like matrix addition when using GCC 4.7+ with -O3 optimization. This significantly speeds up computations on compatible hardware.

```cpp
matrix addition
```

--------------------------------

### Update to Armadillo 0.9.6

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The Armadillo library is updated to version 0.9.6.

```cpp
        * inst/include/* : updated to armadillo 0.9.6
```

--------------------------------

### Update to Armadillo 0.9.8

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The Armadillo library is updated to version 0.9.8.

```cpp
        * inst/include/*: updated to Armadillo 0.9.8
```

--------------------------------

### Matrix Multiplication with Transpose in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Demonstrates faster multiplication of a matrix with its transpose using Armadillo's optimized functions. This is useful for performance-critical linear algebra operations.

```cpp
X*X.t()
X.t()*X
```

--------------------------------

### fastLm() Generic with Formula Interface

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The fastLm() function in R/fastLm.R is made generic and gains a formula interface. It also includes methods for print, summary, and predict, allowing it to behave like a standard model fitting function.

```R
        * R/fastLm.R: fastLm() is now generic and has a formula interface as
        well as methods for print, summary, predict to behave like a standard
        model fitting function
```

--------------------------------

### Handle Dense Matrix Subviews and Sparse Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes a fix for regressions that occurred when interacting between dense matrix subviews and sparse matrices. This ensures correct behavior in mixed matrix type operations.

```cpp
dense matrix subviews
```

```cpp
sparse matrices
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.2.3

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry details an update to Armadillo version 3.2.3. It also notes a reversion of specific header files related to matrix and column/row types due to compilation issues with g++ 4.7.

```R
* DESCRIPTION: Release 0.3.2.3

        * inst/include/*: Upgraded to new release 3.2.3 of Armadillo

        * inst/include/armadillo_bits/: Reverted files {Mat,Col,Row}_bones.hpp
        to the previous versions due to compilation issues under g++ 4.7
```

--------------------------------

### BibTeX Citation for RcppArmadillo Manual

Source: https://cran.r-project.org/web/packages/RcppArmadillo/citation.html

Use this BibTeX entry to cite the R package manual.

```bibtex
@Manual{,
    title = {RcppArmadillo: 'Rcpp' Integration for the 'Armadillo'
      Templated Linear Algebra Library},
    author = {Dirk Eddelbuettel and Romain Francois and Doug Bates and
      Binxiang Ni and Conrad Sanderson},
    year = {2026},
    note = {R package version 15.2.4-1},
    url = {https://CRAN.R-project.org/package=RcppArmadillo},
    doi = {10.32614/CRAN.package.RcppArmadillo},
  }
```

--------------------------------

### Small Fix for Armadillo 0.8.2

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A minor fix is applied to src/fastLm.cpp to ensure compatibility with Armadillo version 0.8.2.

```cpp
        * src/fastLm.cpp: Small fix for Armadillo 0.8.2
```

--------------------------------

### Update Armadillo to 0.9.52

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This commit updates the Armadillo library to version 0.9.52, codenamed 'Monkey Wrench'. It also includes a change in src/fastLm.cpp to use pinv() instead of inv() for handling singular matrices.

```cpp
        * src/fastLm.cpp: Switch from inv() (which now throws a warning on
        (near) singular matrices and returns an empty matrix) to the
        Moore-Penrose pseudo-inverse pinv() which allows us to continue
        the example from the manual page. Of course, for standard use inv() is
        doing the right thing by issueing the warning.
```

--------------------------------

### RcppArmadillo: Force instantiation of arma version vars

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This change in src/RcppArmadillo.cpp forces the instantiation of Armadillo version variables. This is a workaround to help with older OS X compilers.

```C++
* src/RcppArmadillo.cpp: Force instantiation of arma version vars
        which helps with older OS X compiler, as suggested by Gershon Bialer
```

--------------------------------

### RcppArmadillo Package Skeleton Function

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Introduces a new package skeleton function, RcppArmadillo:::RcppArmadillo.package.skeleton(), which is similar to Rcpp::Rcpp.package.skeleton() but specifically targets the use of RcppArmadillo.

```R
        * R/RcppArmadillo.package.skeleton.R: new package skeleton function
        similar to Rcpp::Rcpp.package.skeleton() but targetting use of
        RcppArmadillo
```

--------------------------------

### RcppArmadillo: Adjust unit test output directory

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The unit tests output directory 'fallback' has been changed to '..', and the test files are now located in the top-level of the $pkg.Rcheck/ directory. This change affects how test results are organized.

