Merge pull request #504 from justadudewhohacks/remove-tfjs-image-recognition-base

examples/examples-browser/public/js/faceDetectionControls.js

@@ -26,13 +26,13 @@ function getFaceDetectorOptions() {
 }
 
 function onIncreaseMinConfidence() {
-  minConfidence = Math.min(faceapi.round(minConfidence + 0.1), 1.0)
+  minConfidence = Math.min(faceapi.utils.round(minConfidence + 0.1), 1.0)
   $('#minConfidence').val(minConfidence)
   updateResults()
 }
 
 function onDecreaseMinConfidence() {
-  minConfidence = Math.max(faceapi.round(minConfidence - 0.1), 0.1)
+  minConfidence = Math.max(faceapi.utils.round(minConfidence - 0.1), 0.1)
   $('#minConfidence').val(minConfidence)
   updateResults()
 }
@@ -51,24 +51,24 @@ function changeInputSize(size) {
 }
 
 function onIncreaseScoreThreshold() {
-  scoreThreshold = Math.min(faceapi.round(scoreThreshold + 0.1), 1.0)
+  scoreThreshold = Math.min(faceapi.utils.round(scoreThreshold + 0.1), 1.0)
   $('#scoreThreshold').val(scoreThreshold)
   updateResults()
 }
 
 function onDecreaseScoreThreshold() {
-  scoreThreshold = Math.max(faceapi.round(scoreThreshold - 0.1), 0.1)
+  scoreThreshold = Math.max(faceapi.utils.round(scoreThreshold - 0.1), 0.1)
   $('#scoreThreshold').val(scoreThreshold)
   updateResults()
 }
 
 function onIncreaseMinFaceSize() {
-  minFaceSize = Math.min(faceapi.round(minFaceSize + 20), 300)
+  minFaceSize = Math.min(faceapi.utils.round(minFaceSize + 20), 300)
   $('#minFaceSize').val(minFaceSize)
 }
 
 function onDecreaseMinFaceSize() {
-  minFaceSize = Math.max(faceapi.round(minFaceSize - 20), 50)
+  minFaceSize = Math.max(faceapi.utils.round(minFaceSize - 20), 50)
   $('#minFaceSize').val(minFaceSize)
 }
 

examples/examples-browser/views/ageAndGenderRecognition.html

@@ -161,8 +161,8 @@
         const { age, gender, genderProbability } = result
         new faceapi.draw.DrawTextField(
           [
-            `${faceapi.round(age, 0)} years`,
-            `${gender} (${faceapi.round(genderProbability)})`
+            `${faceapi.utils.round(age, 0)} years`,
+            `${gender} (${faceapi.utils.round(genderProbability)})`
           ],
           result.detection.box.bottomLeft
         ).draw(canvas)

examples/examples-browser/views/bbtFaceMatching.html

@@ -96,7 +96,7 @@
 
     function displayTimeStats(timeInMs) {
       $('#time').val(`${timeInMs} ms`)
-      $('#fps').val(`${faceapi.round(1000 / timeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / timeInMs)}`)
     }
 
     function displayImage(src) {

examples/examples-browser/views/bbtFaceSimilarity.html

@@ -39,7 +39,7 @@
     let descriptors = { desc1: null, desc2: null }
 
     function updateResult() {
-      const distance = faceapi.round(
+      const distance = faceapi.utils.round(
         faceapi.euclideanDistance(descriptors.desc1, descriptors.desc2)
       )
       let text = distance

examples/examples-browser/views/videoFaceTracking.html

@@ -156,7 +156,7 @@
       forwardTimes = [timeInMs].concat(forwardTimes).slice(0, 30)
       const avgTimeInMs = forwardTimes.reduce((total, t) => total + t) / forwardTimes.length
       $('#time').val(`${Math.round(avgTimeInMs)} ms`)
-      $('#fps').val(`${faceapi.round(1000 / avgTimeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / avgTimeInMs)}`)
     }
 
     async function onPlay(videoEl) {

examples/examples-browser/views/webcamAgeAndGenderRecognition.html

@@ -152,7 +152,7 @@
       forwardTimes = [timeInMs].concat(forwardTimes).slice(0, 30)
       const avgTimeInMs = forwardTimes.reduce((total, t) => total + t) / forwardTimes.length
       $('#time').val(`${Math.round(avgTimeInMs)} ms`)
-      $('#fps').val(`${faceapi.round(1000 / avgTimeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / avgTimeInMs)}`)
     }
 
     function interpolateAgePredictions(age) {
@@ -192,8 +192,8 @@
         const interpolatedAge = interpolateAgePredictions(age)
         new faceapi.draw.DrawTextField(
           [
-            `${faceapi.round(interpolatedAge, 0)} years`,
-            `${gender} (${faceapi.round(genderProbability)})`
+            `${faceapi.utils.round(interpolatedAge, 0)} years`,
+            `${gender} (${faceapi.utils.round(genderProbability)})`
           ],
           result.detection.box.bottomLeft
         ).draw(canvas)

examples/examples-browser/views/webcamFaceDetection.html

@@ -139,7 +139,7 @@
       forwardTimes = [timeInMs].concat(forwardTimes).slice(0, 30)
       const avgTimeInMs = forwardTimes.reduce((total, t) => total + t) / forwardTimes.length
       $('#time').val(`${Math.round(avgTimeInMs)} ms`)
-      $('#fps').val(`${faceapi.round(1000 / avgTimeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / avgTimeInMs)}`)
     }
 
     async function onPlay() {

examples/examples-browser/views/webcamFaceExpressionRecognition.html

@@ -151,7 +151,7 @@
       forwardTimes = [timeInMs].concat(forwardTimes).slice(0, 30)
       const avgTimeInMs = forwardTimes.reduce((total, t) => total + t) / forwardTimes.length
       $('#time').val(`${Math.round(avgTimeInMs)} ms`)
-      $('#fps').val(`${faceapi.round(1000 / avgTimeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / avgTimeInMs)}`)
     }
 
     async function onPlay() {

examples/examples-browser/views/webcamFaceLandmarkDetection.html

@@ -151,7 +151,7 @@
       forwardTimes = [timeInMs].concat(forwardTimes).slice(0, 30)
       const avgTimeInMs = forwardTimes.reduce((total, t) => total + t) / forwardTimes.length
       $('#time').val(`${Math.round(avgTimeInMs)} ms`)
-      $('#fps').val(`${faceapi.round(1000 / avgTimeInMs)}`)
+      $('#fps').val(`${faceapi.utils.round(1000 / avgTimeInMs)}`)
     }
 
     async function onPlay() {

examples/examples-nodejs/ageAndGenderRecognition.ts

@@ -19,8 +19,8 @@ async function run() {
     const { age, gender, genderProbability } = result
     new faceapi.draw.DrawTextField(
       [
-        `${faceapi.round(age, 0)} years`,
-        `${gender} (${faceapi.round(genderProbability)})`
+        `${faceapi.utils.round(age, 0)} years`,
+        `${gender} (${faceapi.utils.round(genderProbability)})`
       ],
       result.detection.box.bottomLeft
     ).draw(out)

jasmine-node.js

-let spec_files = ['**/*.test.ts'].concat(
-  process.env.EXCLUDE_UNCOMPRESSED
-    ? ['!**/*.uncompressed.test.ts']
-    : []
-)
+let spec_files = ['**/*.test.ts']
 
