ml-gsd - npm Package Compare versions

Comparing version 4.0.0 to 5.0.0

History.md

@@ -0,1 +1,5 @@
+# [5.0.0](https://github.com/mljs/global-spectral-deconvolution/compare/v4.0.0...v5.0.0) (2020-05-19)
+
+
+
 # [4.0.0](https://github.com/mljs/global-spectral-deconvolution/compare/v3.0.1...v4.0.0) (2020-03-20)
@@ -2,0 +6,0 @@

lib/index.js

@@ -8,18 +8,4 @@ 'use strict';
 var SG = _interopDefault(require('ml-savitzky-golay-generalized'));
-var optimize = _interopDefault(require('ml-optimize-lorentzian'));
+var mlOptimizeLorentzian = require('ml-optimize-lorentzian');
 
-const defaultOptions = {
-  sgOptions: {
-    windowSize: 9,
-    polynomial: 3,
-  },
-  minMaxRatio: 0.00025,
-  broadRatio: 0.0,
-  maxCriteria: true,
-  smoothY: true,
-  realTopDetection: false,
-  heightFactor: 0,
-  derivativeThreshold: -1,
-};
-
 /**
@@ -46,31 +32,28 @@ * Global spectra deconvolution
  */
-function gsd(x, yIn, options) {
-  options = Object.assign({}, defaultOptions, options);
-  let sgOptions = options.sgOptions;
+function gsd(x, yIn, options = {}) {
+  let {
+    noiseLevel,
+    sgOptions = {
+      windowSize: 9,
+      polynomial: 3,
+    },
+    smoothY = true,
+    heightFactor = 0,
+    broadRatio = 0.0,
+    maxCriteria = true,
+    minMaxRatio = 0.00025,
+    derivativeThreshold = -1,
+    realTopDetection = false,
+  } = options;
+
   const y = yIn.slice();
+  let equalSpaced = isEqualSpaced(x);
 
-  let maxDx = 0;
-  let minDx = Number.MAX_VALUE;
-  if (!('noiseLevel' in options)) {
-    // We have to know if x is equally spaced
+  if (noiseLevel === undefined) {
+    noiseLevel = equalSpaced ? getNoiseLevel(y) : 0;
+  }
 
-    let tmp;
-    for (let i = 0; i < x.length - 1; ++i) {
-      tmp = Math.abs(x[i + 1] - x[i]);
-      if (tmp < minDx) {
-        minDx = tmp;
-      }
-      if (tmp > maxDx) {
-        maxDx = tmp;
-      }
-    }
+  const yCorrection = { m: 1, b: noiseLevel };
 
-    if ((maxDx - minDx) / maxDx < 0.05) {
-      options.noiseLevel = getNoiseLevel(y);
-    } else {
-      options.noiseLevel = 0;
-    }
-  }
-  const yCorrection = { m: 1, b: options.noiseLevel };
-  if (!options.maxCriteria) {
+  if (!maxCriteria) {
     yCorrection.m = -1;
@@ -89,10 +72,13 @@ yCorrection.b *= -1;
   }
-  // If the max difference between delta x is less than 5%, then, we can assume it to be equally spaced variable
+  // If the max difference between delta x is less than 5%, then,
+  // we can assume it to be equally spaced variable
   let Y = y;
   let dY, ddY;
-  if ((maxDx - minDx) / maxDx < 0.05) {
-    if (options.smoothY) {
+  const { windowSize, polynomial } = sgOptions;
+
+  if (equalSpaced) {
+    if (smoothY) {
       Y = SG(y, x[1] - x[0], {
-        windowSize: sgOptions.windowSize,
-        polynomial: sgOptions.polynomial,
+        windowSize,
+        polynomial,
         derivative: 0,
@@ -102,16 +88,16 @@ });
     dY = SG(y, x[1] - x[0], {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 1,
     });
     ddY = SG(y, x[1] - x[0], {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 2,
     });
   } else {
-    if (options.smoothY) {
+    if (smoothY) {
       Y = SG(y, x, {
-        windowSize: sgOptions.windowSize,
-        polynomial: sgOptions.polynomial,
+        windowSize,
+        polynomial,
         derivative: 0,
@@ -121,13 +107,13 @@ });
     dY = SG(y, x, {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 1,
     });
     ddY = SG(y, x, {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 2,
     });
   }
-
+  // console.log('this is 2', y)
   const X = x;
@@ -159,3 +145,4 @@ const dx = x[1] - x[0];
     // filter based on derivativeThreshold
-    if (Math.abs(dY[i]) > options.derivativeThreshold) {
+    // console.log('pasa', y[i], dY[i], ddY[i]);
+    if (Math.abs(dY[i]) > derivativeThreshold) {
       // Minimum in first derivative
@@ -196,4 +183,3 @@ if (
       minddY[minddYLen++] = i; // ( [X[i], Y[i], i] );
-      broadMask[broadMaskLen++] =
-        Math.abs(ddY[i]) <= options.broadRatio * maxDdy;
+      broadMask[broadMaskLen++] = Math.abs(ddY[i]) <= broadRatio * maxDdy;
     }
@@ -234,3 +220,3 @@ }
     if (possible !== -1) {
-      if (Math.abs(Y[minddY[j]]) > options.minMaxRatio * maxY) {
+      if (Math.abs(Y[minddY[j]]) > minMaxRatio * maxY) {
         signals[signalsLen++] = {
@@ -247,8 +233,7 @@ index: minddY[j],
 
