normal_distribution 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 662e907768609223b84adda82f20f9dbba538b2b
+  data.tar.gz: d8b1598c480018ae67999e987be9d18d65340e00
+SHA512:
+  metadata.gz: c99cc756d443cc770d661f9c841be2c33c5526625ac53c26c959bcb725b6908ddfc8ef522221810db9dda39641cf4653ff49ce9de0c9d59e0e66e2468de6b6d6
+  data.tar.gz: 258308f35c9436adda2e0023c2f0dcb540014b2e0ffd4f482e8f4e12adfc5a5855f9bb0b3d6216865757bcea36c060435593fa6b068ef44323a5f780fd63ffb9

data/ext/normal_distribution/erf_inv.c

@@ -0,0 +1,91 @@
+#include "erf_inv.h"
+
+long double t_erf_inv(long double x) {
+	if (x < -1 || x > 1) {
+		return NAN;
+	} else if (x == 1.0) {
+		return INFINITY;
+	} else if (x == -1.0) {
+		return -INFINITY;
+	}
+
+	const long double LN2 = 6.931471805599453094172321214581e-1L;
+
+	const long double A0 = 1.1975323115670912564578e0L;
+	const long double A1 = 4.7072688112383978012285e1L;
+	const long double A2 = 6.9706266534389598238465e2L;
+	const long double A3 = 4.8548868893843886794648e3L;
+	const long double A4 = 1.6235862515167575384252e4L;
+	const long double A5 = 2.3782041382114385731252e4L;
+	const long double A6 = 1.1819493347062294404278e4L;
+	const long double A7 = 8.8709406962545514830200e2L;
+
+	const long double B0 = 1.0000000000000000000e0L;
+	const long double B1 = 4.2313330701600911252e1L;
+	const long double B2 = 6.8718700749205790830e2L;
+	const long double B3 = 5.3941960214247511077e3L;
+	const long double B4 = 2.1213794301586595867e4L;
+	const long double B5 = 3.9307895800092710610e4L;
+	const long double B6 = 2.8729085735721942674e4L;
+	const long double B7 = 5.2264952788528545610e3L;
+
+	const long double C0 = 1.42343711074968357734e0L;
+	const long double C1 = 4.63033784615654529590e0L;
+	const long double C2 = 5.76949722146069140550e0L;
+	const long double C3 = 3.64784832476320460504e0L;
+	const long double C4 = 1.27045825245236838258e0L;
+	const long double C5 = 2.41780725177450611770e-1L;
+	const long double C6 = 2.27238449892691845833e-2L;
+	const long double C7 = 7.74545014278341407640e-4L;
+
+	const long double D0 = 1.4142135623730950488016887e0L;
+	const long double D1 = 2.9036514445419946173133295e0L;
+	const long double D2 = 2.3707661626024532365971225e0L;
+	const long double D3 = 9.7547832001787427186894837e-1L;
+	const long double D4 = 2.0945065210512749128288442e-1L;
+	const long double D5 = 2.1494160384252876777097297e-2L;
+	const long double D6 = 7.7441459065157709165577218e-4L;
+	const long double D7 = 1.4859850019840355905497876e-9L;
+
+	const long double E0 = 6.65790464350110377720e0L;
+	const long double E1 = 5.46378491116411436990e0L;
+	const long double E2 = 1.78482653991729133580e0L;
+	const long double E3 = 2.96560571828504891230e-1L;
+	const long double E4 = 2.65321895265761230930e-2L;
+	const long double E5 = 1.24266094738807843860e-3L;
+	const long double E6 = 2.71155556874348757815e-5L;
+	const long double E7 = 2.01033439929228813265e-7L;
+
+	const long double F0 = 1.414213562373095048801689e0L;
+	const long double F1 = 8.482908416595164588112026e-1L;
+	const long double F2 = 1.936480946950659106176712e-1L;
+	const long double F3 = 2.103693768272068968719679e-2L;
+	const long double F4 = 1.112800997078859844711555e-3L;
+	const long double F5 = 2.611088405080593625138020e-5L;
+	const long double F6 = 2.010321207683943062279931e-7L;
+	const long double F7 = 2.891024605872965461538222e-15L;
+
+	long double abs_x = fabsl(x);
+
+	if (abs_x <= 0.85L) {
+		long double r = 0.180625L - 0.25L * x * x;
+		long double num = (((((((A7 * r + A6) * r + A5) * r + A4) * r + A3) * r + A2) * r + A1) * r + A0);
+		long double den = (((((((B7 * r + B6) * r + B5) * r + B4) * r + B3) * r + B2) * r + B1) * r + B0);
+		return x * num / den;
+	}
+
+	long double r = sqrtl(LN2 - logl(1.0L - abs_x));
+
+	long double num, den;
+	if (r <= 5.0L) {
+		r = r - 1.6L;
+		num = (((((((C7 * r + C6) * r + C5) * r + C4) * r + C3) * r + C2) * r + C1) * r + C0);
+		den = (((((((D7 * r + D6) * r + D5) * r + D4) * r + D3) * r + D2) * r + D1) * r + D0);
+	} else {
+		r = r - 5.0L;
+		num = (((((((E7 * r + E6) * r + E5) * r + E4) * r + E3) * r + E2) * r + E1) * r + E0);
+		den = (((((((F7 * r + F6) * r + F5) * r + F4) * r + F3) * r + F2) * r + F1) * r + F0);
+	}
+
+	return copysignl(num / den, x);
+}