 // exclude browser tests
 spec_files = spec_files.concat(['!**/*.browser.test.ts'])
+spec_files = spec_files.concat(['!test/tests.legacy/*'])
 
 module.exports = {
   spec_dir: 'test',

karma.conf.js

@@ -2,6 +2,7 @@ const dataFiles = [
   'test/images/*.jpg',
   'test/images/*.png',
   'test/data/*.json',
+  'test/data/*.weights',
   'test/media/*.mp4',
   'weights/**/*',
   'weights_uncompressed/**/*',
@@ -21,24 +22,17 @@ let exclude = (
         'faceLandmarkNet',
         'faceRecognitionNet',
         'ssdMobilenetv1',
-        'tinyFaceDetector',
-        'mtcnn'
+        'tinyFaceDetector'
       ]
     : []
   )
     .filter(ex => ex !== process.env.UUT)
     .map(ex => `test/tests/${ex}/*.ts`)
 
-
-exclude = exclude.concat(
-  process.env.EXCLUDE_UNCOMPRESSED
-    ? ['**/*.uncompressed.test.ts']
-    : []
-)
-
 // exclude nodejs tests
 exclude = exclude.concat(['**/*.node.test.ts'])
 exclude = exclude.concat(['test/env.node.ts'])
+exclude = exclude.concat(['test/tests-legacy/**/*.ts'])
 
 
 module.exports = function(config) {

package-lock.json

@@ -1487,9 +1487,9 @@
       "dev": true
     },
     "fsevents": {
-      "version": "2.0.7",
-      "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.0.7.tgz",
-      "integrity": "sha512-a7YT0SV3RB+DjYcppwVDLtn13UQnmg0SWZS7ezZD0UjnLwXmy8Zm21GMVGLaFGimIqcvyMQaOJBrop8MyOp1kQ==",
+      "version": "2.1.2",
+      "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.1.2.tgz",
+      "integrity": "sha512-R4wDiBwZ0KzpgOWetKDug1FZcYhqYnUYKtfZYt4mD5SBz76q0KR4Q9o7GIPamsVPGmW3EYPPJ0dOOjvx32ldZA==",
       "dev": true,
       "optional": true
     },
@@ -3917,15 +3917,6 @@
         "yallist": "^3.0.3"
       }
     },
-    "tfjs-image-recognition-base": {
-      "version": "0.6.2",
-      "resolved": "https://registry.npmjs.org/tfjs-image-recognition-base/-/tfjs-image-recognition-base-0.6.2.tgz",
-      "integrity": "sha512-ukxViVfAPw7s0KiGhwr3zrwsm+EVa2Z+4aEwKBWO43Rt48nbPyVvrHdL+WbxRynZYjklEE69ft66C8zzea7vFw==",
-      "requires": {
-        "@tensorflow/tfjs-core": "^1.2.9",
-        "tslib": "^1.10.0"
-      }
-    },
     "through2": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/through2/-/through2-3.0.0.tgz",

package.json

@@ -11,22 +11,16 @@
     "tsc": "tsc",
     "tsc-es6": "tsc --p tsconfig.es6.json",
     "build": "rm -rf ./build && rm -rf ./dist && npm run rollup && npm run rollup-min && npm run tsc && npm run tsc-es6",
-    "test": "karma start",
+    "test": "npm run test-browser && npm run test-node",
     "test-browser": "karma start --single-run",
     "test-node": "ts-node -r ./test/env.node.ts node_modules/jasmine/bin/jasmine --config=jasmine-node.js",
-    "test-all": "npm run test-browser-exclude-uncompressed && npm run test-node-exclude-uncompressed",
-    "test-all-include-uncompressed": "npm run test-browser && npm run test-node",
     "test-facelandmarknets": "set UUT=faceLandmarkNet&& karma start",
     "test-facerecognitionnet": "set UUT=faceRecognitionNet&& karma start",
     "test-agegendernet": "set UUT=ageGenderNet&& karma start",
     "test-ssdmobilenetv1": "set UUT=ssdMobilenetv1&& karma start",
     "test-tinyfacedetector": "set UUT=tinyFaceDetector&& karma start",
     "test-globalapi": "set UUT=globalApi&& karma start",
-    "test-mtcnn": "set UUT=mtcnn&& karma start",
     "test-cpu": "set BACKEND_CPU=true&& karma start",
-    "test-exclude-uncompressed": "set EXCLUDE_UNCOMPRESSED=true&& karma start",
-    "test-browser-exclude-uncompressed": "set EXCLUDE_UNCOMPRESSED=true&& karma start --single-run",
-    "test-node-exclude-uncompressed": "set EXCLUDE_UNCOMPRESSED=true&& npm run test-node",
     "docs": "typedoc --options ./typedoc.config.js ./src"
   },
   "keywords": [
@@ -40,7 +34,6 @@
   "license": "MIT",
   "dependencies": {
     "@tensorflow/tfjs-core": "1.2.9",
-    "tfjs-image-recognition-base": "^0.6.2",
     "tslib": "^1.10.0"
   },
   "devDependencies": {

src/NeuralNetwork.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ParamMapping } from './common';
+import { getModelUris } from './common/getModelUris';
+import { loadWeightMap } from './dom';
+import { env } from './env';
+
+export abstract class NeuralNetwork<TNetParams> {
+
+  protected _params: TNetParams | undefined = undefined
+  protected _paramMappings: ParamMapping[] = []
+
+  constructor(protected _name: string) {}
+
+  public get params(): TNetParams | undefined { return this._params }
+  public get paramMappings(): ParamMapping[] { return this._paramMappings }
+  public get isLoaded(): boolean { return !!this.params }
+
+  public getParamFromPath(paramPath: string): tf.Tensor {
+    const { obj, objProp } = this.traversePropertyPath(paramPath)
+    return obj[objProp]
+  }
+
+  public reassignParamFromPath(paramPath: string, tensor: tf.Tensor) {
+    const { obj, objProp } = this.traversePropertyPath(paramPath)
+    obj[objProp].dispose()
+    obj[objProp] = tensor
+  }
+
+  public getParamList() {
+    return this._paramMappings.map(({ paramPath }) => ({
+      path: paramPath,
+      tensor: this.getParamFromPath(paramPath)
+    }))
+  }
+
+  public getTrainableParams() {
+    return this.getParamList().filter(param => param.tensor instanceof tf.Variable)
+  }
+
+  public getFrozenParams() {
+    return this.getParamList().filter(param => !(param.tensor instanceof tf.Variable))
+  }
+
+  public variable() {
+    this.getFrozenParams().forEach(({ path, tensor }) => {
+      this.reassignParamFromPath(path, tensor.variable())
+    })
+  }
+
+  public freeze() {
+    this.getTrainableParams().forEach(({ path, tensor: variable }) => {
+      const tensor = tf.tensor(variable.dataSync())
+      variable.dispose()
+      this.reassignParamFromPath(path, tensor)
+    })
+  }
+
+  public dispose(throwOnRedispose: boolean = true) {
+    this.getParamList().forEach(param => {
+      if (throwOnRedispose && param.tensor.isDisposed) {
+        throw new Error(`param tensor has already been disposed for path ${param.path}`)
+      }
+      param.tensor.dispose()
+    })
+    this._params = undefined
+  }
+
+  public serializeParams(): Float32Array {
+    return new Float32Array(
+      this.getParamList()
+        .map(({ tensor }) => Array.from(tensor.dataSync()) as number[])
+        .reduce((flat, arr) => flat.concat(arr))
+    )
+  }
+
+  public async load(weightsOrUrl: Float32Array | string | undefined): Promise<void> {
+    if (weightsOrUrl instanceof Float32Array) {
+      this.extractWeights(weightsOrUrl)
+      return
+    }
+
+    await this.loadFromUri(weightsOrUrl)
+  }
+
+  public async loadFromUri(uri: string | undefined) {
+    if (uri && typeof uri !== 'string') {
+      throw new Error(`${this._name}.loadFromUri - expected model uri`)
+    }
+
+    const weightMap = await loadWeightMap(uri, this.getDefaultModelName())
+    this.loadFromWeightMap(weightMap)
+  }
+
+  public async loadFromDisk(filePath: string | undefined) {
+    if (filePath && typeof filePath !== 'string') {
+      throw new Error(`${this._name}.loadFromDisk - expected model file path`)
+    }
+
+    const { readFile } = env.getEnv()
+
+    const { manifestUri, modelBaseUri } = getModelUris(filePath, this.getDefaultModelName())
+
+    const fetchWeightsFromDisk = (filePaths: string[]) => Promise.all(
+      filePaths.map(filePath => readFile(filePath).then(buf => buf.buffer))
+    )
+    const loadWeights = tf.io.weightsLoaderFactory(fetchWeightsFromDisk)
+
+    const manifest = JSON.parse((await readFile(manifestUri)).toString())
+    const weightMap = await loadWeights(manifest, modelBaseUri)
+
+    this.loadFromWeightMap(weightMap)
+  }
+
+  public loadFromWeightMap(weightMap: tf.NamedTensorMap) {
+    const {
+      paramMappings,
+      params
+    } = this.extractParamsFromWeigthMap(weightMap)
+
+    this._paramMappings = paramMappings
+    this._params = params
+  }
+
+  public extractWeights(weights: Float32Array) {
+    const {
+      paramMappings,
+      params
+    } = this.extractParams(weights)
+
+    this._paramMappings = paramMappings
+    this._params = params
+  }
+
+  private traversePropertyPath(paramPath: string) {
+    if (!this.params) {
+      throw new Error(`traversePropertyPath - model has no loaded params`)
+    }
+
+    const result = paramPath.split('/').reduce((res: { nextObj: any, obj?: any, objProp?: string }, objProp) => {
+      if (!res.nextObj.hasOwnProperty(objProp)) {
+        throw new Error(`traversePropertyPath - object does not have property ${objProp}, for path ${paramPath}`)
+      }
+
+      return { obj: res.nextObj, objProp, nextObj: res.nextObj[objProp] }
+    }, { nextObj: this.params })
+
+    const { obj, objProp } = result
+    if (!obj || !objProp || !(obj[objProp] instanceof tf.Tensor)) {
+      throw new Error(`traversePropertyPath - parameter is not a tensor, for path ${paramPath}`)
+    }
+
+    return { obj, objProp }
+  }
+
+  protected abstract getDefaultModelName(): string
+  protected abstract extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap): { params: TNetParams, paramMappings: ParamMapping[] }
+  protected abstract extractParams(weights: Float32Array): { params: TNetParams, paramMappings: ParamMapping[] }
+}
\ No newline at end of file

