From 7ab81249bf95d350916f1b5a26135ef83da05003 Mon Sep 17 00:00:00 2001 From: Ozzie Gooen Date: Fri, 3 Apr 2020 20:41:37 +0100 Subject: [PATCH] Kde with auto kernel width --- __tests__/Bandwidth__Test.re | 4 ++-- package.json | 1 - src/Samples.re | 3 +-- src/symbolic/Jstat.re | 6 ++++-- src/utility/Bandwidth.re | 30 ++++++++++++++++++++++++++++++ src/utility/Science.re | 28 ---------------------------- 6 files changed, 37 insertions(+), 35 deletions(-) create mode 100644 src/utility/Bandwidth.re delete mode 100644 src/utility/Science.re diff --git a/__tests__/Bandwidth__Test.re b/__tests__/Bandwidth__Test.re index 013402fa..54cd111d 100644 --- a/__tests__/Bandwidth__Test.re +++ b/__tests__/Bandwidth__Test.re @@ -4,10 +4,10 @@ open Expect; describe("Bandwidth", () => { test("nrd0()", () => { let data = [|1., 4., 3., 2.|]; - expect(Science.nrd0(data)) |> toEqual(0.7635139420854616); + expect(Bandwidth.nrd0(data)) |> toEqual(0.7625801874014622); }); test("nrd()", () => { let data = [|1., 4., 3., 2.|]; - expect(Science.nrd(data)) |> toEqual(0.899249754011766); + expect(Bandwidth.nrd(data)) |> toEqual(0.8981499984950554); }); }); \ No newline at end of file diff --git a/package.json b/package.json index d483b362..dda61fae 100644 --- a/package.json +++ b/package.json @@ -60,7 +60,6 @@ "react-use": "^13.27.0", "reason-react": ">=0.7.0", "reschema": "1.3.0", - "science": "^1.9.3", "tailwindcss": "1.2.0" }, "alias": { diff --git a/src/Samples.re b/src/Samples.re index d7f3b5d2..360f8279 100644 --- a/src/Samples.re +++ b/src/Samples.re @@ -89,11 +89,10 @@ module T = { }; let kde = (~samples, ~outputXYPoints) => { - let width = Science.nrd0(samples); + let width = Bandwidth.nrd0(samples); let xyPointRange = E.A.Sorted.range(samples) |> E.O.default(0.0); let xyPointWidth = xyPointRange /. float_of_int(outputXYPoints); let kernelWidth = int_of_float(Jstat.max([|(width /. xyPointWidth), 1.0 |])); - Js.log4(samples, width, xyPointWidth, kernelWidth); KDE.normalSampling(samples, outputXYPoints, kernelWidth); }; diff --git a/src/symbolic/Jstat.re b/src/symbolic/Jstat.re index 4c361bd2..5f1c6c51 100644 --- a/src/symbolic/Jstat.re +++ b/src/symbolic/Jstat.re @@ -97,6 +97,8 @@ type binomial = { [@bs.module "jstat"] external deviation: array(float) => float = "deviation"; [@bs.module "jstat"] external stdev: array(float) => float = "stdev"; [@bs.module "jstat"] -external quantiles: (array(float), array(float)) => float = "quantiles"; +external quartiles: (array(float)) => array(float) = "quartiles"; [@bs.module "jstat"] -external percentile: (array(float), array(float)) => float = "percentile"; \ No newline at end of file +external quantiles: (array(float), array(float)) => array(float) = "quantiles"; +[@bs.module "jstat"] +external percentile: (array(float), float, bool) => float = "percentile"; \ No newline at end of file diff --git a/src/utility/Bandwidth.re b/src/utility/Bandwidth.re new file mode 100644 index 00000000..d2315639 --- /dev/null +++ b/src/utility/Bandwidth.re @@ -0,0 +1,30 @@ +//The math here was taken from https://github.com/jasondavies/science.js/blob/master/src/stats/bandwidth.js + +let len = x => E.A.length(x) |> float_of_int; + +let iqr = x => { + Jstat.percentile(x, 0.75, true) -. Jstat.percentile(x, 0.25, true); +}; + +// Silverman, B. W. (1986) Density Estimation. London: Chapman and Hall. +let nrd0 = x => { + let hi = Js_math.sqrt(Jstat.variance(x)); + let lo = Js_math.minMany_float([|hi, iqr(x) /. 1.34|]); + let e = Js_math.abs_float(x[1]); + let lo' = + switch (lo, hi, e) { + | (lo, _, _) when !Js.Float.isNaN(lo) => lo + | (_, hi, _) when !Js.Float.isNaN(hi) => hi + | (_, _, e) when !Js.Float.isNaN(e) => e + | _ => 1.0 + }; + 0.9 *. lo' *. Js.Math.pow_float(~base=len(x), ~exp=-0.2); +}; + +// Scott, D. W. (1992) Multivariate Density Estimation: Theory, Practice, and Visualization. Wiley. +let nrd = x => { + let h = iqr(x) /. 1.34; + 1.06 + *. Js.Math.min_float(Js.Math.sqrt(Jstat.variance(x)), h) + *. Js.Math.pow_float(~base=len(x), ~exp=(-1.0) /. 5.0); +}; \ No newline at end of file diff --git a/src/utility/Science.re b/src/utility/Science.re deleted file mode 100644 index b755ad5b..00000000 --- a/src/utility/Science.re +++ /dev/null @@ -1,28 +0,0 @@ -[@bs.val] [@bs.module "science"] [@bs.scope "stats"] -external variance: array(float) => float = "variance"; - -[@bs.val] [@bs.module "science"] [@bs.scope "stats"] -external iqr: array(float) => float = "iqr"; - -let len = x => E.A.length(x) |> float_of_int; - -let nrd0 = x => { - let hi = Js_math.sqrt(variance(x)); - let lo = Js_math.minMany_float([|hi, iqr(x) /. 1.34|]); - let e = Js_math.abs_float(x[1]); - let lo' = - switch (lo, hi, e) { - | (lo, hi, e) when !Js.Float.isNaN(lo) => lo - | (lo, hi, e) when !Js.Float.isNaN(hi) => hi - | (lo, hi, e) when !Js.Float.isNaN(e) => e - | _ => 1.0 - }; - 0.9 *. lo' *. Js_math.pow_float(len(x), -0.2); -}; - -let nrd = x => { - let h = iqr(x) /. 1.34; - 1.06 - *. Js_math.min_float(Js_math.sqrt(variance(x)), h) - *. Js_math.pow_float(len(x), (-1.0) /. 5.0); -}; \ No newline at end of file