data/ext/normal_distribution/erf_inv.h

@@ -0,0 +1,12 @@
+///
+/// https://github.com/lakshayg/erfinv
+///
+
+#ifndef NORMAL_DISTRIBUTION_ERF_INV_H
+#define NORMAL_DISTRIBUTION_ERF_INV_H
+
+#include <math.h>
+
+long double t_erf_inv(long double x);
+
+#endif //NORMAL_DISTRIBUTION_ERF_INV_H

data/ext/normal_distribution/extconf.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+require "mkmf"
+
+create_makefile("normal_distribution/normal_distribution")

data/ext/normal_distribution/normal_distribution.c

@@ -0,0 +1,108 @@
+#include "normal_distribution.h"
+
+static double t_parse_percentage(VALUE percentage) {
+    double perc = NUM2DBL( percentage );
+
+    if (perc > 100 || perc < 0) {
+        rb_raise(rb_eArgError, "percentage must be between 0 and 100");
+    }
+
+    return perc;
+}
+
+static double t_mean( double * data, long size ) {
+	double sum = .0;
+
+	for ( long i = 0 ; i < size ; ++ i ) {
+		sum += data[i];
+	}
+
+	return sum / size;
+}
+
+static double t_z_score( double percentage ) {
+	return sqrt( 2 ) * t_erf_inv( percentage / 100 );
+}
+
+static double t_variance( double * data, long size, double mean ) {
+	double * squared_diff = ALLOC_N( double, size );
+
+	for ( long i = 0 ; i < size ; ++ i ) {
+		squared_diff[i] = pow( mean - data[i], 2 );
+	}
+
+	double variance = t_mean( squared_diff, size );
+	free( squared_diff );
+
+	return variance;
+}
+
+static double t_stddev( double * data, long size, double mean ) {
+	return sqrt( t_variance( data, size, mean ) );
+}
+
+static double * t_parse_dbl_ary( VALUE ary, long * size ) {
+    Check_Type(ary, T_ARRAY);
+	long len = RARRAY_LEN( ary );
+
+	if (len == 0) {
+	    rb_raise(rb_eArgError, "data must not be empty");
+	}
+
+	VALUE * values = RARRAY_PTR( ary );
+	double * d_data = ALLOC_N( double, len );
+
+	for ( int i = 0 ; i < len ; ++ i ) {
+		d_data[i] = NUM2DBL( values[i] );
+	}
+
+	*size = len;
+
+	return d_data;
+}
+
+static VALUE t_init( VALUE self, VALUE values ) {
+	long size;
+	double * data = t_parse_dbl_ary( values, &size );
+	double mean = t_mean( data, size );
+	double stddev = t_stddev( data, size, mean );
+
+	rb_iv_set( self, "@mean", rb_float_new( mean ) );
+	rb_iv_set( self, "@standard_deviation", rb_float_new( stddev ) );
+	free( data );
+
+	return self;
+}
+
+static VALUE t_confidence_interval( VALUE self, VALUE percentage ) {
+    double perc = t_parse_percentage( percentage );
+	double z = t_z_score( perc );
+	double stddev = NUM2DBL( rb_iv_get( self, "@standard_deviation" ) );
+	double mean = NUM2DBL( rb_iv_get( self, "@mean" ) );
+	double lower_bound = - z * stddev + mean;
+	double upper_bound = z * stddev + mean;
+
+	VALUE pair = rb_ary_new();
+	rb_ary_push( pair, rb_float_new( lower_bound ) );
+	rb_ary_push( pair, rb_float_new( upper_bound ) );
+
+	return pair;
+}
+
+static VALUE t_attr_mean( VALUE self ) {
+	return rb_iv_get( self, "@mean" );
+}
+
+static VALUE t_attr_stddev( VALUE self ) {
+	return rb_iv_get( self, "@standard_deviation" );
+}
+
+void Init_normal_distribution( void ) {
+	VALUE module = rb_define_module( "NormalDistribution" );
+	VALUE model = rb_define_class_under( module, "Model", rb_cObject );
+
+	rb_define_method( model, "initialize", t_init, 1 );
+	rb_define_method( model, "confidence_interval", t_confidence_interval, 1 );
+	rb_define_method( model, "mean", t_attr_mean, 0 );
+	rb_define_method( model, "standard_deviation", t_attr_stddev, 0 );
+}

