renamed package

.gitignore

 node_modules
-_data
 .rpt2_cache
\ No newline at end of file

.npmignore

+node_modules
+.rpt2_cache
+
+examples
+test
+proto
+weights
\ No newline at end of file

README.md

-# face-recognition.min.js
+# face-api.js
 
-**face recognition API for the browser with tensorflow.js**
+**JavaScript API for face detection and face recognition in the browser with tensorflow.js**
 
 This project implements a ResNet-34 like architecture using the tensorflow.js core API ([@tensorflow/tfjs-core](https://github.com/tensorflow/tfjs-core)) for realtime face recognition in the browser. The neural net is equivalent to the **FaceRecognizerNet** used in [face-recognition.js](https://github.com/justadudewhohacks/face-recognition.js) and the net used in the [dlib](https://github.com/davisking/dlib/blob/master/examples/dnn_face_recognition_ex.cpp) face recognition example. The weights have been trained by [davisking](https://github.com/davisking) and the model achieves a prediction accuracy of 99.38% on the LFW (Labeled Faces in the Wild) benchmark for face recognition.
 
@@ -29,7 +29,7 @@ Download the weights file from your server and initialize the net (note, that yo
 ``` javascript
 const res = await axios.get('face_recognition_model.weights', { responseType: 'arraybuffer' })
 const weights = new Float32Array(res.data)
-const net = facerecognition.faceRecognitionNet(weights)
+const net = faceapi.faceRecognitionNet(weights)
 ```
 
 Compute and compare two 150 x 150 sized face images:
@@ -43,7 +43,7 @@ const imgData2 = ...
 
 const descriptor1 = await net.computeFaceDescriptor(imgData1)
 const descriptor2 = await net.computeFaceDescriptor(imgData2)
-const distance = facerecognition.euclidianDistance(descriptor1, descriptor2)
+const distance = faceapi.euclidianDistance(descriptor1, descriptor2)
 
 if (distance < 0.6)
   console.log('match')

dist/face-recognition.js → dist/face-api.js

 (function (global, factory) {
     typeof exports === 'object' && typeof module !== 'undefined' ? factory(exports, require('crypto')) :
     typeof define === 'function' && define.amd ? define(['exports', 'crypto'], factory) :
-    (factory((global.facerecognition = global.facerecognition || {}),global.crypto));
+    (factory((global.faceapi = global.faceapi || {}),global.crypto));
 }(this, (function (exports,crypto) { 'use strict';
 
     crypto = crypto && crypto.hasOwnProperty('default') ? crypto['default'] : crypto;
@@ -1650,11 +1650,11 @@
     seedrandom.xor4096 = xor4096;
     seedrandom.tychei = tychei;
 
-    var C__Users_user_dev_faceRecognition_min_js_node_modules_seedrandom = seedrandom;
+    var C__Users_user_dev_faceApi_js_node_modules_seedrandom = seedrandom;
 
     var seedrandom$1 = /*#__PURE__*/Object.freeze({
-        default: C__Users_user_dev_faceRecognition_min_js_node_modules_seedrandom,
-        __moduleExports: C__Users_user_dev_faceRecognition_min_js_node_modules_seedrandom
+        default: C__Users_user_dev_faceApi_js_node_modules_seedrandom,
+        __moduleExports: C__Users_user_dev_faceApi_js_node_modules_seedrandom
     });
 
     var MPRandGauss = (function () {

dist/transformInputs.d.ts

-import * as tf from '@tensorflow/tfjs-core';
-import { NetInput } from './NetInput';
-import { TNetInput } from './types';
-/**
- * Pads the smaller dimension of an image tensor with zeros, such that width === height.
- *
- * @param imgTensor The image tensor.
- * @param isCenterImage (optional, default: false) If true, add padding on both sides of the image, such that the image
- * @returns The padded tensor with width === height.
- */
-export declare function padToSquare(imgTensor: tf.Tensor4D, isCenterImage?: boolean): tf.Tensor4D;
-export declare function getImageTensor(input: tf.Tensor | NetInput | TNetInput): tf.Tensor4D;