-        if (options.heightFactor) {
+        if (heightFactor) {
           let yLeft = Y[intervalL[possible].index];
           let yRight = Y[intervalR[possible].index];
           signals[signalsLen - 1].height =
-            options.heightFactor *
-            (signals[signalsLen - 1].y - (yLeft + yRight) / 2);
+            heightFactor * (signals[signalsLen - 1].y - (yLeft + yRight) / 2);
         }
@@ -260,4 +245,4 @@ }
 
-  if (options.realTopDetection) {
-    realTopDetection(signals, X, Y);
+  if (realTopDetection) {
+    determineRealTop(signals, X, Y);
   }
@@ -267,6 +252,6 @@
   for (let j = 0; j < signals.length; j++) {
-    signals[j].base = options.noiseLevel;
+    signals[j].base = noiseLevel;
   }
 
-  signals.sort(function(a, b) {
+  signals.sort(function (a, b) {
     return a.x - b.x;
@@ -278,3 +263,19 @@ });
 
-function getNoiseLevel(y) {
+const isEqualSpaced = (x) => {
+  let tmp;
+  let maxDx = 0;
+  let minDx = Number.MAX_SAFE_INTEGER;
+  for (let i = 0; i < x.length - 1; ++i) {
+    tmp = Math.abs(x[i + 1] - x[i]);
+    if (tmp < minDx) {
+      minDx = tmp;
+    }
+    if (tmp > maxDx) {
+      maxDx = tmp;
+    }
+  }
+  return (maxDx - minDx) / maxDx < 0.05;
+};
+
+const getNoiseLevel = (y) => {
   let mean = 0;
@@ -303,5 +304,5 @@
   return stddev;
-}
+};
 
-function realTopDetection(peakList, x, y) {
+const determineRealTop = (peakList, x, y) => {
   let alpha, beta, gamma, p, currentPoint;
@@ -360,57 +361,27 @@ for (let j = 0; j < peakList.length; j++) {
   }
-}
+};
 
-function sampleFunction(from, to, x, y, lastIndex) {
-  let nbPoints = x.length;
-  let sampleX = [];
-  let sampleY = [];
-  let direction = Math.sign(x[1] - x[0]); // Direction of the derivative
-  if (direction === -1) {
-    lastIndex[0] = x.length - 1;
-  }
+function optimizePeaks(peakList, x, y, options = {}) {
+  const {
+    functionName = 'gaussian',
+    factorWidth = 4,
+    optimizationOptions = {
+      damping: 1.5,
+      maxIterations: 100,
+      errorTolerance: 10e-5,
+    },
+  } = options;
 
