r2rs-nmath 0.1.1

mod d;
mod p;
mod q;
mod r;

use strafe_type::{LogProbability64, Probability64, Rational64, Real64};

pub(crate) use self::{d::*, p::*, q::*, r::*};
use crate::traits::{Distribution, RNG};

/// # The Logistic Distribution
///
/// ## Description
///
/// Density, distribution function, quantile function and random generation for the logistic
/// distribution with parameters location and scale.
///
/// ## Arguments
///
/// * location, scale: location and scale parameters.
///
/// ## Details
///
/// If location or scale are omitted, they assume the default values of 0 and 1 respectively.
///
/// The Logistic distribution with location = m and scale = s has distribution function
///
/// $ F(x) = \frac{1}{1 + exp(-\frac{x-m}{s})} $
///
/// and density
///
/// $ f(x) = \frac{1}{s} exp(\frac{x-m}{s}) (1 + exp(\frac{x-m}{s}))^{-2} $.
///
/// It is a long-tailed distribution with mean m and variance $ \frac{\pi^2}{3} s^2 $.
///
/// ## Value
///
/// dlogis gives the density, plogis gives the distribution function, qlogis gives the quantile
/// function, and rlogis generates random deviates.
///
/// The length of the result is determined by n for rlogis, and is the maximum of the lengths of
/// the numerical arguments for the other functions.
///
/// The numerical arguments other than n are recycled to the length of the result. Only the first
/// elements of the logical arguments are used.
///
/// ## Density Plot
///
/// ```rust
/// # use r2rs_base::traits::StatisticalSlice;
/// # use r2rs_nmath::{distribution::LogisticBuilder, traits::Distribution};
/// # use strafe_plot::prelude::{IntoDrawingArea, Line, Plot, PlotOptions, SVGBackend, BLACK};
/// # use strafe_type::FloatConstraint;
/// let logis = LogisticBuilder::new().build();
/// let x = <[f64]>::sequence(-6.0, 6.0, 1000);
/// let y = x
///     .iter()
///     .map(|x| logis.density(x).unwrap())
///     .collect::<Vec<_>>();
///
/// let root = SVGBackend::new("density.svg", (1024, 768)).into_drawing_area();
/// Plot::new()
///     .with_options(PlotOptions {
///         x_axis_label: "x".to_string(),
///         y_axis_label: "density".to_string(),
///         ..Default::default()
///     })
///     .with_plottable(Line {
///         x,
///         y,
///         color: BLACK,
///         ..Default::default()
///     })
///     .plot(&root)
///     .unwrap();
/// # use std::fs::rename;
/// #     drop(root);
/// #     rename(
/// #             format!("density.svg"),
/// #             format!("src/distribution/logis/doctest_out/density.svg"),
/// #     )
/// #     .unwrap();
/// ```
#[cfg_attr(feature = "doc_outputs", cfg_attr(all(), doc = embed_doc_image::embed_image!("density", "src/distribution/logis/doctest_out/density.svg")))]
#[cfg_attr(feature = "doc_outputs", cfg_attr(all(), doc = "![Density][density]"))]
///
/// ## Note
///
/// qlogis(p) is the same as the well known ‘logit’ function, $ logit(p) = log(p/(1-p)) $, and
/// plogis(x) has consequently been called the ‘inverse logit’.
///
/// The distribution function is a rescaled hyperbolic tangent, $ plogis(x) == (1+ tanh(x/2))/2 $,
/// and it is called a sigmoid function in contexts such as neural networks.
///
/// ## Source
///
/// \[dpq\]logis are calculated directly from the definitions.
///
/// rlogis uses inversion.
///
/// # References
///
/// Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) The New S Language. Wadsworth &
/// Brooks/Cole.
///
/// Johnson, N. L., Kotz, S. and Balakrishnan, N. (1995) Continuous Univariate Distributions,
/// volume 2, chapter 23. Wiley, New York.
///
/// ## See Also
///
/// Distributions for other standard distributions.
///
/// ## Examples
///
/// Approximately equal (+/- 3)
/// ```rust
/// # use num_traits::FloatConst;
/// # use r2rs_nmath::{
/// #     distribution::LogisticBuilder,
/// #     rng::MersenneTwister,
/// #     traits::{Distribution, RNG},
/// # };
/// # use r2rs_stats::funcs::variance;
/// # use strafe_type::FloatConstraint;
/// let logis = LogisticBuilder::new()
///     .with_location(0)
///     .with_scale(5)
///     .build();
///
/// let mut rng = MersenneTwister::new();
/// rng.set_seed(1);
///
/// let r = (0..4000)
///     .map(|_| logis.random_sample(&mut rng).unwrap())
///     .collect::<Vec<_>>();
/// println!("{}", variance(&r));
/// println!("{}", f64::PI().powi(2) / 3.0 * 5.0_f64.powi(2));
/// # use std::{fs::File, io::Write};
/// # let mut f = File::create("src/distribution/logis/doctest_out/rand.md").unwrap();
/// # writeln!(f, "```output").unwrap();
/// # writeln!(f, "{}", variance(&r)).unwrap();
/// # writeln!(f, "{}", f64::PI().powi(2) / 3.0 * 5.0_f64.powi(2)).unwrap();
/// # writeln!(f, "```").unwrap();
/// ```
#[cfg_attr(feature = "doc_outputs", cfg_attr(all(), doc = include_str!("doctest_out/rand.md")))]
pub struct Logistic {
    location: Real64,
    scale: Rational64,
}

impl Distribution for Logistic {
    fn density<R: Into<Real64>>(&self, x: R) -> Real64 {
        dlogis(x, self.location, self.scale, false)
    }

    fn log_density<R: Into<Real64>>(&self, x: R) -> Real64 {
        dlogis(x, self.location, self.scale, true)
    }

    fn probability<R: Into<Real64>>(&self, q: R, lower_tail: bool) -> Probability64 {
        plogis(q, self.location, self.scale, lower_tail)
    }

    fn log_probability<R: Into<Real64>>(&self, q: R, lower_tail: bool) -> LogProbability64 {
        log_plogis(q, self.location, self.scale, lower_tail)
    }

    fn quantile<P: Into<Probability64>>(&self, p: P, lower_tail: bool) -> Real64 {
        qlogis(p, self.location, self.scale, lower_tail)
    }

    fn log_quantile<LP: Into<LogProbability64>>(&self, p: LP, lower_tail: bool) -> Real64 {
        log_qlogis(p, self.location, self.scale, lower_tail)
    }

    fn random_sample<R: RNG>(&self, rng: &mut R) -> Real64 {
        rlogis(self.location, self.scale, rng)
    }
}

pub struct LogisticBuilder {
    location: Option<Real64>,
    scale: Option<Rational64>,
}

impl LogisticBuilder {
    pub fn new() -> Self {
        Self {
            location: None,
            scale: None,
        }
    }

    pub fn with_location<R: Into<Real64>>(&mut self, location: R) -> &mut Self {
        self.location = Some(location.into());
        self
    }

    pub fn with_scale<R: Into<Rational64>>(&mut self, scale: R) -> &mut Self {
        self.scale = Some(scale.into());
        self
    }

    pub fn build(&self) -> Logistic {
        let location = self.location.unwrap_or(0.0.into());
        let scale = self.scale.unwrap_or(1.0.into());

        Logistic { location, scale }
    }
}

#[cfg(test)]
mod tests;

#[cfg(all(test, feature = "enable_proptest"))]
mod proptests;

#[cfg(all(test, feature = "enable_covtest"))]
mod covtests;