dist/transformInputs.js

-import * as tf from '@tensorflow/tfjs-core';
-import { NetInput } from './NetInput';
-/**
- * Pads the smaller dimension of an image tensor with zeros, such that width === height.
- *
- * @param imgTensor The image tensor.
- * @param isCenterImage (optional, default: false) If true, add padding on both sides of the image, such that the image
- * @returns The padded tensor with width === height.
- */
-export function padToSquare(imgTensor, isCenterImage) {
-    if (isCenterImage === void 0) { isCenterImage = false; }
-    return tf.tidy(function () {
-        var _a = imgTensor.shape.slice(1), height = _a[0], width = _a[1];
-        if (height === width) {
-            return imgTensor;
-        }
-        var paddingAmount = Math.floor(Math.abs(height - width) * (isCenterImage ? 0.5 : 1));
-        var paddingAxis = height > width ? 2 : 1;
-        var paddingTensorShape = imgTensor.shape.slice();
-        paddingTensorShape[paddingAxis] = paddingAmount;
-        var tensorsToStack = (isCenterImage ? [tf.fill(paddingTensorShape, 0)] : [])
-            .concat([imgTensor, tf.fill(paddingTensorShape, 0)]);
-        console.log(tensorsToStack);
-        return tf.concat(tensorsToStack, paddingAxis);
-    });
-}
-export function getImageTensor(input) {
-    return tf.tidy(function () {
-        if (input instanceof tf.Tensor) {
-            var rank = input.shape.length;
-            if (rank !== 3 && rank !== 4) {
-                throw new Error('input tensor must be of rank 3 or 4');
-            }
-            return (rank === 3 ? input.expandDims(0) : input).toFloat();
-        }
-        var netInput = input instanceof NetInput ? input : new NetInput(input);
-        return tf.concat(netInput.canvases.map(function (canvas) {
-            return tf.fromPixels(canvas).expandDims(0).toFloat();
-        }));
-    });
-}
-//# sourceMappingURL=transformInputs.js.map
\ No newline at end of file

dist/transformInputs.js.map

-{"version":3,"file":"transformInputs.js","sourceRoot":"","sources":["../src/transformInputs.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,MAAM,uBAAuB,CAAC;AAE5C,OAAO,EAAE,QAAQ,EAAE,MAAM,YAAY,CAAC;AAGtC;;;;;;GAMG;AACH,MAAM,sBACJ,SAAsB,EACtB,aAA8B;IAA9B,8BAAA,EAAA,qBAA8B;IAE9B,OAAO,EAAE,CAAC,IAAI,CAAC;QAEP,IAAA,6BAA0C,EAAzC,cAAM,EAAE,aAAK,CAA4B;QAChD,IAAI,MAAM,KAAK,KAAK,EAAE;YACpB,OAAO,SAAS,CAAA;SACjB;QAED,IAAM,aAAa,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,MAAM,GAAG,KAAK,CAAC,GAAG,CAAC,aAAa,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QACtF,IAAM,WAAW,GAAG,MAAM,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QAC1C,IAAM,kBAAkB,GAAG,SAAS,CAAC,KAAK,CAAC,KAAK,EAAsC,CAAA;QACtF,kBAAkB,CAAC,WAAW,CAAC,GAAG,aAAa,CAAA;QAC/C,IAAM,cAAc,GAAG,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,kBAAkB,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC;aAC3E,MAAM,CAAC,CAAC,SAAS,EAAG,EAAE,CAAC,IAAI,CAAC,kBAAkB,EAAE,CAAC,CAAC,CAAC,CAAkB,CAAA;QACxE,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC,CAAA;QAC3B,OAAO,EAAE,CAAC,MAAM,CAAC,cAAc,EAAE,WAAW,CAAC,CAAA;IAC/C,CAAC,CAAC,CAAA;AACJ,CAAC;AAED,MAAM,yBAAyB,KAAuC;IACpE,OAAO,EAAE,CAAC,IAAI,CAAC;QACb,IAAI,KAAK,YAAY,EAAE,CAAC,MAAM,EAAE;YAC9B,IAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,CAAA;YAC/B,IAAI,IAAI,KAAK,CAAC,IAAI,IAAI,KAAK,CAAC,EAAE;gBAC5B,MAAM,IAAI,KAAK,CAAC,qCAAqC,CAAC,CAAA;aACvD;YAED,OAAO,CAAC,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,OAAO,EAAiB,CAAA;SAC3E;QAED,IAAM,QAAQ,GAAG,KAAK,YAAY,QAAQ,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,QAAQ,CAAC,KAAK,CAAC,CAAA;QACxE,OAAO,EAAE,CAAC,MAAM,CACd,QAAQ,CAAC,QAAQ,CAAC,GAAG,CAAC,UAAA,MAAM;YAC1B,OAAA,EAAE,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE;QAA7C,CAA6C,CAC9C,CACa,CAAA;IAClB,CAAC,CAAC,CAAA;AACJ,CAAC"}
\ No newline at end of file

examples/public/commons.js

@@ -15,13 +15,13 @@ async function fetchImage(uri) {
 async function initFaceDetectionNet() {
   const res = await axios.get('face_detection_model.weights', { responseType: 'arraybuffer' })
   const weights = new Float32Array(res.data)
-  return facerecognition.faceDetectionNet(weights)
+  return faceapi.faceDetectionNet(weights)
 }
 
 async function initFaceRecognitionNet() {
   const res = await axios.get('face_recognition_model.weights', { responseType: 'arraybuffer' })
   const weights = new Float32Array(res.data)
-  return facerecognition.faceRecognitionNet(weights)
+  return faceapi.faceRecognitionNet(weights)
 }
 