src/ageGenderNet/AgeGenderNet.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
 import { fullyConnectedLayer } from '../common/fullyConnectedLayer';
 import { seperateWeightMaps } from '../faceProcessor/util';
@@ -7,6 +6,8 @@ import { TinyXception } from '../xception/TinyXception';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';
 import { AgeAndGenderPrediction, Gender, NetOutput, NetParams } from './types';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { NetInput, TNetInput, toNetInput } from '../dom';
 
 export class AgeGenderNet extends NeuralNetwork<NetParams> {
 

src/ageGenderNet/extractParams.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import { extractFCParamsFactory, extractWeightsFactory, ParamMapping } from '../common';
 import { NetParams } from './types';
 
-export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
-  const extractFCParams = TfjsImageRecognitionBase.extractFCParamsFactory(extractWeights, paramMappings)
+  const extractFCParams = extractFCParamsFactory(extractWeights, paramMappings)
 
   const age = extractFCParams(512, 1, 'fc/age')
   const gender = extractFCParams(512, 2, 'fc/gender')

src/ageGenderNet/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { disposeUnusedWeightTensors, extractWeightEntryFactory, FCParams, ParamMapping } from '../common';
 import { NetParams } from './types';
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
-  function extractFcParams(prefix: string): TfjsImageRecognitionBase.FCParams {
+  function extractFcParams(prefix: string): FCParams {
     const weights = extractWeightEntry<tf.Tensor2D>(`${prefix}/weights`, 2)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
     return { weights, bias }
@@ -24,7 +24,7 @@ export function extractParamsFromWeigthMap(
     }
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/ageGenderNet/types.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { FCParams } from '../common';
 
 export type AgeAndGenderPrediction = {
   age: number
@@ -16,7 +17,7 @@ export type NetOutput = { age: tf.Tensor1D, gender: tf.Tensor2D }
 
 export type NetParams = {
   fc: {
-    age: TfjsImageRecognitionBase.FCParams
-    gender: TfjsImageRecognitionBase.FCParams
+    age: FCParams
+    gender: FCParams
   }
 }
\ No newline at end of file

src/classes/BoundingBox.ts

+import { Box } from './Box';
+
+export interface IBoundingBox {
+  left: number
+  top: number
+  right: number
+  bottom: number
+}
+
+export class BoundingBox extends Box<BoundingBox> implements IBoundingBox {
+  constructor(left: number, top: number, right: number, bottom: number, allowNegativeDimensions: boolean = false) {
+    super({ left, top, right, bottom }, allowNegativeDimensions)
+  }
+}
\ No newline at end of file