-  let delta = Math.abs(to - from) / 2;
-  let mid = (from + to) / 2;
-  let stop = false;
-  let index = lastIndex[0];
-  while (!stop && index < nbPoints && index >= 0) {
-    if (Math.abs(x[index] - mid) <= delta) {
-      sampleX.push(x[index]);
-      sampleY.push(y[index]);
-      index += direction;
-    } else {
-      // It is outside the range.
-      if (Math.sign(mid - x[index]) === 1) {
-        // We'll reach the mid going in the current direction
-        index += direction;
-      } else {
-        // There is not more peaks in the current range
-        stop = true;
-      }
-    }
-    // console.log(sampleX);
-  }
-  lastIndex[0] = index;
-  return [sampleX, sampleY];
-}
-
-function optimizePeaks(peakList, x, y, n, fnType) {
-  let i;
-  let j;
-
   let lastIndex = [0];
-  let groups = groupPeaks(peakList, n);
+  let groups = groupPeaks(peakList, factorWidth);
   let result = [];
   let factor = 1;
-  if (fnType === 'gaussian') {
+  if (functionName === 'gaussian') {
     factor = 1.17741;
   } // From https://en.wikipedia.org/wiki/Gaussian_function#Properties
-  let sampling, error, opts;
-  for (i = 0; i < groups.length; i++) {
+  let sampling;
+  for (let i = 0; i < groups.length; i++) {
     let peaks = groups[i].group;
     if (peaks.length > 1) {
       // Multiple peaks
-      // console.log("Pending group of overlaped peaks "+peaks.length);
-      // console.log("here1");
-      // console.log(groups[i].limits);
       sampling = sampleFunction(
@@ -423,17 +394,16 @@ groups[i].limits[0] - groups[i].limits[1],
       );
-      // console.log(sampling);
       if (sampling[0].length > 5) {
-        error = peaks[0].width / 1000;
-        opts = [3, 100, error, error, error, error * 10, error * 10, 11, 9, 1];
-        // var gauss = Opt.optimizeSingleGaussian(sampling[0], sampling[1], opts, peaks);
         let optPeaks = [];
-        if (fnType === 'gaussian') {
-          optPeaks = optimize.optimizeGaussianSum(sampling, peaks, opts);
+        if (functionName === 'gaussian') {
+          optPeaks = mlOptimizeLorentzian.optimizeGaussianSum(sampling, peaks, optimizationOptions);
         } else {
-          if (fnType === 'lorentzian') {
-            optPeaks = optimize.optimizeLorentzianSum(sampling, peaks, opts);
+          if (functionName === 'lorentzian') {
+            optPeaks = mlOptimizeLorentzian.optimizeLorentzianSum(
+              sampling,
+              peaks,
+              optimizationOptions,
+            );
           }
         }
-        // console.log(optPeak);
-        for (j = 0; j < optPeaks.length; j++) {
+        for (let j = 0; j < optPeaks.length; j++) {
           result.push({
@@ -450,4 +420,4 @@ x: optPeaks[j][0][0],
       sampling = sampleFunction(
-        peaks.x - n * peaks.width,
-        peaks.x + n * peaks.width,
+        peaks.x - factorWidth * peaks.width,
+        peaks.x + factorWidth * peaks.width,
         x,
@@ -457,30 +427,25 @@ y,
       );
-      // console.log("here2");
-      // console.log(groups[i].limits);
+
       if (sampling[0].length > 5) {
-        error = peaks.width / 1000;
-        opts = [3, 100, error, error, error, error * 10, error * 10, 11, 9, 1];
-        // var gauss = Opt.optimizeSingleGaussian(sampling[0], sampling[1], opts, peaks);
-        // var gauss = Opt.optimizeSingleGaussian([sampling[0],sampling[1]], peaks, opts);
-        let optPeak = [];
-        if (fnType === 'gaussian') {
-          optPeak = optimize.optimizeSingleGaussian(
+        let fitResult = [];
+        if (functionName === 'gaussian') {
+          fitResult = mlOptimizeLorentzian.optimizeSingleGaussian(
             [sampling[0], sampling[1]],
             peaks,
-            opts,
+            optimizationOptions,
           );
         } else {
-          if (fnType === 'lorentzian') {
-            optPeak = optimize.optimizeSingleLorentzian(
+          if (functionName === 'lorentzian') {
+            fitResult = mlOptimizeLorentzian.optimizeSingleLorentzian(
               [sampling[0], sampling[1]],
               peaks,
-              opts,
+              optimizationOptions,
             );
           }
         }
-        // console.log(optPeak);
+        let { parameters } = fitResult;
         result.push({
-          x: optPeak[0][0],
-          y: optPeak[1][0],
-          width: optPeak[2][0] * factor,
+          x: parameters[0],
+          y: parameters[1],
+          width: parameters[2] * factor,
         }); // From https://en.wikipedia.org/wiki/Gaussian_function#Properties}
@@ -493,10 +458,42 @@ }
 