 // fetch first image of each class and compute their descriptors
@@ -32,7 +32,7 @@ async function initTrainDescriptorsByClass(net, numImagesForTraining = 1) {
     async className => {
       const descriptors = []
       for (let i = 1; i < (numImagesForTraining + 1); i++) {
-        const img = await facerecognition.bufferToImage(
+        const img = await faceapi.bufferToImage(
           await fetchImage(getFaceImageUri(className, i))
         )
         descriptors.push(await net.computeFaceDescriptor(img))
@@ -47,9 +47,9 @@ async function initTrainDescriptorsByClass(net, numImagesForTraining = 1) {
 
 function getBestMatch(descriptorsByClass, queryDescriptor) {
   function computeMeanDistance(descriptorsOfClass) {
-    return facerecognition.round(
+    return faceapi.round(
       descriptorsOfClass
-        .map(d => facerecognition.euclideanDistance(d, queryDescriptor))
+        .map(d => faceapi.euclideanDistance(d, queryDescriptor))
         .reduce((d1, d2) => d1 + d2, 0)
           / (descriptorsOfClass.length || 1)
       )

examples/views/detectAndDrawFaces.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -46,13 +46,13 @@
     let net
 
     function onIncreaseThreshold() {
-      minConfidence = Math.min(facerecognition.round(minConfidence + 0.1), 1.0)
+      minConfidence = Math.min(faceapi.round(minConfidence + 0.1), 1.0)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
 
     function onDecreaseThreshold() {
-      minConfidence = Math.max(facerecognition.round(minConfidence - 0.1), 0.1)
+      minConfidence = Math.max(faceapi.round(minConfidence - 0.1), 0.1)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
@@ -64,18 +64,18 @@
       canvas.width = width
       canvas.height = height
 
-      const input = new facerecognition.NetInput(inputImgEl)
+      const input = new faceapi.NetInput(inputImgEl)
       const detections = await net.locateFaces(input, minConfidence)
-      facerecognition.drawDetection('overlay', detections.map(det => det.forSize(width, height)))
+      faceapi.drawDetection('overlay', detections.map(det => det.forSize(width, height)))
 
-      const faceImages = await facerecognition.extractFaces(input.canvases[0], detections)
+      const faceImages = await faceapi.extractFaces(input.canvases[0], detections)
       $('#facesContainer').empty()
       faceImages.forEach(canvas => $('#facesContainer').append(canvas))
     }
 
     async function onSelectionChanged(uri) {
       const imgBuf = await fetchImage(uri)
-      $(`#inputImg`).get(0).src = (await facerecognition.bufferToImage(imgBuf)).src
+      $(`#inputImg`).get(0).src = (await faceapi.bufferToImage(imgBuf)).src
       updateResults()
     }
 

examples/views/detectAndRecognizeFaces.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -47,13 +47,13 @@
     let trainDescriptorsByClass = []
 
     function onIncreaseThreshold() {
-      minConfidence = Math.min(facerecognition.round(minConfidence + 0.1), 1.0)
+      minConfidence = Math.min(faceapi.round(minConfidence + 0.1), 1.0)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
 
     function onDecreaseThreshold() {
-      minConfidence = Math.max(facerecognition.round(minConfidence - 0.1), 0.1)
+      minConfidence = Math.max(faceapi.round(minConfidence - 0.1), 0.1)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
@@ -65,13 +65,13 @@
       canvas.width = width
       canvas.height = height
 
-      const input = new facerecognition.NetInput(inputImgEl)
+      const input = new faceapi.NetInput(inputImgEl)
       const detections = await detectionNet.locateFaces(input, minConfidence)
 
       const detectionsForSize = detections.map(det => det.forSize(width, height))
-      facerecognition.drawDetection('overlay', detectionsForSize, { withScore: false })
+      faceapi.drawDetection('overlay', detectionsForSize, { withScore: false })
 
-      const faceTensors = (await facerecognition.extractFaceTensors(input, detections))
+      const faceTensors = (await faceapi.extractFaceTensors(input, detections))
       const descriptors = await Promise.all(faceTensors.map(t => recognitionNet.computeFaceDescriptor(t)))
 