src/classes/Box.ts

+import { isDimensions, isValidNumber } from '../utils';
+import { IBoundingBox } from './BoundingBox';
+import { IDimensions } from './Dimensions';
+import { Point } from './Point';
+import { IRect } from './Rect';
+
+export class Box<BoxType = any> implements IBoundingBox, IRect {
+
+  public static isRect(rect: any): boolean {
+    return !!rect && [rect.x, rect.y, rect.width, rect.height].every(isValidNumber)
+  }
+
+  public static assertIsValidBox(box: any, callee: string, allowNegativeDimensions: boolean = false) {
+    if (!Box.isRect(box)) {
+      throw new Error(`${callee} - invalid box: ${JSON.stringify(box)}, expected object with properties x, y, width, height`)
+    }
+
+    if (!allowNegativeDimensions && (box.width < 0 || box.height < 0)) {
+      throw new Error(`${callee} - width (${box.width}) and height (${box.height}) must be positive numbers`)
+    }
+  }
+
+  private _x: number
+  private _y: number
+  private _width: number
+  private _height: number
+
+  constructor(_box: IBoundingBox | IRect, allowNegativeDimensions: boolean = true) {
+    const box = (_box || {}) as any
+
+    const isBbox = [box.left, box.top, box.right, box.bottom].every(isValidNumber)
+    const isRect = [box.x, box.y, box.width, box.height].every(isValidNumber)
+
+    if (!isRect && !isBbox) {
+      throw new Error(`Box.constructor - expected box to be IBoundingBox | IRect, instead have ${JSON.stringify(box)}`)
+    }
+
+    const [x, y, width, height] = isRect
+      ? [box.x, box.y, box.width, box.height]
+      : [box.left, box.top, box.right - box.left, box.bottom - box.top]
+
+    Box.assertIsValidBox({ x, y, width, height }, 'Box.constructor', allowNegativeDimensions)
+
+    this._x = x
+    this._y = y
+    this._width = width
+    this._height = height
+  }
+
+  public get x(): number { return this._x }
+  public get y(): number { return this._y }
+  public get width(): number { return this._width }
+  public get height(): number { return this._height }
+  public get left(): number { return this.x }
+  public get top(): number { return this.y }
+  public get right(): number { return this.x + this.width }
+  public get bottom(): number { return this.y + this.height }
+  public get area(): number { return this.width * this.height }
+  public get topLeft(): Point { return new Point(this.left, this.top) }
+  public get topRight(): Point { return new Point(this.right, this.top) }
+  public get bottomLeft(): Point { return new Point(this.left, this.bottom) }
+  public get bottomRight(): Point { return new Point(this.right, this.bottom) }
+
+  public round(): Box<BoxType> {
+    const [x, y, width, height] = [this.x, this.y, this.width, this.height]
+      .map(val => Math.round(val))
+    return new Box({ x, y, width, height })
+  }
+
+  public floor(): Box<BoxType> {
+    const [x, y, width, height] = [this.x, this.y, this.width, this.height]
+      .map(val => Math.floor(val))
+    return new Box({ x, y, width, height })
+  }
+
+  public toSquare(): Box<BoxType> {
+    let { x, y, width, height } = this
+    const diff = Math.abs(width - height)
+    if (width < height) {
+      x -= (diff / 2)
+      width += diff
+    }
+    if (height < width) {
+      y -= (diff / 2)
+      height += diff
+    }
+
+    return new Box({ x, y, width, height })
+  }
+
+  public rescale(s: IDimensions | number): Box<BoxType> {
+    const scaleX = isDimensions(s) ? (s as IDimensions).width : s as number
+    const scaleY = isDimensions(s) ? (s as IDimensions).height : s as number
+    return new Box({
+      x: this.x * scaleX,
+      y: this.y * scaleY,
+      width: this.width * scaleX,
+      height: this.height * scaleY
+    })
+  }
+
+  public pad(padX: number, padY: number): Box<BoxType> {
+    let [x, y, width, height] = [
+      this.x - (padX / 2),
+      this.y - (padY / 2),
+      this.width + padX,
+      this.height + padY
+    ]
+    return new Box({ x, y, width, height })
+  }
+
+  public clipAtImageBorders(imgWidth: number, imgHeight: number): Box<BoxType> {
+    const { x, y, right, bottom } = this
+    const clippedX = Math.max(x, 0)
+    const clippedY = Math.max(y, 0)
+
+    const newWidth = right - clippedX
+    const newHeight = bottom - clippedY
+    const clippedWidth = Math.min(newWidth, imgWidth - clippedX)
+    const clippedHeight = Math.min(newHeight, imgHeight - clippedY)
+
+    return (new Box({ x: clippedX, y: clippedY, width: clippedWidth, height: clippedHeight})).floor()
+  }
+
+  public shift(sx: number, sy: number): Box<BoxType> {
+    const { width, height } = this
+    const x = this.x + sx
+    const y = this.y + sy
+
+    return new Box({ x, y, width, height })
+  }
+
+  public padAtBorders(imageHeight: number, imageWidth: number) {
+    const w = this.width + 1
+    const h = this.height + 1
+
+    let dx = 1
+    let dy = 1
+    let edx = w
+    let edy = h
+
+    let x = this.left
+    let y = this.top
+    let ex = this.right
+    let ey = this.bottom
+
+    if (ex > imageWidth) {
+      edx = -ex + imageWidth + w
+      ex = imageWidth
+    }
+    if (ey > imageHeight) {
+      edy = -ey + imageHeight + h
+      ey = imageHeight
+    }
+    if (x < 1) {
+      edy = 2 - x
+      x = 1
+    }
+    if (y < 1) {
+      edy = 2 - y
+      y = 1
+    }
+
+    return { dy, edy, dx, edx, y, ey, x, ex, w, h }
+  }
+
+  public calibrate(region: Box) {
+    return new Box({
+      left: this.left + (region.left * this.width),
+      top: this.top + (region.top * this.height),
+      right: this.right + (region.right * this.width),
+      bottom: this.bottom + (region.bottom * this.height)
+    }).toSquare().round()
+  }
+}
\ No newline at end of file

src/classes/Dimensions.ts

+import { isValidNumber } from '../utils';
+
+export interface IDimensions {
+  width: number
+  height: number
+}
+
+export class Dimensions implements IDimensions {
+
+  private _width: number
+  private _height: number
+
+  constructor(width: number, height: number) {
+    if (!isValidNumber(width) || !isValidNumber(height)) {
+      throw new Error(`Dimensions.constructor - expected width and height to be valid numbers, instead have ${JSON.stringify({ width, height })}`)
+    }
+
+    this._width = width
+    this._height = height
+  }
+
+  public get width(): number { return this._width }
+  public get height(): number { return this._height }
+
+  public reverse(): Dimensions {
+    return new Dimensions(1 / this.width, 1 / this.height)
+  }
+}
\ No newline at end of file

src/classes/FaceDetection.ts

-import { Box, IDimensions, ObjectDetection, Rect } from 'tfjs-image-recognition-base';
+import { Box } from './Box';
+import { IDimensions } from './Dimensions';
+import { ObjectDetection } from './ObjectDetection';
+import { Rect } from './Rect';
 
 export interface IFaceDetecion {
   score: number

src/classes/FaceLandmarks.ts

-import { Box, Dimensions, getCenterPoint, IBoundingBox, IDimensions, IRect, Point, Rect } from 'tfjs-image-recognition-base';
-
-import { minBbox } from '../minBbox';
+import { minBbox } from '../ops';
+import { getCenterPoint } from '../utils';
+import { IBoundingBox } from './BoundingBox';
+import { Box } from './Box';
+import { Dimensions, IDimensions } from './Dimensions';
 import { FaceDetection } from './FaceDetection';
+import { Point } from './Point';
+import { IRect, Rect } from './Rect';
 
 // face alignment constants
 const relX = 0.5

src/classes/FaceLandmarks5.ts

-import { getCenterPoint, Point } from 'tfjs-image-recognition-base';
-
+import { getCenterPoint } from '../utils';
 import { FaceLandmarks } from './FaceLandmarks';
+import { Point } from './Point';
+
 
 export class FaceLandmarks5 extends FaceLandmarks {
 

src/classes/FaceLandmarks68.ts

-import { getCenterPoint, Point } from 'tfjs-image-recognition-base';
-
-import { FaceLandmarks } from '../classes/FaceLandmarks';
+import { getCenterPoint } from '../utils';
+import { FaceLandmarks } from './FaceLandmarks';
+import { Point } from './Point';
 
 export class FaceLandmarks68 extends FaceLandmarks {
   public getJawOutline(): Point[] {

src/classes/FaceMatch.ts

-import { round } from 'tfjs-image-recognition-base';
+import { round } from '../utils';
 
 export interface IFaceMatch {
   label: string

src/classes/LabeledBox.ts

+import { isValidNumber } from '../utils';
+import { IBoundingBox } from './BoundingBox';
+import { Box } from './Box';
+import { IRect } from './Rect';
+
+export class LabeledBox extends Box<LabeledBox> {
+
+  public static assertIsValidLabeledBox(box: any, callee: string) {
+    Box.assertIsValidBox(box, callee)
+
+    if (!isValidNumber(box.label)) {
+      throw new Error(`${callee} - expected property label (${box.label}) to be a number`)
+    }
+  }
+
+  private _label: number
+
+  constructor(box: IBoundingBox | IRect | any, label: number) {
+    super(box)
+    this._label = label
+  }
+
+  public get label(): number { return this._label }
+
+}
\ No newline at end of file