```R
* inst/unitTests/runTests.R: unit tests output 'fallback' directory
        changed to '..' and files are now in top-level of $pkg.Rcheck/
```

--------------------------------

### Faster Handling of Subcubes in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes performance improvements in handling subcubes. This optimization benefits operations that involve extracting or manipulating portions of 3D arrays.

```cpp
subcubes
```

--------------------------------

### Remove OS_type Restriction and Add Maintainers

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The OS_type restriction is removed from the DESCRIPTION file now that RcppArmadillo includes Armadillo headers directly. Douglas Bates is added to the Authors and Maintainers list.

```text
        * DESCRIPTION: Remove OS_type restriction now that RcppArmadillo
          includes Armadillo headers, add Doug to Authors and Maintainers
```

--------------------------------

### Use Pivoted Decomposition with chol() in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Describes the expansion of the `chol()` function to optionally use pivoted decomposition. This can be beneficial for numerical stability with certain matrices.

```cpp
chol()
```

--------------------------------

### Fix Alias Handling in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Reports a fix for a regression in alias handling for functions like `fliplr()`, `flipud()`, and `reverse()`. These functions manipulate the order of matrix elements.

```cpp
fliplr()
```

```cpp
flipud()
```

```cpp
reverse()
```

--------------------------------

### Fast Fourier Transform (FFT) in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Adds support for the Fast Fourier Transform (`fft()`) function, enabling efficient frequency-domain analysis of signals and data.

```cpp
fft()
```

--------------------------------

### Update to Armadillo 0.9.10

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

The Armadillo library is upgraded to version 0.9.10. This version introduces support for new 'mtOp' and 'mtGlue' operations.

```cpp
        * inst/include/* : upgrade to armadillo 0.9.10
```

```cpp
        * inst/include/* : support for new mtOp and mtGlue, introduced in
        armadillo 0.9.10
```

--------------------------------

### SCALAR Macro for API Change

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Introduces the SCALAR macro in inst/include/RcppArmadilloDefines.in.h to handle the 0.9.0 API change in Armadillo, ensuring compatibility.

```cpp
        * inst/include/RcppArmadilloDefines.in.h added the SCALAR macro
        to take care of the 0.9.0 api change
```

--------------------------------

### Update fastLm.cpp

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This commit includes a small polish to the src/fastLm.cpp file and moves its content into its own dedicated file.

```cpp
        * src/fastLm.cpp: Moved into its own file, some more polish
```

--------------------------------

### Faster Determinant for Diagonal/Triangular Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes the optimization for calculating the determinant of matrices marked as diagonal or triangular, leading to faster computations for these specific matrix types.

```cpp
determinant
```

--------------------------------

### Imbue and Transform Operations on Submatrices/Subcubes in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Enables `.imbue()` and `.transform()` operations on submatrices and subcubes. This allows for efficient in-place modification of specific parts of matrices and cubes.

```cpp
`.imbue()`
`.transform()`
```

--------------------------------

### Expanded Standalone clamp() for Complex Values in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes that the standalone `clamp()` function has been extended to handle complex numbers. This allows for clamping the real and imaginary parts of complex values independently or jointly.

```cpp
clamp()
```

--------------------------------

### Standard Linear Model Fitting (Formula Interface)

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

The `fastLm` function provides a standard R interface for linear model fitting using formulas. It supports data frames and returns an object similar to `lm`, including residuals and fitted values.

```R
flmmod <- fastLm( log(Volume) ~ log(Girth), data=trees)
```

```R
summary(flmmod)
```

--------------------------------

### RcppArmadillo: Update Armadillo to beta 3.1.92

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update incorporates a beta release of the Armadillo library, version 3.1.92. Beta versions may include new functionalities or fixes that are under testing.

```R
* DESCRIPTION: Release 0.3.1.92

        * inst/include/*: Upgraded to new (beta) release 3.1.92 of Armadillo
```

--------------------------------

### Correct Dimensions for arma::field in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Addresses an issue with correct dimension setting during import/export of `arma::field` types. This fix is protected by a `#define` directive.

```cpp
#define
```

--------------------------------

### Configure Armadillo Random Number Generator

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Sets the internal Armadillo random number generator seed. Requires undefining ARMA_RNG_ALT to function.

```R
armadillo_set_seed_random()

armadillo_set_seed(val)
```

--------------------------------

### Add arma:: Prefix

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

In src/fastLm.cpp, the 'arma::' prefix is added to Armadillo types and functions to ensure proper namespacing and avoid potential conflicts.