       // free memory for face image tensors after we computed their descriptors
@@ -81,13 +81,13 @@
         const bestMatch = getBestMatch(trainDescriptorsByClass, descriptor)
         const text = `${bestMatch.distance < threshold ? bestMatch.className : 'unkown'} (${bestMatch.distance})`
         const { x, y } = detectionsForSize[i].box
-        facerecognition.drawText(canvas.getContext('2d'), x, y, text, facerecognition.getDefaultDrawOptions())
+        faceapi.drawText(canvas.getContext('2d'), x, y, text, faceapi.getDefaultDrawOptions())
       })
     }
 
     async function onSelectionChanged(uri) {
       const imgBuf = await fetchImage(uri)
-      $(`#inputImg`).get(0).src = (await facerecognition.bufferToImage(imgBuf)).src
+      $(`#inputImg`).get(0).src = (await faceapi.bufferToImage(imgBuf)).src
       updateResults()
     }
 

examples/views/faceDetection.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -45,13 +45,13 @@
     let net, result
 
     function onIncreaseThreshold() {
-      minConfidence = Math.min(facerecognition.round(minConfidence + 0.1), 1.0)
+      minConfidence = Math.min(faceapi.round(minConfidence + 0.1), 1.0)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
 
     function onDecreaseThreshold() {
-      minConfidence = Math.max(facerecognition.round(minConfidence - 0.1), 0.1)
+      minConfidence = Math.max(faceapi.round(minConfidence - 0.1), 0.1)
       $('#minConfidence').val(minConfidence)
       updateResults()
     }
@@ -63,14 +63,14 @@
       canvas.width = width
       canvas.height = height
 
-      const input = new facerecognition.NetInput(inputImgEl)
+      const input = new faceapi.NetInput(inputImgEl)
       result = await net.locateFaces(input, minConfidence)
-      facerecognition.drawDetection('overlay', result.map(det => det.forSize(width, height)))
+      faceapi.drawDetection('overlay', result.map(det => det.forSize(width, height)))
     }
 
     async function onSelectionChanged(uri) {
       const imgBuf = await fetchImage(uri)
-      $(`#inputImg`).get(0).src = (await facerecognition.bufferToImage(imgBuf)).src
+      $(`#inputImg`).get(0).src = (await faceapi.bufferToImage(imgBuf)).src
       updateResults()
     }
 

examples/views/faceDetectionVideo.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -55,25 +55,25 @@
     let net, result
 
     function onIncreaseThreshold() {
-      minConfidence = Math.min(facerecognition.round(minConfidence + 0.1), 1.0)
+      minConfidence = Math.min(faceapi.round(minConfidence + 0.1), 1.0)
       $('#minConfidence').val(minConfidence)
     }
 
     function onDecreaseThreshold() {
-      minConfidence = Math.max(facerecognition.round(minConfidence - 0.1), 0.1)
+      minConfidence = Math.max(faceapi.round(minConfidence - 0.1), 0.1)
       $('#minConfidence').val(minConfidence)
     }
 
     function displayTimeStats(timeInMs) {
       $('#time').val(`${timeInMs} ms`)
-      $('#fps').val(`${facerecognition.round(1000 / timeInMs)}`)
+      $('#fps').val(`${faceapi.round(1000 / timeInMs)}`)
     }
 
     async function onPlay(videoEl) {
       if(videoEl.paused || videoEl.ended)
         return false
 
-      const input = new facerecognition.NetInput(videoEl)
+      const input = new faceapi.NetInput(videoEl)
       const { width, height } = input
       const canvas = $('#overlay').get(0)
       canvas.width = width
@@ -83,7 +83,7 @@
       result = await net.locateFaces(input, minConfidence)
       displayTimeStats(Date.now() - ts)
 
-      facerecognition.drawDetection('overlay', result.map(det => det.forSize(width, height)))
+      faceapi.drawDetection('overlay', result.map(det => det.forSize(width, height)))
       setTimeout(() => onPlay(videoEl))
     }
 

examples/views/faceRecognition.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -101,7 +101,7 @@
 
     function displayTimeStats(timeInMs) {
       $('#time').val(`${timeInMs} ms`)
-      $('#fps').val(`${facerecognition.round(1000 / timeInMs)}`)
+      $('#fps').val(`${faceapi.round(1000 / timeInMs)}`)
     }
 
     function displayImage(src) {
@@ -111,7 +111,7 @@
     async function runFaceRecognition() {
       async function next() {
         const imgBuf = await fetchImage(getFaceImageUri(classes[currClassIdx], currImageIdx))
-        const input = await facerecognition.bufferToImage(imgBuf)
+        const input = await faceapi.bufferToImage(imgBuf)
         const imgEl = $('#face').get(0)
         imgEl.src = input.src
 

examples/views/faceSimilarity.html

 <!DOCTYPE html>
 <html>
 <head>
-  <script src="face-recognition.js"></script>
+  <script src="face-api.js"></script>
   <script src="axios.min.js"></script>
   <script src="commons.js"></script>
   <link rel="stylesheet" href="styles.css">
@@ -40,8 +40,8 @@
     let net, descriptors = { desc1: null, desc2: null }
 
     function updateResult() {
-      const distance = facerecognition.round(
-        facerecognition.euclideanDistance(descriptors.desc1, descriptors.desc2)
+      const distance = faceapi.round(
+        faceapi.euclideanDistance(descriptors.desc1, descriptors.desc2)
       )
       let text = distance
       let bgColor = '#ffffff'
@@ -55,7 +55,7 @@
 