src/classes/ObjectDetection.ts

+import { Box } from './Box';
+import { Dimensions, IDimensions } from './Dimensions';
+import { IRect, Rect } from './Rect';
+
+export class ObjectDetection {
+  private _score: number
+  private _classScore: number
+  private _className: string
+  private _box: Rect
+  private _imageDims: Dimensions
+
+  constructor(
+    score: number,
+    classScore: number,
+    className: string,
+    relativeBox: IRect,
+    imageDims: IDimensions
+  ) {
+    this._imageDims = new Dimensions(imageDims.width, imageDims.height)
+    this._score = score
+    this._classScore = classScore
+    this._className = className
+    this._box = new Box(relativeBox).rescale(this._imageDims)
+  }
+
+  public get score(): number { return this._score }
+  public get classScore(): number { return this._classScore }
+  public get className(): string { return this._className }
+  public get box(): Box { return this._box }
+  public get imageDims(): Dimensions { return this._imageDims }
+  public get imageWidth(): number { return this.imageDims.width }
+  public get imageHeight(): number { return this.imageDims.height }
+  public get relativeBox(): Box { return new Box(this._box).rescale(this.imageDims.reverse()) }
+
+  public forSize(width: number, height: number): ObjectDetection {
+    return new ObjectDetection(
+      this.score,
+      this.classScore,
+      this.className,
+      this.relativeBox,
+      { width, height}
+    )
+  }
+}
\ No newline at end of file

src/classes/Point.ts

+export interface IPoint {
+  x: number
+  y: number
+}
+
+export class Point implements IPoint {
+  private _x: number
+  private _y: number
+
+  constructor(x: number, y: number) {
+    this._x = x
+    this._y = y
+  }
+
+  get x(): number { return this._x }
+  get y(): number { return this._y }
+
+  public add(pt: IPoint): Point {
+    return new Point(this.x + pt.x, this.y + pt.y)
+  }
+
+  public sub(pt: IPoint): Point {
+    return new Point(this.x - pt.x, this.y - pt.y)
+  }
+
+  public mul(pt: IPoint): Point {
+    return new Point(this.x * pt.x, this.y * pt.y)
+  }
+
+  public div(pt: IPoint): Point {
+    return new Point(this.x / pt.x, this.y / pt.y)
+  }
+
+  public abs(): Point {
+    return new Point(Math.abs(this.x), Math.abs(this.y))
+  }
+
+  public magnitude(): number {
+    return Math.sqrt(Math.pow(this.x, 2) + Math.pow(this.y, 2))
+  }
+
+  public floor(): Point {
+    return new Point(Math.floor(this.x), Math.floor(this.y))
+  }
+}
\ No newline at end of file

src/classes/PredictedBox.ts

+import { isValidProbablitiy } from '../utils';
+import { IBoundingBox } from './BoundingBox';
+import { LabeledBox } from './LabeledBox';
+import { IRect } from './Rect';
+
+export class PredictedBox extends LabeledBox {
+
+  public static assertIsValidPredictedBox(box: any, callee: string) {
+    LabeledBox.assertIsValidLabeledBox(box, callee)
+
+    if (
+      !isValidProbablitiy(box.score)
+      || !isValidProbablitiy(box.classScore)
+    ) {
+      throw new Error(`${callee} - expected properties score (${box.score}) and (${box.classScore}) to be a number between [0, 1]`)
+    }
+  }
+
+  private _score: number
+  private _classScore: number
+
+  constructor(box: IBoundingBox | IRect | any, label: number, score: number, classScore: number) {
+    super(box, label)
+    this._score = score
+    this._classScore = classScore
+  }
+
+  public get score(): number { return this._score }
+  public get classScore(): number { return this._classScore }
+
+}
\ No newline at end of file

src/classes/Rect.ts

+import { Box } from './Box';
+
+export interface IRect {
+  x: number
+  y: number
+  width: number
+  height: number
+}
+
+export class Rect extends Box<Rect> implements IRect {
+  constructor(x: number, y: number, width: number, height: number, allowNegativeDimensions: boolean = false) {
+    super({ x, y, width, height }, allowNegativeDimensions)
+  }
+}
\ No newline at end of file

src/classes/index.ts

+export * from './BoundingBox'
+export * from './Box'
+export * from './Dimensions'
 export * from './FaceDetection';
 export * from './FaceLandmarks';
 export * from './FaceLandmarks5';
 export * from './FaceLandmarks68';
 export * from './FaceMatch';
+export * from './LabeledBox'
 export * from './LabeledFaceDescriptors';
+export * from './ObjectDetection'
+export * from './Point'
+export * from './PredictedBox'
+export * from './Rect'
\ No newline at end of file

src/common/convLayer.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ConvParams } from './types';
+
+export function convLayer(
+  x: tf.Tensor4D,
+  params: ConvParams,
+  padding: 'valid' | 'same' = 'same',
+  withRelu: boolean = false
+): tf.Tensor4D {
+  return tf.tidy(() => {
+    const out = tf.add(
+      tf.conv2d(x, params.filters, [1, 1], padding),
+      params.bias
+    ) as tf.Tensor4D
+
+    return withRelu ? tf.relu(out) : out
+  })
+}
\ No newline at end of file

src/common/depthwiseSeparableConv.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { SeparableConvParams } from './types';
 
 export function depthwiseSeparableConv(
   x: tf.Tensor4D,
-  params: TfjsImageRecognitionBase.SeparableConvParams,
+  params: SeparableConvParams,
   stride: [number, number]
 ): tf.Tensor4D {
   return tf.tidy(() => {

src/common/disposeUnusedWeightTensors.ts

+import { ParamMapping } from './types';
+
+export function disposeUnusedWeightTensors(weightMap: any, paramMappings: ParamMapping[]) {
+  Object.keys(weightMap).forEach(path => {
+    if (!paramMappings.some(pm => pm.originalPath === path)) {
+      weightMap[path].dispose()
+    }
+  })
+}

src/common/extractConvParamsFactory.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ConvParams, ExtractWeightsFunction, ParamMapping } from './types';
+
+export function extractConvParamsFactory(
+  extractWeights: ExtractWeightsFunction,
+  paramMappings: ParamMapping[]
+) {
+
+  return function(
+    channelsIn: number,
+    channelsOut: number,
+    filterSize: number,
+    mappedPrefix: string
+  ): ConvParams {
+
+    const filters = tf.tensor4d(
+      extractWeights(channelsIn * channelsOut * filterSize * filterSize),
+      [filterSize, filterSize, channelsIn, channelsOut]
+    )
+    const bias = tf.tensor1d(extractWeights(channelsOut))
+
+    paramMappings.push(
+      { paramPath: `${mappedPrefix}/filters` },
+      { paramPath: `${mappedPrefix}/bias` }
+    )
+
+    return { filters, bias }
+  }
+
+}

src/common/extractFCParamsFactory.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ExtractWeightsFunction, FCParams, ParamMapping } from './types';
+
+
+export function extractFCParamsFactory(
+  extractWeights: ExtractWeightsFunction,
+  paramMappings: ParamMapping[]
+) {
+
+  return function(
+    channelsIn: number,
+    channelsOut: number,
+    mappedPrefix: string
+  ): FCParams {
+
+    const fc_weights = tf.tensor2d(extractWeights(channelsIn * channelsOut), [channelsIn, channelsOut])
+    const fc_bias = tf.tensor1d(extractWeights(channelsOut))
+
+    paramMappings.push(
+      { paramPath: `${mappedPrefix}/weights` },
+      { paramPath: `${mappedPrefix}/bias` }
+    )
+
+    return {
+      weights: fc_weights,
+      bias: fc_bias
+    }
+  }
+
+}