```cpp
        * src/fastLm.cpp: added some arma:: prefix
```

--------------------------------

### Control Warning Level in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `ARMA_WARN_LEVEL` configuration option for controlling the verbosity of warning messages emitted by Armadillo. This allows users to suppress warnings for specific issues like failed decompositions.

```cpp
ARMA_WARN_LEVEL
```

--------------------------------

### Faster CSV Loading with OpenMP in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Reports faster loading of CSV files when OpenMP is enabled. This indicates parallel processing improvements for I/O operations.

```cpp
OpenMP
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.0.0

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry marks the release of RcppArmadillo version 0.3.0, which includes an upgrade to Armadillo library version 3.0.0. This is a significant update to the core linear algebra library.

```R
* DESCRIPTION: Release 0.3.0

        * inst/include/*: Upgraded to new release 3.0.0 of Armadillo
```

--------------------------------

### Reshaping Matrices to Vectors in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `vectorise()` function for reshaping matrices into vectors. This is helpful when data needs to be flattened for certain algorithms or operations.

```cpp
vectorise()
```

--------------------------------

### RcppArmadillo: Update Armadillo to beta 3.1.94

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry marks an update to a beta release of Armadillo, version 3.1.94. Beta releases often contain experimental features or early fixes.

```R
* DESCRIPTION: Release 0.3.1.94

        * inst/include/*: Upgraded to new (beta) release 3.1.94 of Armadillo
```

--------------------------------

### Avoid Extra Memory Copy in wrap()

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Optimizes the wrap() function in inst/include/RcppArmadillo.h to avoid an extra memory copy when possible, specifically for 'int' or 'double' element types in eGlue and eOp.

```cpp
        * inst/include/RcppArmadillo.h: avoid an extra memory copy when
        possible (i.e. in wrap( eGlue) and wrap( eOp ) when the elem_type
        is int or double).
```

--------------------------------

### Added log_det_sympd() in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces `log_det_sympd()`, a function specifically for calculating the log determinant of symmetric positive definite matrices. This function is optimized for this matrix type.

```cpp
log_det_sympd()
```

--------------------------------

### Added .clamp() Member Function in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `.clamp()` member function, which restricts the values within a matrix or vector to a specified range. This is useful for data normalization or enforcing bounds.

```cpp
.clamp()
```

--------------------------------

### OpenMP Thread Management

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Functions to report and set the maximum number of OpenMP threads used by Armadillo.

```APIDOC
## Report (or Set) Maximum Number of OpenMP Threads

### Description
Report (or Set) Maximum Number of OpenMP Threads 

### Usage
```R
armadillo_get_number_of_omp_threads()

armadillo_set_number_of_omp_threads(n)
```

### Arguments
`n` |  Number of threads to be set  

### Value
For the getter, and on a system with OpenMP, the maximum number of threads that OpenMP may be using and on systems without it, one. The setter does not return a value.
```

--------------------------------

### Mozilla Public License 2.0 Licensing

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes the change in licensing for Armadillo to the Mozilla Public License 2.0. This affects how the library can be used and distributed.

```text
Mozilla Public License 2.0
```

--------------------------------

### Save/Load CSV with Semicolon Separator in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Illustrates the use of `csv_opts::semicolon` for saving and loading CSV files where the delimiter is a semicolon instead of a comma. This is useful for compatibility with regional settings or specific file formats.

```cpp
csv_opts::semicolon
```

--------------------------------

### Added tgamma() Function in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Introduces the `tgamma()` function, which computes the gamma function. The gamma function is a generalization of the factorial function to real and complex numbers.

```cpp
tgamma()
```

--------------------------------

### RcppArmadillo: Update Armadillo to test release 2.99.1

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update integrates Armadillo test release 2.99.1 into RcppArmadillo. Test releases are often used for pre-release testing and feedback.

```R
* inst/include/*: Upgraded to new test release 2.99.1 of Armadillo
```

--------------------------------

### Patch for MinGW Compiler on Win32

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A small patch is added to inst/include/armadillo_bits/config.hpp to prevent the use of asinh, acosh, and atanh on win32 systems when using an older MinGW compiler that cannot handle them. This issue is resolved in newer R versions.

```cpp
        * inst/include/armadillo_bits: Added small patch to not use asinh,
        acosh and atanh on win32 as the (older) MinGW compiler cannot cope.
        This problem will go away once R 2.12 is released as it requires
        gcc / g++ 4.5.
```

--------------------------------

### Use Shift-Invert Mode with eigs_sym() and eigs_gen() in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Explains the addition of an optional shift-invert mode to `eigs_sym()` and `eigs_gen()`. This mode is useful for finding eigenvalues close to a specific value.

```cpp
eigs_sym()
```

```cpp
eigs_gen()
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.2.0

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This log entry signifies an upgrade to Armadillo version 3.2.0. Such updates often bring performance enhancements and bug fixes from the core Armadillo library.