     async function onSelectionChanged(which, uri) {
       const imgBuf = await fetchImage(uri)
-      const input = await facerecognition.bufferToImage(imgBuf)
+      const input = await faceapi.bufferToImage(imgBuf)
       const imgEl = $(`#face${which}`).get(0)
       imgEl.src = input.src
       descriptors[`desc${which}`] = await net.computeFaceDescriptor(input)

package.json

 {
-  "name": "face-recognition.min",
+  "name": "face-api",
   "version": "0.0.0",
-  "description": "face recognition API for the browser with tensorflow.js",
+  "description": "JavaScript API for face detection and face recognition in the browser with tensorflow.js",
   "main": "./dist/index.js",
   "typings": "./dist/index.d.ts",
   "scripts": {

rollup.config.js

@@ -15,9 +15,9 @@ export default {
   ].concat(minify ? uglify() : []),
   output: {
     extend: true,
-    file: `dist/face-recognition${minify ? '.min' : ''}.js`,
+    file: `dist/face-api${minify ? '.min' : ''}.js`,
     format: 'umd',
-    name: 'facerecognition',
+    name: 'faceapi',
     globals: {
       'crypto': 'crypto'
     }

test/descriptors/howard.json

-descriptorHoward=[-0.08900658041238785,0.10903991758823395,0.027176208794116974,0.0440075621008873,-0.14542894065380096,0.11052004992961884,-0.044826459139585495,-0.05154901742935181,0.10313282907009125,-0.09580706059932709,0.11335687339305878,-0.027231775224208832,-0.20172204077243805,0.0940278172492981,-0.02581452578306198,0.07219456881284714,-0.12272307276725769,-0.07349634170532227,-0.17236188054084778,-0.17453305423259735,-0.034208014607429504,0.1051197499036789,0.026275131851434708,0.014430046081542969,-0.20353534817695618,-0.2949812114238739,-0.04833771288394928,-0.10960748046636581,0.08448511362075806,-0.03991013765335083,-0.03964321315288544,-0.09928630292415619,-0.1602567881345749,0.026378951966762543,0.09079921245574951,0.07745552062988281,-0.054152462631464005,-0.017411045730113983,0.16053830087184906,0.010681785643100739,-0.11814303696155548,0.03829622268676758,0.08098047226667404,0.2989161014556885,0.12581878900527954,0.0647912546992302,0.023303285241127014,-0.07838225364685059,0.13633489608764648,-0.21215596795082092,0.07675531506538391,0.1447518765926361,0.14686475694179535,0.0699121281504631,0.08843745291233063,-0.11935200542211533,-0.015284918248653412,0.16930952668190002,-0.04400303214788437,0.1650175303220749,0.10481946915388107,-0.01336788758635521,-0.050796136260032654,-0.0797152891755104,0.2541898190975189,0.07128539681434631,-0.14587090909481049,-0.15604129433631897,0.11365237832069397,-0.16018034517765045,-0.03458000719547272,0.05678928643465042,-0.0719192773103714,-0.15881866216659546,-0.1955045610666275,0.06456597149372101,0.5308966636657715,0.13605226576328278,-0.18340086936950684,-0.05473683401942253,-0.09668048471212387,-0.000602424144744873,0.06609033793210983,0.08351708948612213,-0.13018563389778137,-0.07167275249958038,-0.043135225772857666,0.08809376507997513,0.2999389171600342,-0.07008984684944153,0.005112119019031525,0.1464608609676361,0.030642826110124588,0.005341166630387306,-0.03758299723267555,-0.0027411580085754395,-0.19020092487335205,-0.005203835666179657,-0.03693883866071701,0.01771560311317444,0.0251515731215477,-0.13933824002742767,0.04255777597427368,0.08094561100006104,-0.23745450377464294,0.21049562096595764,-0.016159698367118835,-0.06422223895788193,0.0915207490324974,0.10660701990127563,-0.14731749892234802,-0.027426909655332565,0.23789143562316895,-0.2964036166667938,0.2034282386302948,0.2009483426809311,0.04705991595983505,0.1396426558494568,0.05233515799045563,0.11507779359817505,0.045886870473623276,0.12765640020370483,-0.15917259454727173,-0.13223722577095032,-0.023241132497787476,-0.1298847794532776,-0.027176383882761,0.009421631693840027]
\ No newline at end of file