src/common/extractSeparableConvParamsFactory.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ExtractWeightsFunction, ParamMapping, SeparableConvParams } from './types';
+
+export function extractSeparableConvParamsFactory(
+  extractWeights: ExtractWeightsFunction,
+  paramMappings: ParamMapping[]
+) {
+
+  return function(channelsIn: number, channelsOut: number, mappedPrefix: string): SeparableConvParams {
+    const depthwise_filter = tf.tensor4d(extractWeights(3 * 3 * channelsIn), [3, 3, channelsIn, 1])
+    const pointwise_filter = tf.tensor4d(extractWeights(channelsIn * channelsOut), [1, 1, channelsIn, channelsOut])
+    const bias = tf.tensor1d(extractWeights(channelsOut))
+
+    paramMappings.push(
+      { paramPath: `${mappedPrefix}/depthwise_filter` },
+      { paramPath: `${mappedPrefix}/pointwise_filter` },
+      { paramPath: `${mappedPrefix}/bias` }
+    )
+
+    return new SeparableConvParams(
+      depthwise_filter,
+      pointwise_filter,
+      bias
+    )
+  }
+
+}
+
+export function loadSeparableConvParamsFactory(
+  extractWeightEntry: <T>(originalPath: string, paramRank: number) => T
+) {
+
+  return function (prefix: string): SeparableConvParams {
+    const depthwise_filter = extractWeightEntry<tf.Tensor4D>(`${prefix}/depthwise_filter`, 4)
+    const pointwise_filter = extractWeightEntry<tf.Tensor4D>(`${prefix}/pointwise_filter`, 4)
+    const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
+
+    return new SeparableConvParams(
+      depthwise_filter,
+      pointwise_filter,
+      bias
+    )
+  }
+
+}

src/common/extractWeightEntryFactory.ts

+import { isTensor } from '../utils';
+import { ParamMapping } from './types';
+
+export function extractWeightEntryFactory(weightMap: any, paramMappings: ParamMapping[]) {
+
+  return function<T> (originalPath: string, paramRank: number, mappedPath?: string): T {
+    const tensor = weightMap[originalPath]
+
+    if (!isTensor(tensor, paramRank)) {
+      throw new Error(`expected weightMap[${originalPath}] to be a Tensor${paramRank}D, instead have ${tensor}`)
+    }
+
+    paramMappings.push(
+      { originalPath, paramPath: mappedPath || originalPath }
+    )
+
+    return tensor
+  }
+
+}

src/common/extractWeightsFactory.ts

+export function extractWeightsFactory(weights: Float32Array) {
+  let remainingWeights = weights
+
+  function extractWeights(numWeights: number): Float32Array {
+    const ret = remainingWeights.slice(0, numWeights)
+    remainingWeights = remainingWeights.slice(numWeights)
+    return ret
+  }
+
+  function getRemainingWeights(): Float32Array {
+    return remainingWeights
+  }
+
+  return {
+    extractWeights,
+    getRemainingWeights
+  }
+}
\ No newline at end of file

src/common/fullyConnectedLayer.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { FCParams } from './types';
 
 export function fullyConnectedLayer(
   x: tf.Tensor2D,
-  params: TfjsImageRecognitionBase.FCParams
+  params: FCParams
 ): tf.Tensor2D {
   return tf.tidy(() =>
     tf.add(

src/common/getModelUris.ts

+export function getModelUris(uri: string | undefined, defaultModelName: string) {
+  const defaultManifestFilename = `${defaultModelName}-weights_manifest.json`
+
+  if (!uri) {
+    return {
+      modelBaseUri: '',
+      manifestUri: defaultManifestFilename
+    }
+  }
+
+  if (uri === '/') {
+    return {
+      modelBaseUri: '/',
+      manifestUri: `/${defaultManifestFilename}`
+    }
+  }
+  const protocol = uri.startsWith('http://') ? 'http://' : uri.startsWith('https://') ? 'https://' : '';
+  uri = uri.replace(protocol, '');
+
+  const parts = uri.split('/').filter(s => s)
+
+  const manifestFile = uri.endsWith('.json')
+    ? parts[parts.length - 1]
+    : defaultManifestFilename
+
+  let modelBaseUri = protocol + (uri.endsWith('.json') ? parts.slice(0, parts.length - 1) : parts).join('/')
+  modelBaseUri = uri.startsWith('/') ? `/${modelBaseUri}` : modelBaseUri
+
+  return {
+    modelBaseUri,
+    manifestUri: modelBaseUri === '/' ? `/${manifestFile}` : `${modelBaseUri}/${manifestFile}`
+  }
+}
\ No newline at end of file

src/common/index.ts

+export * from './convLayer'
+export * from './depthwiseSeparableConv'
+export * from './disposeUnusedWeightTensors'
+export * from './extractConvParamsFactory'
+export * from './extractFCParamsFactory'
+export * from './extractSeparableConvParamsFactory'
+export * from './extractWeightEntryFactory'
+export * from './extractWeightsFactory'
+export * from './getModelUris'
+export * from './types'
\ No newline at end of file

src/common/loadConvParamsFactory.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { ConvParams } from './types';
 
 export function loadConvParamsFactory(extractWeightEntry: <T>(originalPath: string, paramRank: number) => T) {
-  return function(prefix: string): TfjsImageRecognitionBase.ConvParams {
+  return function(prefix: string): ConvParams {
     const filters = extractWeightEntry<tf.Tensor4D>(`${prefix}/filters`, 4)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
 

src/common/types.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+export type ExtractWeightsFunction = (numWeights: number) => Float32Array
+
+export type ParamMapping = {
+  originalPath?: string
+  paramPath: string
+}
+
+export type ConvParams = {
+  filters: tf.Tensor4D
+  bias: tf.Tensor1D
+}
+
+export type FCParams = {
+  weights: tf.Tensor2D
+  bias: tf.Tensor1D
+}
+
+export class SeparableConvParams {
+  constructor(
+    public depthwise_filter: tf.Tensor4D,
+    public pointwise_filter: tf.Tensor4D,
+    public bias: tf.Tensor1D
+  ) {}
+}
\ No newline at end of file