+function sampleFunction(from, to, x, y, lastIndex) {
+  let nbPoints = x.length;
+  let sampleX = [];
+  let sampleY = [];
+  let direction = Math.sign(x[1] - x[0]); // Direction of the derivative
+  if (direction === -1) {
+    lastIndex[0] = x.length - 1;
+  }
+
+  let delta = Math.abs(to - from) / 2;
+  let mid = (from + to) / 2;
+  let stop = false;
+  let index = lastIndex[0];
+  while (!stop && index < nbPoints && index >= 0) {
+    if (Math.abs(x[index] - mid) <= delta) {
+      sampleX.push(x[index]);
+      sampleY.push(y[index]);
+      index += direction;
+    } else {
+      // It is outside the range.
+      if (Math.sign(mid - x[index]) === 1) {
+        // We'll reach the mid going in the current direction
+        index += direction;
+      } else {
+        // There is not more peaks in the current range
+        stop = true;
+      }
+    }
+  }
+  lastIndex[0] = index;
+  return [sampleX, sampleY];
+}
+
 function groupPeaks(peakList, nL) {
   let group = [];
   let groups = [];
-  let i, j;
   let limits = [peakList[0].x, nL * peakList[0].width];
   let upperLimit, lowerLimit;
   // Merge forward
-  for (i = 0; i < peakList.length; i++) {
+  for (let i = 0; i < peakList.length; i++) {
     // If the 2 things overlaps
@@ -531,3 +528,3 @@ if (
   // Merge backward
-  for (i = groups.length - 2; i >= 0; i--) {
+  for (let i = groups.length - 2; i >= 0; i--) {
     // The groups overlaps
@@ -538,3 +535,3 @@ if (
     ) {
-      for (j = 0; j < groups[i + 1].group.length; j++) {
+      for (let j = 0; j < groups[i + 1].group.length; j++) {
         groups[i].group.push(groups[i + 1].group[j]);
@@ -550,3 +547,3 @@ }
       }
-      // console.log(limits);
+
       groups[i].limits = [
@@ -600,3 +597,3 @@ (upperLimit + lowerLimit) / 2,
       if (count > 2) {
-        let fitted = optimize.optimizeSingleLorentzian(candidates, {
+        let fitted = mlOptimizeLorentzian.optimizeSingleLorentzian(candidates, {
           x: broadLines[maxI].x,
@@ -608,6 +605,7 @@ y: max,
         });
+        let { parameters } = fitted;
         peakList.push({
-          x: fitted[0][0],
-          y: fitted[1][0],
-          width: fitted[2][0],
+          x: parameters[0],
+          y: parameters[1],
+          width: parameters[2],
           soft: false,
@@ -629,3 +627,3 @@ });
 
-  peakList.sort(function(a, b) {
+  peakList.sort(function (a, b) {
     return a.x - b.x;
@@ -632,0 +630,0 @@ });

package.json

 {
   "name": "ml-gsd",
-  "version": "4.0.0",
+  "version": "5.0.0",
   "description": "Global Spectra Deconvolution",
@@ -44,20 +44,20 @@ "main": "lib/index.js",
   "devDependencies": {
-    "@babel/plugin-transform-modules-commonjs": "^7.8.3",
+    "@babel/plugin-transform-modules-commonjs": "^7.9.6",
     "chemcalc": "^3.4.1",
     "cheminfo-tools": "^1.23.3",
-    "eslint": "^6.8.0",
-    "eslint-config-cheminfo": "^2.0.4",
-    "eslint-plugin-import": "^2.20.1",
-    "eslint-plugin-jest": "^23.8.2",
-    "eslint-plugin-prettier": "^3.1.2",
-    "jest": "^25.1.0",
+    "eslint": "^7.0.0",
+    "eslint-config-cheminfo": "^3.0.0",
+    "eslint-plugin-import": "^2.20.2",
+    "eslint-plugin-jest": "^23.13.1",
+    "eslint-plugin-prettier": "^3.1.3",
+    "jest": "^26.0.1",
     "ml-stat": "^1.3.3",
-    "prettier": "^1.19.1",
-    "rollup": "^2.1.0",
-    "spectrum-generator": "^4.0.0",
+    "prettier": "^2.0.5",
+    "rollup": "^2.10.4",
+    "spectrum-generator": "^4.0.2",
     "xy-parser": "^3.0.0"
   },
   "dependencies": {
-    "ml-optimize-lorentzian": "0.1.4",
-    "ml-savitzky-golay-generalized": "2.0.1"
+    "ml-optimize-lorentzian": "^0.2.0",
+    "ml-savitzky-golay-generalized": "2.0.2"
   },
@@ -64,0 +64,0 @@ "jest": {

src/__tests__/infraRed.js

@@ -5,5 +5,5 @@ import { readFileSync } from 'fs';
 
-describe('Global spectra deconvolution Infrared spectra', function() {
+describe('Global spectra deconvolution Infrared spectra', function () {
   // Test case obtained from Pag 443, Chap 8.
-  it('Should get the correct result', function() {
+  it('Should get the correct result', function () {
     let spectrum = JSON.parse(
@@ -10,0 +10,0 @@ readFileSync(`${__dirname}/data/infraRed.json`, 'utf-8'),

src/__tests__/massPeakPicking.js

@@ -35,4 +35,6 @@ import { gsd, optimizePeaks } from '..';
     });
-    result = optimizePeaks(result, x, y, 1, 'gaussian');
-
+    result = optimizePeaks(result, x, y, {
+      factorWidth: 1,
+      functionName: 'gaussian',
+    });
     expect(result[0].x).toBeCloseTo(69.938, 1);
@@ -39,0 +41,0 @@ expect(result[0].y).toBeCloseTo(max, 2);

src/__tests__/simple.js

@@ -29,20 +29,5 @@ import { gsd } from '..';
 