test/descriptors/leonard.json

-descriptorLeonard=[0.016118615865707397,0.1272888332605362,-0.013150867074728012,-0.03657906502485275,-0.10901328921318054,-0.004170142114162445,-0.010215671733021736,0.006740286946296692,0.1793280392885208,-0.06382005661725998,0.20762376487255096,-0.016507171094417572,-0.2747085690498352,-0.026863690465688705,-0.0744708999991417,0.1404581367969513,-0.198239266872406,-0.12842532992362976,-0.15176594257354736,-0.12798485159873962,0.07055014371871948,-0.021530020982027054,-0.013443628326058388,0.05855056643486023,-0.10384566336870193,-0.26168593764305115,-0.04933137446641922,-0.10280363261699677,0.02998245507478714,-0.21771246194839478,0.07944433391094208,0.03573431074619293,-0.1271427869796753,-0.02638978511095047,-0.01610453426837921,0.05286967754364014,0.024109765887260437,-0.08603353798389435,0.18842941522598267,0.02356734871864319,-0.16014565527439117,0.07457999885082245,0.04670367389917374,0.32030463218688965,0.18153166770935059,0.031080730259418488,-0.01877094805240631,-0.09274949133396149,0.13673648238182068,-0.20436367392539978,0.03125997632741928,0.1889854222536087,0.07329613715410233,0.03783072903752327,0.12429258227348328,-0.10134827345609665,0.060723669826984406,0.13368329405784607,-0.22554075717926025,0.032615188509225845,0.05100584030151367,-0.008536417037248611,-0.028306663036346436,-0.09284669905900955,0.10282410681247711,0.05005515366792679,-0.05751366168260574,-0.16138313710689545,0.1641443520784378,-0.21432684361934662,-0.1301480382680893,0.051546234637498856,-0.1041136085987091,-0.1479661911725998,-0.32189327478408813,0.0080157071352005,0.40646892786026,0.17670349776744843,-0.11870051920413971,0.06668459624052048,-0.0077753327786922455,-0.0853877067565918,0.03622785955667496,0.022207416594028473,-0.1716664731502533,0.00936036929488182,-0.12650445103645325,0.1116786003112793,0.1721886694431305,0.018712684512138367,-0.029012983664870262,0.19890916347503662,0.004050761461257935,-0.06176470220088959,0.03496668487787247,0.010774612426757812,-0.10733450204133987,0.034917011857032776,-0.1856822967529297,-0.0436706580221653,0.08982815593481064,-0.16002188622951508,-0.01822887361049652,0.05696277320384979,-0.2110522985458374,0.03147541731595993,-0.009835068136453629,-0.059302788227796555,-0.060756754130125046,0.07512637972831726,-0.20573465526103973,0.023113828152418137,0.2548554837703705,-0.23949584364891052,0.17401018738746643,0.2319977581501007,0.09833789616823196,0.024068880826234818,0.12989574670791626,0.04740560054779053,-0.012434778735041618,-0.0918111503124237,-0.15748904645442963,-0.08378004282712936,-0.004624858498573303,-0.005299612879753113,0.055329449474811554,0.04065752774477005]
\ No newline at end of file

test/descriptors/penny.json

-descriptorPenny=[-0.005881071090698242,0.035252682864665985,0.07666284590959549,-0.06133250892162323,-0.08197010308504105,0.014184653759002686,0.005242734216153622,-0.19035089015960693,0.13593840599060059,-0.09199994057416916,0.055665574967861176,-0.09802958369255066,-0.25256460905075073,0.05049600824713707,0.003990292549133301,0.15637768805027008,-0.051415905356407166,-0.28471335768699646,-0.16898605227470398,-0.06020534038543701,-0.006151877343654633,-0.011471755802631378,-0.005520425736904144,0.1003769263625145,-0.20014473795890808,-0.26222628355026245,-0.08959437161684036,-0.12957212328910828,-0.029134854674339294,-0.062256816774606705,-0.04312460869550705,0.08265750110149384,-0.14303170144557953,0.011085912585258484,0.10712425410747528,0.04088181257247925,0.0007619466632604599,-0.1466120481491089,0.260177880525589,-0.009320661425590515,-0.25915971398353577,-0.019178807735443115,0.11162054538726807,0.2115343064069748,0.21638226509094238,0.0045770928263664246,0.07745259255170822,-0.08094276487827301,0.13608475029468536,-0.31285080313682556,0.1226205825805664,0.14514890313148499,0.10254833102226257,0.11901462823152542,0.07176060229539871,-0.21035197377204895,-0.016285259276628494,0.11171358823776245,-0.22099058330059052,0.14237764477729797,0.06349137425422668,-0.09368033707141876,-0.07181331515312195,-0.12528066337108612,0.22090691328048706,0.16226458549499512,-0.16031241416931152,-0.2046929895877838,0.1601838767528534,-0.2115079164505005,-0.07380392402410507,0.06603492796421051,-0.19789603352546692,-0.15601742267608643,-0.285112589597702,-0.03280213475227356,0.41335564851760864,0.17521080374717712,-0.2080937772989273,0.026215683668851852,-0.11048824340105057,0.019838571548461914,-0.020542338490486145,0.11541588604450226,0.03076190873980522,-0.0232694149017334,0.0033110976219177246,0.028120767325162888,0.27188798785209656,-0.0759110227227211,-0.06393658369779587,0.22212064266204834,-0.048158712685108185,-0.03439009189605713,-0.08470702916383743,0.07175514847040176,-0.09915667027235031,-0.019413530826568604,-0.08236535638570786,-0.0213082954287529,-0.030551567673683167,0.1010485291481018,0.07094472646713257,0.15863054990768433,-0.200668603181839,0.16280920803546906,-0.07447126507759094,-0.007395192980766296,-0.023165743798017502,0.01478651911020279,-0.1056482344865799,-0.07202574610710144,0.17325882613658905,-0.28997519612312317,0.12274238467216492,0.18649500608444214,0.01748467981815338,0.03461623191833496,-0.009980626404285431,0.06018088012933731,0.11674463003873825,0.033887870609760284,-0.218181312084198,-0.05439690500497818,0.10635554790496826,-0.0007329434156417847,-0.021141983568668365,0.0820411890745163]
\ No newline at end of file