src/dom/NetInput.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { Dimensions } from '../classes/Dimensions';
+import { env } from '../env';
+import { padToSquare } from '../ops/padToSquare';
+import { computeReshapedDimensions, isTensor3D, isTensor4D, range } from '../utils';
+import { createCanvasFromMedia } from './createCanvas';
+import { imageToSquare } from './imageToSquare';
+import { TResolvedNetInput } from './types';
+
+export class NetInput {
+  private _imageTensors: Array<tf.Tensor3D | tf.Tensor4D> = []
+  private _canvases: HTMLCanvasElement[] = []
+  private _batchSize: number
+  private _treatAsBatchInput: boolean = false
+
+  private _inputDimensions: number[][] = []
+  private _inputSize: number
+
+  constructor(
+    inputs: Array<TResolvedNetInput>,
+    treatAsBatchInput: boolean = false
+  ) {
+    if (!Array.isArray(inputs)) {
+      throw new Error(`NetInput.constructor - expected inputs to be an Array of TResolvedNetInput or to be instanceof tf.Tensor4D, instead have ${inputs}`)
+    }
+
+    this._treatAsBatchInput = treatAsBatchInput
+    this._batchSize = inputs.length
+
+    inputs.forEach((input, idx) => {
+
+      if (isTensor3D(input)) {
+        this._imageTensors[idx] = input
+        this._inputDimensions[idx] = input.shape
+        return
+      }
+
+      if (isTensor4D(input)) {
+        const batchSize = input.shape[0]
+        if (batchSize !== 1) {
+          throw new Error(`NetInput - tf.Tensor4D with batchSize ${batchSize} passed, but not supported in input array`)
+        }
+
+        this._imageTensors[idx] = input
+        this._inputDimensions[idx] = input.shape.slice(1)
+        return
+      }
+
+      const canvas = input instanceof env.getEnv().Canvas ? input : createCanvasFromMedia(input)
+      this._canvases[idx] = canvas
+      this._inputDimensions[idx] = [canvas.height, canvas.width, 3]
+    })
+  }
+
+  public get imageTensors(): Array<tf.Tensor3D | tf.Tensor4D> {
+    return this._imageTensors
+  }
+
+  public get canvases(): HTMLCanvasElement[] {
+    return this._canvases
+  }
+
+  public get isBatchInput(): boolean {
+    return this.batchSize > 1 || this._treatAsBatchInput
+  }
+
+  public get batchSize(): number {
+    return this._batchSize
+  }
+
+  public get inputDimensions(): number[][] {
+    return this._inputDimensions
+  }
+
+  public get inputSize(): number | undefined {
+    return this._inputSize
+  }
+
+  public get reshapedInputDimensions(): Dimensions[] {
+    return range(this.batchSize, 0, 1).map(
+      (_, batchIdx) => this.getReshapedInputDimensions(batchIdx)
+    )
+  }
+
+  public getInput(batchIdx: number): tf.Tensor3D  | tf.Tensor4D | HTMLCanvasElement {
+    return this.canvases[batchIdx] || this.imageTensors[batchIdx]
+  }
+
+  public getInputDimensions(batchIdx: number): number[] {
+    return this._inputDimensions[batchIdx]
+  }
+
+  public getInputHeight(batchIdx: number): number {
+    return this._inputDimensions[batchIdx][0]
+  }
+
+  public getInputWidth(batchIdx: number): number {
+    return this._inputDimensions[batchIdx][1]
+  }
+
+  public getReshapedInputDimensions(batchIdx: number): Dimensions {
+    if (typeof this.inputSize !== 'number') {
+      throw new Error('getReshapedInputDimensions - inputSize not set, toBatchTensor has not been called yet')
+    }
+
+    const width = this.getInputWidth(batchIdx)
+    const height = this.getInputHeight(batchIdx)
+    return computeReshapedDimensions({ width, height }, this.inputSize)
+  }
+
+  /**
+   * Create a batch tensor from all input canvases and tensors
+   * with size [batchSize, inputSize, inputSize, 3].
+   *
+   * @param inputSize Height and width of the tensor.
+   * @param isCenterImage (optional, default: false) If true, add an equal amount of padding on
+   * both sides of the minor dimension oof the image.
+   * @returns The batch tensor.
+   */
+  public toBatchTensor(inputSize: number, isCenterInputs: boolean = true): tf.Tensor4D {
+
+    this._inputSize = inputSize
+
+    return tf.tidy(() => {
+
+      const inputTensors = range(this.batchSize, 0, 1).map(batchIdx => {
+        const input = this.getInput(batchIdx)
+
+        if (input instanceof tf.Tensor) {
+          let imgTensor = isTensor4D(input) ? input : input.expandDims<tf.Rank.R4>()
+          imgTensor = padToSquare(imgTensor, isCenterInputs)
+
+          if (imgTensor.shape[1] !== inputSize || imgTensor.shape[2] !== inputSize) {
+            imgTensor = tf.image.resizeBilinear(imgTensor, [inputSize, inputSize])
+          }
+
+          return imgTensor.as3D(inputSize, inputSize, 3)
+        }
+
+        if (input instanceof env.getEnv().Canvas) {
+          return tf.browser.fromPixels(imageToSquare(input, inputSize, isCenterInputs))
+        }
+
+        throw new Error(`toBatchTensor - at batchIdx ${batchIdx}, expected input to be instanceof tf.Tensor or instanceof HTMLCanvasElement, instead have ${input}`)
+      })
+
+      const batchTensor = tf.stack(inputTensors.map(t => t.toFloat())).as4D(this.batchSize, inputSize, inputSize, 3)
+
+      return batchTensor
+    })
+  }
+}
\ No newline at end of file

src/dom/awaitMediaLoaded.ts

+import { env } from '../env';
+import { isMediaLoaded } from './isMediaLoaded';
+
+export function awaitMediaLoaded(media: HTMLImageElement | HTMLVideoElement | HTMLCanvasElement) {
+
+  return new Promise((resolve, reject) => {
+    if (media instanceof env.getEnv().Canvas || isMediaLoaded(media)) {
+      return resolve()
+    }
+
+    function onLoad(e: Event) {
+      if (!e.currentTarget) return
+      e.currentTarget.removeEventListener('load', onLoad)
+      e.currentTarget.removeEventListener('error', onError)
+      resolve(e)
+    }
+
+    function onError(e: Event) {
+      if (!e.currentTarget) return
+      e.currentTarget.removeEventListener('load', onLoad)
+      e.currentTarget.removeEventListener('error', onError)
+      reject(e)
+    }
+
+    media.addEventListener('load', onLoad)
+    media.addEventListener('error', onError)
+  })
+}
\ No newline at end of file

src/dom/bufferToImage.ts

+import { env } from '../env';
+
+export function bufferToImage(buf: Blob): Promise<HTMLImageElement> {
+  return new Promise((resolve, reject) => {
+    if (!(buf instanceof Blob)) {
+      return reject('bufferToImage - expected buf to be of type: Blob')
+    }
+
+    const reader = new FileReader()
+    reader.onload = () => {
+      if (typeof reader.result !== 'string') {
+        return reject('bufferToImage - expected reader.result to be a string, in onload')
+      }
+
+      const img = env.getEnv().createImageElement()
+      img.onload = () => resolve(img)
+      img.onerror = reject
+      img.src = reader.result
+    }
+    reader.onerror = reject
+    reader.readAsDataURL(buf)
+  })
+}
\ No newline at end of file

src/dom/createCanvas.ts

+import { IDimensions } from '../classes/Dimensions';
+import { env } from '../env';
+import { getContext2dOrThrow } from './getContext2dOrThrow';
+import { getMediaDimensions } from './getMediaDimensions';
+import { isMediaLoaded } from './isMediaLoaded';
+
+export function createCanvas({ width, height }: IDimensions): HTMLCanvasElement {
+
+  const { createCanvasElement } = env.getEnv()
+  const canvas = createCanvasElement()
+  canvas.width = width
+  canvas.height = height
+  return canvas
+}
+
+export function createCanvasFromMedia(media: HTMLImageElement | HTMLVideoElement | ImageData, dims?: IDimensions): HTMLCanvasElement {
+
+  const { ImageData } = env.getEnv()
+
+  if (!(media instanceof ImageData) && !isMediaLoaded(media)) {
+    throw new Error('createCanvasFromMedia - media has not finished loading yet')
+  }
+
+  const { width, height } = dims || getMediaDimensions(media)
+  const canvas = createCanvas({ width, height })
+
+  if (media instanceof ImageData) {
+    getContext2dOrThrow(canvas).putImageData(media, 0, 0)
+  } else {
+    getContext2dOrThrow(canvas).drawImage(media, 0, 0, width, height)
+  }
+  return canvas
+}
\ No newline at end of file

src/dom/extractFaceTensors.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { isTensor4D, Rect, isTensor3D } from 'tfjs-image-recognition-base';
 
+import { Rect } from '../classes';
 import { FaceDetection } from '../classes/FaceDetection';
+import { isTensor3D, isTensor4D } from '../utils';
 
 /**
  * Extracts the tensors of the image regions containing the detected faces.

src/dom/extractFaces.ts

-import {
-  createCanvas,
-  env,
-  getContext2dOrThrow,
-  imageTensorToCanvas,
-  Rect,
-  TNetInput,
-  toNetInput,
-} from 'tfjs-image-recognition-base';
-
 import { FaceDetection } from '../classes/FaceDetection';
+import { Rect } from '../classes/Rect';
+import { env } from '../env';
+import { createCanvas } from './createCanvas';
+import { getContext2dOrThrow } from './getContext2dOrThrow';
+import { imageTensorToCanvas } from './imageTensorToCanvas';
+import { toNetInput } from './toNetInput';
+import { TNetInput } from './types';
 
 /**
  * Extracts the image regions containing the detected faces.