-    expect(peaks).toStrictEqual([
-      {
-        base: 1.1956287811760204,
-        index: 15,
-        soft: false,
-        width: 2,
-        x: 15,
-        y: 4.657142857142857,
-        left: {
-          index: 14,
-          x: 14,
-        },
-        right: {
-          index: 16,
-          x: 16,
-        },
-      },
-    ]);
+    expect(peaks[0].y).toBeCloseTo(4.657, 3);
+    expect(peaks[0].base).toBeCloseTo(1.1956, 3);
+    expect(peaks[0].x).toStrictEqual(15);
   });
@@ -87,3 +72,3 @@
       realTopDetection: true,
-      smoothY: true,
+      smoothY: false,
       sgOptions: {
@@ -94,3 +79,2 @@ windowSize: 5,
     });
-
     expect(peaks).toStrictEqual([
@@ -110,4 +94,4 @@ {
         width: 3,
-        x: 14.755485084898725,
-        y: 3.7914767697342637,
+        x: 14.5,
+        y: 4.006546067576939,
       },
@@ -114,0 +98,0 @@ ]);

src/gsd.js

 import SG from 'ml-savitzky-golay-generalized';
 
-const defaultOptions = {
-  sgOptions: {
-    windowSize: 9,
-    polynomial: 3,
-  },
-  minMaxRatio: 0.00025,
-  broadRatio: 0.0,
-  maxCriteria: true,
-  smoothY: true,
-  realTopDetection: false,
-  heightFactor: 0,
-  derivativeThreshold: -1,
-};
-
 /**
@@ -38,31 +24,28 @@ * Global spectra deconvolution
  */
-export function gsd(x, yIn, options) {
-  options = Object.assign({}, defaultOptions, options);
-  let sgOptions = options.sgOptions;
+export function gsd(x, yIn, options = {}) {
+  let {
+    noiseLevel,
+    sgOptions = {
+      windowSize: 9,
+      polynomial: 3,
+    },
+    smoothY = true,
+    heightFactor = 0,
+    broadRatio = 0.0,
+    maxCriteria = true,
+    minMaxRatio = 0.00025,
+    derivativeThreshold = -1,
+    realTopDetection = false,
+  } = options;
+
   const y = yIn.slice();
+  let equalSpaced = isEqualSpaced(x);
 
-  let maxDx = 0;
-  let minDx = Number.MAX_VALUE;
-  if (!('noiseLevel' in options)) {
-    // We have to know if x is equally spaced
+  if (noiseLevel === undefined) {
+    noiseLevel = equalSpaced ? getNoiseLevel(y) : 0;
+  }
 
-    let tmp;
-    for (let i = 0; i < x.length - 1; ++i) {
-      tmp = Math.abs(x[i + 1] - x[i]);
-      if (tmp < minDx) {
-        minDx = tmp;
-      }
-      if (tmp > maxDx) {
-        maxDx = tmp;
-      }
-    }
+  const yCorrection = { m: 1, b: noiseLevel };
 