test/descriptors/raj.json

-descriptorRaj=[-0.15496353805065155,0.043691255152225494,0.03724939376115799,0.016656994819641113,-0.049970414489507675,0.08455478399991989,-0.045141857117414474,-0.0610889196395874,0.11937284469604492,-0.08181658387184143,0.27349838614463806,0.025957435369491577,-0.2195511758327484,-0.015493396669626236,-0.10711826384067535,0.07597102224826813,-0.08671779185533524,-0.13454240560531616,-0.06495959311723709,-0.036077070981264114,-0.006286047399044037,0.02592059224843979,-0.014918237924575806,0.009378507733345032,-0.11585316807031631,-0.3628930449485779,-0.10048630833625793,-0.07691314816474915,0.001719092484563589,-0.03902893140912056,-0.03469265252351761,-0.020092345774173737,-0.19703790545463562,0.008729912340641022,0.03802715986967087,0.09341692179441452,-0.11134380102157593,-0.009015034884214401,0.14202813804149628,0.026977384462952614,-0.09668654203414917,0.0022040903568267822,0.031198769807815552,0.2793624997138977,0.17439183592796326,0.07265886664390564,0.006268925964832306,-0.13531382381916046,0.05125536024570465,-0.2435196340084076,0.10533908754587173,0.1630076915025711,0.0991237536072731,0.11345860362052917,0.07629163563251495,-0.16580615937709808,-0.024815142154693604,0.15595972537994385,-0.14484508335590363,0.04327581822872162,-0.053320854902267456,0.03464396297931671,-0.005613304674625397,-0.11897846311330795,0.18600404262542725,0.15069319307804108,-0.11564309895038605,-0.18291205167770386,0.1789712905883789,-0.18346670269966125,-0.017721213400363922,0.0858432874083519,-0.07925683259963989,-0.2650236487388611,-0.24118715524673462,0.10106492042541504,0.41431039571762085,0.2085058093070984,-0.09770824015140533,0.02742883935570717,-0.05898953229188919,-0.07701464742422104,0.028122015297412872,-0.0040946416556835175,-0.10734926909208298,-0.016904285177588463,-0.06405238062143326,0.008443355560302734,0.201746866106987,0.04558864235877991,-0.04860920086503029,0.1356211006641388,-0.06206119433045387,0.005768755450844765,0.07818647474050522,0.05297157168388367,-0.10372574627399445,0.042196549475193024,-0.16179141402244568,-0.031241916120052338,-0.11580236256122589,-0.08878964930772781,0.02395106852054596,0.059674158692359924,-0.14112776517868042,0.16106471419334412,-0.01805580034852028,-0.036857277154922485,-0.029434598982334137,0.0012199431657791138,-0.13350501656532288,0.02282126620411873,0.1603221446275711,-0.33153676986694336,0.22605177760124207,0.11830978095531464,0.11549906432628632,0.21082323789596558,0.11426876485347748,0.05354222655296326,-0.03111746534705162,-0.025937475264072418,-0.2007763385772705,-0.04948417842388153,-0.0020427852869033813,-0.045253459364175797,0.04411966726183891,0.02899453043937683]
\ No newline at end of file