```R
* DESCRIPTION: Release 0.3.2.0

        * inst/include/*: Upgraded to new release 3.2.0 of Armadillo
```

--------------------------------

### Apply Upstream Corrections for randn and randg in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Mentions the application of upstream corrections for `arma::randn` draws when using alternative generators (like R's) and for `arma::randg`. This ensures more accurate random number generation.

```cpp
arma::randn
```

```cpp
arma::randg
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.2.4

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry indicates an upgrade of the Armadillo C++ linear algebra library to version 3.2.4 within the RcppArmadillo package. This typically involves incorporating new features, bug fixes, or performance improvements from the underlying Armadillo library.

```R
* DESCRIPTION: Release 0.3.2.4

        * inst/include/*: Upgraded to new release 3.2.4 of Armadillo
```

--------------------------------

### RcppArmadillo: Protect ARMA_DEFAULT_OSTREAM define

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This modification to RcppArmadilloConfig.h adds a #ifndef test to protect the define of ARMA_DEFAULT_OSTREAM. This prevents potential redefinition issues.

```C++
* inst/include/RcppArmadilloConfig.h: Protect define of
        ARMA_DEFAULT_OSTREAM by a #ifndef test
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.0.2

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry signifies an update to Armadillo version 3.0.2. Updates to the underlying library often include bug fixes and performance enhancements.

```R
* DESCRIPTION: Release 0.3.0.2

        * inst/include/*: Upgraded to new release 3.0.2 of Armadillo
```

--------------------------------

### Manage OpenMP Threads

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Functions to report or set the maximum number of OpenMP threads used by Armadillo.

```R
armadillo_get_number_of_omp_threads()

armadillo_set_number_of_omp_threads(n)
```

--------------------------------

### Set Armadillo RNG from a given value in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Provides functions to set Armadillo's system RNG from a specified value or randomize it. Note: These functions may not work within RStudio due to RStudio's own use of the system RNG.

```cpp
rcpparma_hello_world()
```

--------------------------------

### RcppArmadillo: Update Armadillo to test release 2.99.3

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry signifies an update to Armadillo test release 2.99.3. Test releases are typically used for pre-release evaluation of the Armadillo library.

```R
* DESCRIPTION: Release 0.2.39

        * inst/include/*: Upgraded to new test release 2.99.3 of Armadillo
```

--------------------------------

### Throttle or Reset Core Usage

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Helper functions to limit or restore the number of cores used by RcppArmadillo-internal code on OpenMP-enabled systems.

```R
armadillo_throttle_cores(n = 2)

armadillo_reset_cores()
```

--------------------------------

### RcppArmadillo: Use Rcpp::Rcout for output

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This change replaces std::cout with Rcpp::Rcout in several Armadillo header files. This ensures that output from Armadillo functions is correctly handled within the R environment.

```C++
* inst/include/armadillo_bits/Cube_meat.hpp : changes related to
        std::cout -> Rcpp::Rcout

        * inst/include/armadillo_bits/Mat_meat.hpp : idem

        * inst/include/armadillo_bits/config.hpp : idem

        * inst/include/armadillo_bits/debug.hpp : idem

        * inst/include/armadillo_bits/field_meat.hpp : idem

        * inst/include/armadillo_bits/forward_bones.hpp : idem
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.0.1

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update incorporates Armadillo version 3.0.1. Such updates typically include bug fixes and performance improvements from the core Armadillo library.

```R
* DESCRIPTION: Release 0.3.0.1

        * inst/include/*: Upgraded to new release 3.0.1 of Armadillo
```

--------------------------------

### Core Throttling for RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/refman/RcppArmadillo.html

Helper functions to throttle or reset the number of cores used by RcppArmadillo-internal code, particularly for systems with OpenMP.

```APIDOC
## Throttle (or Reset) (Rcpp)Armadillo to Two Cores

### Description
Helper functions to throttle use of cores by RcppArmadillo-internal code on systems with OpenMP. On package load, the initial value is saved and used to reset the value. 

### Usage
```R
armadillo_throttle_cores(n = 2)

armadillo_reset_cores()
```
```

--------------------------------

### RcppArmadillo: Add missing semicolon for C++0x

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A missing semicolon was added to a code block within RcppArmadillo/Mat_meat.h that is ifdef'ed for C++0x mode. This corrects a syntax error.

```C++
* inst/include/RcppArmadillo/Mat_meat.h: Add missing semicolon in
        code ifdef'ed for C++0x mode, with thanks to Teo Guo Ci
```