-    if ((maxDx - minDx) / maxDx < 0.05) {
-      options.noiseLevel = getNoiseLevel(y);
-    } else {
-      options.noiseLevel = 0;
-    }
-  }
-  const yCorrection = { m: 1, b: options.noiseLevel };
-  if (!options.maxCriteria) {
+  if (!maxCriteria) {
     yCorrection.m = -1;
@@ -81,10 +64,13 @@ yCorrection.b *= -1;
   }
-  // If the max difference between delta x is less than 5%, then, we can assume it to be equally spaced variable
+  // If the max difference between delta x is less than 5%, then,
+  // we can assume it to be equally spaced variable
   let Y = y;
   let dY, ddY;
-  if ((maxDx - minDx) / maxDx < 0.05) {
-    if (options.smoothY) {
+  const { windowSize, polynomial } = sgOptions;
+
+  if (equalSpaced) {
+    if (smoothY) {
       Y = SG(y, x[1] - x[0], {
-        windowSize: sgOptions.windowSize,
-        polynomial: sgOptions.polynomial,
+        windowSize,
+        polynomial,
         derivative: 0,
@@ -94,16 +80,16 @@ });
     dY = SG(y, x[1] - x[0], {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 1,
     });
     ddY = SG(y, x[1] - x[0], {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 2,
     });
   } else {
-    if (options.smoothY) {
+    if (smoothY) {
       Y = SG(y, x, {
-        windowSize: sgOptions.windowSize,
-        polynomial: sgOptions.polynomial,
+        windowSize,
+        polynomial,
         derivative: 0,
@@ -113,13 +99,13 @@ });
     dY = SG(y, x, {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 1,
     });
     ddY = SG(y, x, {
-      windowSize: sgOptions.windowSize,
-      polynomial: sgOptions.polynomial,
+      windowSize,
+      polynomial,
       derivative: 2,
     });
   }
-
+  // console.log('this is 2', y)
   const X = x;
@@ -151,3 +137,4 @@ const dx = x[1] - x[0];
     // filter based on derivativeThreshold
-    if (Math.abs(dY[i]) > options.derivativeThreshold) {
+    // console.log('pasa', y[i], dY[i], ddY[i]);
+    if (Math.abs(dY[i]) > derivativeThreshold) {
       // Minimum in first derivative
@@ -188,4 +175,3 @@ if (
       minddY[minddYLen++] = i; // ( [X[i], Y[i], i] );
-      broadMask[broadMaskLen++] =
-        Math.abs(ddY[i]) <= options.broadRatio * maxDdy;
+      broadMask[broadMaskLen++] = Math.abs(ddY[i]) <= broadRatio * maxDdy;
     }
@@ -226,3 +212,3 @@ }
     if (possible !== -1) {
-      if (Math.abs(Y[minddY[j]]) > options.minMaxRatio * maxY) {
+      if (Math.abs(Y[minddY[j]]) > minMaxRatio * maxY) {
         signals[signalsLen++] = {
@@ -239,8 +225,7 @@ index: minddY[j],
 
-        if (options.heightFactor) {
+        if (heightFactor) {
           let yLeft = Y[intervalL[possible].index];
           let yRight = Y[intervalR[possible].index];
           signals[signalsLen - 1].height =
-            options.heightFactor *
-            (signals[signalsLen - 1].y - (yLeft + yRight) / 2);
+            heightFactor * (signals[signalsLen - 1].y - (yLeft + yRight) / 2);
         }
@@ -252,4 +237,4 @@ }
 
-  if (options.realTopDetection) {
-    realTopDetection(signals, X, Y);
+  if (realTopDetection) {
+    determineRealTop(signals, X, Y);
   }
@@ -259,6 +244,6 @@
   for (let j = 0; j < signals.length; j++) {
-    signals[j].base = options.noiseLevel;
+    signals[j].base = noiseLevel;
   }
 
-  signals.sort(function(a, b) {
+  signals.sort(function (a, b) {
     return a.x - b.x;
@@ -270,3 +255,19 @@ });
 
-function getNoiseLevel(y) {
+const isEqualSpaced = (x) => {
+  let tmp;
+  let maxDx = 0;
+  let minDx = Number.MAX_SAFE_INTEGER;
+  for (let i = 0; i < x.length - 1; ++i) {
+    tmp = Math.abs(x[i + 1] - x[i]);
+    if (tmp < minDx) {
+      minDx = tmp;
+    }
+    if (tmp > maxDx) {
+      maxDx = tmp;
+    }
+  }
+  return (maxDx - minDx) / maxDx < 0.05;
+};
+
+const getNoiseLevel = (y) => {
   let mean = 0;
@@ -295,5 +296,5 @@
   return stddev;
-}
+};
 
-function realTopDetection(peakList, x, y) {
+const determineRealTop = (peakList, x, y) => {
   let alpha, beta, gamma, p, currentPoint;
@@ -352,2 +353,2 @@ for (let j = 0; j < peakList.length; j++) {
   }
-}
+};

src/post/joinBroadPeaks.js

@@ -1,2 +0,2 @@
-import Opt from 'ml-optimize-lorentzian';
+import { optimizeSingleLorentzian } from 'ml-optimize-lorentzian';
 
@@ -40,3 +40,3 @@ /**
       if (count > 2) {
-        let fitted = Opt.optimizeSingleLorentzian(candidates, {
+        let fitted = optimizeSingleLorentzian(candidates, {
           x: broadLines[maxI].x,
@@ -48,6 +48,7 @@ y: max,
         });
+        let { parameters } = fitted;
         peakList.push({
-          x: fitted[0][0],
-          y: fitted[1][0],
-          width: fitted[2][0],
+          x: parameters[0],
+          y: parameters[1],
+          width: parameters[2],
           soft: false,
@@ -69,3 +70,3 @@ });
 