data/ext/normal_distribution/normal_distribution.h

@@ -0,0 +1,10 @@
+#ifndef NORMAL_DISTRIBUTION_H
+#define NORMAL_DISTRIBUTION_H 1
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <math.h>
+#include "ruby.h"
+#include "erf_inv.h"
+
+#endif /* NORMAL_DISTRIBUTION_H */

data/lib/normal_distribution.rb

@@ -0,0 +1,7 @@
+# frozen_string_literal: true
+
+require "normal_distribution/version"
+require "normal_distribution/normal_distribution"
+
+module NormalDistribution
+end

data/lib/normal_distribution/model.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+module NormalDistribution
+  class Model
+    # :nodoc:
+    # attr_reader :mean
+
+    # :nodoc:
+    # attr_reader :standard_deviation
+
+    # Initializes normal distribution from given data
+    # Params:
+    # +values+:: Non-empty array of numbers representing data for normal distribution construction
+    # def initialize(values); end
+
+    # Creates confidence interval for given percentage
+    # Params:
+    # +percentage+:: Number in interval <0, 100>
+    # def confidence_interval(percentage); end
+  end
+end

data/lib/normal_distribution/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+module NormalDistribution
+  VERSION = "0.1.0"
+end

metadata

@@ -0,0 +1,125 @@
+--- !ruby/object:Gem::Specification
+name: normal_distribution
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Dominik Čermák
+autorequire:
+bindir: exe
+cert_chain: []
+date: 2020-12-16 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.16'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.16'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+- !ruby/object:Gem::Dependency
+  name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+- !ruby/object:Gem::Dependency
+  name: rspec-benchmark
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description:
+email:
+- domcermak64@gmail.com
+executables: []
+extensions:
+- ext/normal_distribution/extconf.rb
+extra_rdoc_files: []
+files:
+- ext/normal_distribution/erf_inv.c
+- ext/normal_distribution/erf_inv.h
+- ext/normal_distribution/extconf.rb
+- ext/normal_distribution/normal_distribution.c
+- ext/normal_distribution/normal_distribution.h
+- lib/normal_distribution.rb
+- lib/normal_distribution/model.rb
+- lib/normal_distribution/version.rb
+homepage:
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+- ext
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project:
+rubygems_version: 2.6.14
+signing_key:
+specification_version: 4
+summary: This gem provides library to detect anomalies in numerical data via normal
+  distribution
+test_files: []