--------------------------------

### RcppArmadillo: Update Armadillo to 2.4.3

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This entry signifies an update to Armadillo version 2.4.3. Updates to the underlying library often include bug fixes and performance enhancements.

```R
* DESCRIPTION: Release 0.2.35 (depending on Rcpp 0.9.10)

        * inst/include/*: Upgraded to new release 2.4.3 of Armadillo
```

--------------------------------

### Fix constexpr for Fixed-Size Matrices in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Corrects the incorrect use of `constexpr` for handling fixed-size matrices and vectors. This ensures proper compile-time evaluation where applicable.

```cpp
constexpr
```

--------------------------------

### Handle Out-of-Bounds Access with std::out_of_range

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Notes that element accessors in RcppArmadillo now throw `std::out_of_range` exceptions for out-of-bounds access, providing more specific error information than the previous `std::logic_error`.

```cpp
std::out_of_range
```

--------------------------------

### RcppArmadillo: Update Armadillo to 3.0.3

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update brings the Armadillo library to version 3.0.3. Such updates typically include bug fixes and performance improvements.

```R
* DESCRIPTION: Release 0.3.0.3

        * inst/include/*: Upgraded to new release 3.0.3 of Armadillo
```

--------------------------------

### RcppArmadillo: Update Armadillo to test release 2.99.2

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update incorporates Armadillo test release 2.99.2. Test releases are often used to gather feedback on new features or fixes before a stable release.

```R
* DESCRIPTION: Release 0.2.38

        * inst/include/*: Upgraded to new test release 2.99.2 of Armadillo
```

--------------------------------

### Rounding Matrix Elements in RcppArmadillo

Source: https://cran.r-project.org/web/packages/RcppArmadillo/news.html

Adds the `round()` function for rounding matrix elements to the nearest integer. This is useful for data discretization or preparing data for specific analyses.

```cpp
round()
```

--------------------------------

### BibTeX Citation for RcppArmadillo Article

Source: https://cran.r-project.org/web/packages/RcppArmadillo/citation.html

Use this BibTeX entry to cite the 2014 Computational Statistics and Data Analysis article.

```bibtex
@Article{,
    title = {RcppArmadillo: Accelerating R with high-performance C++
      linear algebra},
    author = {Dirk Eddelbuettel and Conrad Sanderson},
    journal = {Computational Statistics and Data Analysis},
    year = {2014},
    volume = {71},
    month = {March},
    pages = {1054--1063},
    doi = {10.1016/j.csda.2013.02.005},
  }
```

--------------------------------

### Fix Runtime Error with SCALAR Macro

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

Corrects a runtime error in src/RcppArmadillo.cpp that occurred when using the SCALAR macro, ensuring proper functionality.

```cpp
        * src/RcppArmadillo.cpp : fix a runtime error uisng the SCALAR macro
```

--------------------------------

### RcppArmadillo: Update Armadillo to test release 2.99.4

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update incorporates a test release of Armadillo, version 2.99.4. Test releases are often used for pre-release testing of new features or fixes.

```R
* DESCRIPTION: Release 0.2.40

        * inst/include/*: Upgraded to new test release 2.99.4 of Armadillo
```

--------------------------------

### Compute sig2 via std::inner_product

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

In src/fastLm.cpp, the calculation of sig2 is changed to use std::inner_product for potentially improved precision or performance.

```cpp
        * src/fastLm.cpp: Compute sig2 via std::inner_product
```

--------------------------------

### Add CxxFlags() Function

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

A new function, RcppArmadillo:::CxxFlags(), is added to R/flags.R, likely to provide or manage C++ compiler flags for RcppArmadillo.

```R
        * R/flags.R: added RcppArmadillo:::CxxFlags()
```

--------------------------------

### RcppArmadillo: Update Armadillo to 2.4.4

Source: https://cran.r-project.org/web/packages/RcppArmadillo/ChangeLog

This update incorporates Armadillo version 2.4.4. Such updates typically include bug fixes and performance improvements from the core Armadillo library.

```R
* DESCRIPTION: Release 0.2.36

        * inst/include/*: Upgraded to new release 2.4.4 of Armadillo
```