-  peakList.sort(function(a, b) {
+  peakList.sort(function (a, b) {
     return a.x - b.x;
@@ -72,0 +73,0 @@ });

src/post/optimizePeaks.js

@@ -1,56 +0,31 @@
-import optimize from 'ml-optimize-lorentzian';
+import {
+  optimizeGaussianSum,
+  optimizeLorentzianSum,
+  optimizeSingleGaussian,
+  optimizeSingleLorentzian,
+} from 'ml-optimize-lorentzian';
 
-function sampleFunction(from, to, x, y, lastIndex) {
-  let nbPoints = x.length;
-  let sampleX = [];
-  let sampleY = [];
-  let direction = Math.sign(x[1] - x[0]); // Direction of the derivative
-  if (direction === -1) {
-    lastIndex[0] = x.length - 1;
-  }
+export function optimizePeaks(peakList, x, y, options = {}) {
+  const {
+    functionName = 'gaussian',
+    factorWidth = 4,
+    optimizationOptions = {
+      damping: 1.5,
+      maxIterations: 100,
+      errorTolerance: 10e-5,
+    },
+  } = options;
 
-  let delta = Math.abs(to - from) / 2;
-  let mid = (from + to) / 2;
-  let stop = false;
-  let index = lastIndex[0];
-  while (!stop && index < nbPoints && index >= 0) {
-    if (Math.abs(x[index] - mid) <= delta) {
-      sampleX.push(x[index]);
-      sampleY.push(y[index]);
-      index += direction;
-    } else {
-      // It is outside the range.
-      if (Math.sign(mid - x[index]) === 1) {
-        // We'll reach the mid going in the current direction
-        index += direction;
-      } else {
-        // There is not more peaks in the current range
-        stop = true;
-      }
-    }
-    // console.log(sampleX);
-  }
-  lastIndex[0] = index;
-  return [sampleX, sampleY];
-}
-
-export function optimizePeaks(peakList, x, y, n, fnType) {
-  let i;
-  let j;
-
   let lastIndex = [0];
-  let groups = groupPeaks(peakList, n);
+  let groups = groupPeaks(peakList, factorWidth);
   let result = [];
   let factor = 1;
-  if (fnType === 'gaussian') {
+  if (functionName === 'gaussian') {
     factor = 1.17741;
   } // From https://en.wikipedia.org/wiki/Gaussian_function#Properties
-  let sampling, error, opts;
-  for (i = 0; i < groups.length; i++) {
+  let sampling;
+  for (let i = 0; i < groups.length; i++) {
     let peaks = groups[i].group;
     if (peaks.length > 1) {
       // Multiple peaks
-      // console.log("Pending group of overlaped peaks "+peaks.length);
-      // console.log("here1");
-      // console.log(groups[i].limits);
       sampling = sampleFunction(
@@ -63,17 +38,16 @@ groups[i].limits[0] - groups[i].limits[1],
       );
-      // console.log(sampling);
       if (sampling[0].length > 5) {
-        error = peaks[0].width / 1000;
-        opts = [3, 100, error, error, error, error * 10, error * 10, 11, 9, 1];
-        // var gauss = Opt.optimizeSingleGaussian(sampling[0], sampling[1], opts, peaks);
         let optPeaks = [];
-        if (fnType === 'gaussian') {
-          optPeaks = optimize.optimizeGaussianSum(sampling, peaks, opts);
+        if (functionName === 'gaussian') {
+          optPeaks = optimizeGaussianSum(sampling, peaks, optimizationOptions);
         } else {
-          if (fnType === 'lorentzian') {
-            optPeaks = optimize.optimizeLorentzianSum(sampling, peaks, opts);
+          if (functionName === 'lorentzian') {
+            optPeaks = optimizeLorentzianSum(
+              sampling,
+              peaks,
+              optimizationOptions,
+            );
           }
         }
-        // console.log(optPeak);
-        for (j = 0; j < optPeaks.length; j++) {
+        for (let j = 0; j < optPeaks.length; j++) {
           result.push({
@@ -90,4 +64,4 @@ x: optPeaks[j][0][0],
       sampling = sampleFunction(
-        peaks.x - n * peaks.width,
-        peaks.x + n * peaks.width,
+        peaks.x - factorWidth * peaks.width,
+        peaks.x + factorWidth * peaks.width,
         x,
@@ -97,30 +71,25 @@ y,
       );
-      // console.log("here2");
-      // console.log(groups[i].limits);
+
       if (sampling[0].length > 5) {
-        error = peaks.width / 1000;
-        opts = [3, 100, error, error, error, error * 10, error * 10, 11, 9, 1];
-        // var gauss = Opt.optimizeSingleGaussian(sampling[0], sampling[1], opts, peaks);
-        // var gauss = Opt.optimizeSingleGaussian([sampling[0],sampling[1]], peaks, opts);
-        let optPeak = [];
-        if (fnType === 'gaussian') {
-          optPeak = optimize.optimizeSingleGaussian(
+        let fitResult = [];
+        if (functionName === 'gaussian') {
+          fitResult = optimizeSingleGaussian(
             [sampling[0], sampling[1]],
             peaks,
-            opts,
+            optimizationOptions,
           );
         } else {
-          if (fnType === 'lorentzian') {
-            optPeak = optimize.optimizeSingleLorentzian(
+          if (functionName === 'lorentzian') {
+            fitResult = optimizeSingleLorentzian(
               [sampling[0], sampling[1]],
               peaks,
-              opts,
+              optimizationOptions,
             );
           }
         }
-        // console.log(optPeak);
+        let { parameters } = fitResult;
         result.push({
-          x: optPeak[0][0],
-          y: optPeak[1][0],
-          width: optPeak[2][0] * factor,
+          x: parameters[0],
+          y: parameters[1],
+          width: parameters[2] * factor,
         }); // From https://en.wikipedia.org/wiki/Gaussian_function#Properties}
@@ -133,10 +102,42 @@ }
 