test/descriptors/sheldon.json

-descriptorSheldon=[-0.04457738250494003,-0.043990347534418106,-0.025750618427991867,0.016650959849357605,-0.05218123272061348,-0.051934950053691864,0.009930811822414398,-0.016631752252578735,0.11751081049442291,0.040901102125644684,0.221496120095253,-0.08035802841186523,-0.23618969321250916,-0.025104589760303497,-0.02730524353682995,0.10455028712749481,-0.12426766008138657,-0.08866177499294281,-0.14048157632350922,-0.0403841957449913,-0.041014257818460464,0.11965540796518326,-0.015900693833827972,0.08545821160078049,-0.10701243579387665,-0.29199138283729553,-0.09832726418972015,-0.15806680917739868,0.0072528645396232605,-0.09897308796644211,0.02866358309984207,0.17405939102172852,-0.17985643446445465,-0.11151746660470963,0.12910649180412292,0.023578159511089325,-0.048720214515924454,-0.023027973249554634,0.24042245745658875,0.09909329563379288,-0.1479608118534088,-0.03488050401210785,0.023121362552046776,0.35197967290878296,0.1343260109424591,0.025992773473262787,0.027310341596603394,-0.07258803397417068,0.09854952991008759,-0.265401691198349,0.09129731357097626,0.18322107195854187,0.14081576466560364,0.08943489193916321,0.07813340425491333,-0.18660837411880493,0.012191173620522022,0.13849547505378723,-0.19889040291309357,0.14307448267936707,0.04158569499850273,-0.06411264091730118,-0.0019705994054675102,-0.10619327425956726,0.1386486440896988,0.019395824521780014,-0.10675810277462006,-0.06835392862558365,0.19760599732398987,-0.10146252810955048,-0.005423944443464279,0.13664257526397705,-0.10927100479602814,-0.24490387737751007,-0.27884605526924133,0.025820117443799973,0.4047747850418091,0.08949815481901169,-0.21505051851272583,-0.004949783906340599,-0.055362775921821594,-0.06972619891166687,-0.008085280656814575,0.027283571660518646,-0.0860084667801857,0.017455322667956352,-0.07179910689592361,0.0487312376499176,0.2027968317270279,0.011654987931251526,-0.007634447887539864,0.1842338740825653,-0.06324949115514755,-0.07316185534000397,0.122040756046772,0.08807042986154556,-0.050539594143629074,-0.08226519823074341,-0.11426769942045212,0.000004231929779052734,0.027701571583747864,-0.2017858624458313,-0.021970629692077637,0.056306734681129456,-0.17306512594223022,0.20115011930465698,0.008511713705956936,-0.07041959464550018,-0.08531112968921661,-0.04338337481021881,-0.07298384606838226,0.07277841866016388,0.2568773627281189,-0.24525469541549683,0.21955132484436035,0.07719040662050247,0.011913388967514038,0.17791587114334106,-0.05779615789651871,0.10519659519195557,-0.1535077542066574,-0.09388455748558044,-0.1587708741426468,-0.14040732383728027,0.0359625518321991,0.04092983901500702,0.04888205975294113,0.014610011130571365]
\ No newline at end of file

test/test.html

-<!DOCTYPE html>
-<html>
-<head>
-  <script src="./imgdata/howard.json"></script>
-  <script src="./imgdata/leonard.json"></script>
-  <script src="./imgdata/penny.json"></script>
-  <script src="./imgdata/raj.json"></script>
-  <script src="./imgdata/sheldon.json"></script>
-  <script src="./descriptors/howard.json"></script>
-  <script src="./descriptors/leonard.json"></script>
-  <script src="./descriptors/penny.json"></script>
-  <script src="./descriptors/raj.json"></script>
-  <script src="./descriptors/sheldon.json"></script>
-  <script src="../dist/face-recognition.min.js"></script>
-</head>
-<body>
-  <label for="file">Load Weights File: </label>
-  <input type="file" onchange="onWeightsSelected(event)"/>
-  <script>
-    function onWeightsSelected(e) {
-      const selectedFile = e.target.files[0]
-      const reader = new FileReader()
-      reader.onload = function(re) {
-        const weights = new Float32Array(re.target.result)
-        runTests(weights)
-      }
-      reader.readAsArrayBuffer(selectedFile)
-    }
-
-    async function runTests(weights) {
-      console.log('running...')
-      const net = facerecognition.faceRecognitionNet(weights)
-      await run(imgdataHoward, descriptorHoward, net, 'howard')
-      await run(imgdataLeonard, descriptorLeonard, net, 'leonard')
-      await run(imgdataPenny, descriptorPenny, net, 'penny')
-      await run(imgdataRaj, descriptorRaj, net, 'raj')
-      await run(imgdataSheldon, descriptorSheldon, net, 'howard')
-      console.log('done')
-    }
-
-    async function run(data, refDescriptor, net, name) {
-      const input = flatten(flatten(data))
-      console.time('computeFaceDescriptor')
-      const desc = await net.computeFaceDescriptor(input)
-      console.timeEnd('computeFaceDescriptor')
-      const distance = facerecognition.euclideanDistance(desc, refDescriptor)
-
-      if (distance > 1e-6)
-        console.error('failed for descriptor %s with distance %s', name, distance)
-    }
-
-    function flatten(arr) {
-      return arr.reduce((res, curr) => res.concat(curr), [])
-    }
-  </script>
-
-</body>
-</html>
\ No newline at end of file