Ver 0.30.11

* Implement Cholesky Decomposition (#46)
* Add `is_symmetric` method to Matrix

Author: Axect

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "peroxide"
-version = "0.30.10"
+version = "0.30.11"
 authors = ["axect <[email protected]>"]
 edition = "2018"
 description = "Rust comprehensive scientific computation library contains linear algebra, numerical analysis, statistics and machine learning tools with farmiliar syntax"

README.md

@@ -128,6 +128,7 @@ Peroxide can do many things.
     * LU Decomposition, Inverse matrix, Block partitioning
     * QR Decomposition (`O3` feature)
     * Singular Value Decomposition (SVD) (`O3` feature)
+    * Cholesky Decomposition (`O3` feature)
     * Reduced Row Echelon form
     * Column, Row operations
     * Eigenvalue, Eigenvector
@@ -210,7 +211,7 @@ then Rust become great choice.
 
 ## Latest README version
 
-Corresponding to `0.30.2`
+Corresponding to `0.30.11`
 
 ## Pre-requisite
 
@@ -265,7 +266,7 @@ Corresponding to `0.30.2`
 
 ## Useful tips for features
 
-* If you want to use `QR` or `SVD` then should use `O3` feature (there are no implementations for these decompositions in `default`)
+* If you want to use *QR* or *SVD* or *Cholesky Decomposition* then should use `O3` feature (there are no implementations for these decompositions in `default`)
 * If you want to write your numerical results, then use `nc` feature and `netcdf` format. (It is much more effective than `csv` and `json`.)
 * To plot, use `nc` feature to export data as netcdf format and use python to draw plot.
     * `plot` feature has limited plot abilities.

RELEASES.md

@@ -1,3 +1,8 @@
+# Release 0.30.11 (2022-02-10)
+
+* Implement Cholesky decomposition in `O3` feature.
+* Implement symmetricity check method - `is_symmetric` for Matrix. 
+
 # Release 0.30.10 (2022-02-05)
 
 * Update `netcdf` dependency to `0.7`

examples/chebyshev.rs

@@ -3,6 +3,7 @@ use peroxide::fuga::*;
 
 fn main() {
     let mut c = chebyshev_polynomial(0, SpecialKind::First);
+    c.print();
     for i in 1 .. 11 {
         c = chebyshev_polynomial(i, SpecialKind::First);
         c.print();

examples/hessian.rs

@@ -8,6 +8,9 @@ fn main() {
 
     let dy = f(x, y);
     dy.print();
+
+    let dy0 = f(x0, y0);
+    dy0.print();
 }
 
 fn f(x: AD, y: AD) -> AD {

src/fuga/mod.rs

@@ -170,6 +170,7 @@ pub use crate::statistics::stat::QType::{
 pub use crate::structure::matrix::{
     Form::{Diagonal, Identity},
     SolveKind::{LU, WAZ},
+    UPLO::{Upper, Lower}
 };
 pub use crate::structure::dataframe::DType::*;
 pub use crate::structure::ad::AD::*;

src/lib.rs

@@ -7,9 +7,10 @@
 //! * Linear Algebra (with BLAS & LAPACK)
 //!     * [Matrix](structure/matrix/index.html) operations
 //!         * `+,-,*,/`
-//!         * LU, Determinant, Inverse
+//!         * LU Decomposition, Determinant, Inverse
 //!         * QR Decomposition (`O3` feature needed)
 //!         * Singular Value Decomposition (`O3` feature needed)
+//!         * Cholesky Decomposition (`O3` feature needed)
 //!         * Reduced Row Echelon Form
 //!     * [Vector](structure/vector/index.html) operations
 //!     * [Eigenvalue, Eigenvector](numerical/eigen/index.html) algorithms
@@ -80,7 +81,7 @@
 //! ### Cargo.toml
 //!
 //! * To use `peroxide`, you should edit `Cargo.toml`
-//! * Current document version is corresponding to `0.30.2`
+//! * Current document version is corresponding to `0.30.11`
 //!
 //! 1. Default
 //!     ```toml
@@ -159,7 +160,7 @@
 //! ## Useful tips for features
 //!
 //! * After `0.28.0`, `dataframe` feature is replaced by `nc` feature.
-//! * If you want to use `QR` or `SVD` then should use `O3` feature (there are no implementations for these decompositions in `default`)
+//! * If you want to use *QR* or *SVD* or *Cholesky Decomposition* then should use `O3` feature (there are no implementations for these decompositions in `default`)
 //! * If you want to write your numerical results, then use `nc` feature and `netcdf` format. (It is much more effective than `csv` and `json`.)
 //! * After `0.23.0`, there are two options - `fuga`, `prelude`. Choose proper option for your computations.
 //! * To plot, use `nc` feature to export data as netcdf format and use python to draw plot.

src/prelude/simpler.rs

@@ -27,13 +27,15 @@ pub trait SimplerLinearAlgebra {
     fn waz_diag(&self) -> Option<matrix::WAZD>;
     fn waz(&self) -> Option<matrix::WAZD>;
     fn qr(&self) -> matrix::QR;
+    fn cholesky(&self) -> Matrix;
     fn rref(&self) -> Matrix;
     fn det(&self) -> f64;
     fn block(&self) -> (Matrix, Matrix, Matrix, Matrix);
     fn inv(&self) -> Matrix;
     fn pseudo_inv(&self) -> Matrix;
     fn solve(&self, b: &Vec<f64>) -> Vec<f64>;
     fn solve_mat(&self, m: &Matrix) -> Matrix;
+    fn is_symmetric(&self) -> bool;
 }
 
 /// Simple Eigenpair
@@ -115,6 +117,14 @@ impl SimplerLinearAlgebra for Matrix {
     fn solve_mat(&self, m: &Matrix) -> Matrix {
         matrix::LinearAlgebra::solve_mat(self, m, matrix::SolveKind::LU)
     }
+
+    fn cholesky(&self) -> Matrix {
+        matrix::LinearAlgebra::cholesky(self, matrix::UPLO::Lower)
+    }
+
+    fn is_symmetric(&self) -> bool {
+        matrix::LinearAlgebra::is_symmetric(self)
+    }
 }
 
 /// Simple solve