+function sampleFunction(from, to, x, y, lastIndex) {
+  let nbPoints = x.length;
+  let sampleX = [];
+  let sampleY = [];
+  let direction = Math.sign(x[1] - x[0]); // Direction of the derivative
+  if (direction === -1) {
+    lastIndex[0] = x.length - 1;
+  }
+
+  let delta = Math.abs(to - from) / 2;
+  let mid = (from + to) / 2;
+  let stop = false;
+  let index = lastIndex[0];
+  while (!stop && index < nbPoints && index >= 0) {
+    if (Math.abs(x[index] - mid) <= delta) {
+      sampleX.push(x[index]);
+      sampleY.push(y[index]);
+      index += direction;
+    } else {
+      // It is outside the range.
+      if (Math.sign(mid - x[index]) === 1) {
+        // We'll reach the mid going in the current direction
+        index += direction;
+      } else {
+        // There is not more peaks in the current range
+        stop = true;
+      }
+    }
+  }
+  lastIndex[0] = index;
+  return [sampleX, sampleY];
+}
+
 function groupPeaks(peakList, nL) {
   let group = [];
   let groups = [];
-  let i, j;
   let limits = [peakList[0].x, nL * peakList[0].width];
   let upperLimit, lowerLimit;
   // Merge forward
-  for (i = 0; i < peakList.length; i++) {
+  for (let i = 0; i < peakList.length; i++) {
     // If the 2 things overlaps
@@ -171,3 +172,3 @@ if (
   // Merge backward
-  for (i = groups.length - 2; i >= 0; i--) {
+  for (let i = groups.length - 2; i >= 0; i--) {
     // The groups overlaps
@@ -178,3 +179,3 @@ if (
     ) {
-      for (j = 0; j < groups[i + 1].group.length; j++) {
+      for (let j = 0; j < groups[i + 1].group.length; j++) {
         groups[i].group.push(groups[i + 1].group[j]);
@@ -190,3 +191,3 @@ }
       }
-      // console.log(limits);
+
       groups[i].limits = [
@@ -193,0 +194,0 @@ (upperLimit + lowerLimit) / 2,

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Worsened metrics

Total package byte prevSize

5855830

decreased by-0.02%

Lines of code

1720

decreased by-0.98%

Dependency changes

• + Addedis-any-array@0.1.12.0.1(transitive)

• + Addedml-array-max@1.2.4(transitive)

• + Addedml-array-min@1.2.3(transitive)

• + Addedml-array-rescale@1.3.7(transitive)

• + Addedml-levenberg-marquardt@2.1.1(transitive)

• + Addedml-matrix@6.12.0(transitive)

• + Addedml-optimize-lorentzian@0.2.0(transitive)

• + Addedml-savitzky-golay-generalized@2.0.2(transitive)

• - Removedml-array-utils@0.3.0(transitive)

• - Removedml-curve-fitting@0.0.7(transitive)

• - Removedml-matrix@2.3.0(transitive)

• - Removedml-optimize-lorentzian@0.1.4(transitive)

• - Removedml-savitzky-golay-generalized@2.0.1(transitive)

• - Removedml-stat@1.3.3(transitive)

• Updatedml-optimize-lorentzian@^0.2.0

• Updatedml-savitzky-golay-generalized@2.0.2