// Cargo.toml[dependencies]compute ="0.2"

[dependencies]compute ={ git ="https://github.com/al-jshen/compute"}

// example with BLAS onlycompute ={version ="0.2", features =["blas"]}

Examples

Statistical distributions

use compute::distributions::*;let beta =Beta::new(2.,2.);let betadata = b.sample_n(1000);// vector of 1000 variatesprintln!("{}", beta.mean());// analytic meanprintln!("{}", beta.var());// analytic varianceprintln!("{}", beta.pdf(0.5));// probability distribution functionlet binom =Binomial::new(4,0.5);println!("{}", p.sample());// sample single valueprintln!("{}", p.pmf(2));// probability mass function

Linear algebra

use compute::linalg::*;let x =arange(1.,4.,0.1).ln_1p().reshape(-1,3);// automatic shape detectionlet y =Vector::from([1.,2.,3.]);// vector structlet pd = x.t().dot(x);// transpose and matrix multiplylet jitter =Matrix::eye(3)*1e-6;// elementwise operationslet c =(pd + jitter).cholesky();// matrix decompositionslet s = c.solve(&y.exp());// linear solversprintln!("{}", s);

Linear models

use compute::prelude::*;let x =vec![0.50,0.75,1.00,1.25,1.50,1.75,1.75,2.00,2.25,2.50,2.75,3.00,3.25,3.50,4.00,4.25,4.50,4.75,5.00,5.50,];let y =vec![0.,0.,0.,0.,0.,0.,1.,0.,1.,0.,1.,0.,1.,0.,1.,1.,1.,1.,1.,1.,];let n = y.len();let xd =design(&x, n);letmut glm =GLM::new(ExponentialFamily::Bernoulli);// logistic regressionglm.set_penalty(1.);// L2 penaltyglm.fit(&xd,&y,25).unwrap();// with fit scoring algorithm (MLE)let coef = glm.coef().unwrap();// get estimated parameterslet errors = glm.coef_standard_error().unwrap();// get errors on parametersprintln!("{:?}", coef);println!("{:?}", errors);

Optimization

use compute::optimize::*;// define a function using a consistent optimization interfacefnrosenbrock<'a>(p:&[Var<'a>],d:&[&[f64]]) ->Var<'a>{assert_eq!(p.len(),2);assert_eq!(d.len(),1);assert_eq!(d[0].len(),2);let(x, y) =(p[0], p[1]);let(a, b) =(d[0][0], d[0][1]);(a - x).powi(2) + b*(y - x.powi(2)).powi(2)}// set up and run optimizerlet init =[0.,0.];let optim =Adam::with_stepsize(5e-4);let popt = optim.optimize(rosenbrock,&init,&[&[1.,100.]],10000);println!("{:?}", popt);

Time series models

use compute::timeseries::*;let x =vec![-2.584, -3.474, -1.977, -0.226,1.166,0.923, -1.075,0.732,0.959];letmut ar =AR::new(1);// AR(1) modelar.fit(&x);// fit model with Yule-Walker equationsprintln!("{:?}", ar.coeffs);// get model coefficientsprintln!("{:?}", ar.predict(&x,5));// forecast 5 steps ahead

Numerical integration

use compute::integrate::*;let f = |x:f64| x.sqrt() + x.sin() -(3.* x).cos() - x.powi(2);println!("{}", trapz(f,0.,1.,100));// trapezoid integration with 100 segmentsprintln!("{}", quad5(f,0.,1.));// gaussian quadrature integrationprintln!("{}", romberg(f,0.,1.,1e-8,10));// romberg integration with tolerance and max steps

Data summary functions

use compute::statistics::*;use compute::linalg::Vector;let x =Vector::from([2.2,3.4,5.,10., -2.1,0.1]);println!("{}", x.mean());println!("{}", x.var());println!("{}", x.max());println!("{}", x.argmax());

Mathematical and statistical functions

use compute::functions::*;println!("{}", logistic(4.));println!("{}", boxcox(5.,2.);// boxcox transformprintln!("{}", digamma(2.));println!("{}", binom_coeff(10,4));// n choose k