src/dom/fetchImage.ts

+import { bufferToImage } from './bufferToImage';
+import { fetchOrThrow } from './fetchOrThrow';
+
+export async function fetchImage(uri: string): Promise<HTMLImageElement> {
+  const res = await fetchOrThrow(uri)
+  const blob = await (res).blob()
+
+  if (!blob.type.startsWith('image/')) {
+    throw new Error(`fetchImage - expected blob type to be of type image/*, instead have: ${blob.type}, for url: ${res.url}`)
+  }
+  return bufferToImage(blob)
+}

src/dom/fetchJson.ts

+import { fetchOrThrow } from './fetchOrThrow';
+
+export async function fetchJson<T>(uri: string): Promise<T> {
+  return (await fetchOrThrow(uri)).json()
+}

src/dom/fetchNetWeights.ts

+import { fetchOrThrow } from './fetchOrThrow';
+
+export async function fetchNetWeights(uri: string): Promise<Float32Array> {
+  return new Float32Array(await (await fetchOrThrow(uri)).arrayBuffer())
+}

src/dom/fetchOrThrow.ts

+import { env } from '../env';
+
+export async function fetchOrThrow(
+  url: string,
+  init?: RequestInit
+): Promise<Response> {
+
+  const fetch = env.getEnv().fetch
+  const res = await fetch(url, init)
+  if (!(res.status < 400)) {
+    throw new Error(`failed to fetch: (${res.status}) ${res.statusText}, from url: ${res.url}`)
+  }
+  return res
+}
\ No newline at end of file

src/dom/getContext2dOrThrow.ts

+import { env } from '../env';
+import { resolveInput } from './resolveInput';
+
+export function getContext2dOrThrow(canvasArg: string | HTMLCanvasElement | CanvasRenderingContext2D): CanvasRenderingContext2D {
+
+  const { Canvas, CanvasRenderingContext2D } = env.getEnv()
+
+  if (canvasArg instanceof CanvasRenderingContext2D) {
+    return canvasArg
+  }
+
+  const canvas = resolveInput(canvasArg)
+
+  if (!(canvas instanceof Canvas)) {
+    throw new Error('resolveContext2d - expected canvas to be of instance of Canvas')
+  }
+
+  const ctx = canvas.getContext('2d')
+  if (!ctx) {
+    throw new Error('resolveContext2d - canvas 2d context is null')
+  }
+
+  return ctx
+}
\ No newline at end of file

src/dom/getMediaDimensions.ts

+import { Dimensions, IDimensions } from '../classes/Dimensions';
+import { env } from '../env';
+
+export function getMediaDimensions(input: HTMLImageElement | HTMLCanvasElement | HTMLVideoElement | IDimensions): Dimensions {
+
+  const { Image, Video } = env.getEnv()
+
+  if (input instanceof Image) {
+    return new Dimensions(input.naturalWidth, input.naturalHeight)
+  }
+  if (input instanceof Video) {
+    return new Dimensions(input.videoWidth, input.videoHeight)
+  }
+  return new Dimensions(input.width, input.height)
+}

src/dom/imageTensorToCanvas.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { env } from '../env';
+import { isTensor4D } from '../utils';
+
+export async function imageTensorToCanvas(
+  imgTensor: tf.Tensor,
+  canvas?: HTMLCanvasElement
+): Promise<HTMLCanvasElement> {
+
+  const targetCanvas = canvas || env.getEnv().createCanvasElement()
+
+  const [height, width, numChannels] = imgTensor.shape.slice(isTensor4D(imgTensor) ? 1 : 0)
+  const imgTensor3D = tf.tidy(() => imgTensor.as3D(height, width, numChannels).toInt())
+  await tf.browser.toPixels(imgTensor3D, targetCanvas)
+
+  imgTensor3D.dispose()
+
+  return targetCanvas
+}
\ No newline at end of file

src/dom/imageToSquare.ts

+import { env } from '../env';
+import { createCanvas, createCanvasFromMedia } from './createCanvas';
+import { getContext2dOrThrow } from './getContext2dOrThrow';
+import { getMediaDimensions } from './getMediaDimensions';
+
+export function imageToSquare(input: HTMLImageElement | HTMLCanvasElement, inputSize: number, centerImage: boolean = false) {
+
+  const { Image, Canvas } = env.getEnv()
+
+  if (!(input instanceof Image || input instanceof Canvas)) {
+    throw new Error('imageToSquare - expected arg0 to be HTMLImageElement | HTMLCanvasElement')
+  }
+
+  const dims = getMediaDimensions(input)
+  const scale = inputSize / Math.max(dims.height, dims.width)
+  const width = scale * dims.width
+  const height = scale * dims.height
+
+  const targetCanvas = createCanvas({ width: inputSize, height: inputSize })
+  const inputCanvas = input instanceof Canvas ? input : createCanvasFromMedia(input)
+
+  const offset = Math.abs(width - height) / 2
+  const dx = centerImage && width < height ? offset : 0
+  const dy = centerImage && height < width ? offset : 0
+  getContext2dOrThrow(targetCanvas).drawImage(inputCanvas, dx, dy, width, height)
+
+  return targetCanvas
+}
\ No newline at end of file

src/dom/index.ts

+export * from './awaitMediaLoaded'
+export * from './bufferToImage'
+export * from './createCanvas'
 export * from './extractFaces'
 export * from './extractFaceTensors'
+export * from './fetchImage'
+export * from './fetchJson'
+export * from './fetchNetWeights'
+export * from './fetchOrThrow'
+export * from './getContext2dOrThrow'
+export * from './getMediaDimensions'
+export * from './imageTensorToCanvas'
+export * from './imageToSquare'
+export * from './isMediaElement'
+export * from './isMediaLoaded'
+export * from './loadWeightMap'
+export * from './matchDimensions'
+export * from './NetInput'
+export * from './resolveInput'
+export * from './toNetInput'
+export * from './types'
\ No newline at end of file

src/dom/isMediaElement.ts

+import { env } from '../env';
+
+export function isMediaElement(input: any) {
+
+  const { Image, Canvas, Video } = env.getEnv()
+
+  return input instanceof Image
+    || input instanceof Canvas
+    || input instanceof Video
+}
\ No newline at end of file

src/dom/isMediaLoaded.ts

+import { env } from '../env';
+
+export function isMediaLoaded(media: HTMLImageElement | HTMLVideoElement) : boolean {
+
+  const { Image, Video } = env.getEnv()
+
+  return (media instanceof Image && media.complete)
+    || (media instanceof Video && media.readyState >= 3)
+}

src/dom/loadWeightMap.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { getModelUris } from '../common/getModelUris';
+import { fetchJson } from './fetchJson';
+
+export async function loadWeightMap(
+  uri: string | undefined,
+  defaultModelName: string,
+): Promise<tf.NamedTensorMap> {
+  const { manifestUri, modelBaseUri } = getModelUris(uri, defaultModelName)
+
+  const manifest = await fetchJson<tf.io.WeightsManifestConfig>(manifestUri)
+
+  return tf.io.loadWeights(manifest, modelBaseUri)
+}
\ No newline at end of file

src/dom/matchDimensions.ts

+import { IDimensions } from '../classes';
+import { getMediaDimensions } from './getMediaDimensions';
+
+export function matchDimensions(input: IDimensions, reference: IDimensions, useMediaDimensions: boolean = false) {
+  const { width, height } = useMediaDimensions
+    ? getMediaDimensions(reference)
+    : reference
+  input.width = width
+  input.height = height
+  return { width, height }
+}
\ No newline at end of file

src/dom/resolveInput.ts

+import { env } from '../env';
+
+export function resolveInput(arg: string | any) {
+  if (!env.isNodejs() && typeof arg === 'string') {
+    return document.getElementById(arg)
+  }
+  return arg
+}
\ No newline at end of file

src/dom/toNetInput.ts

+import { isTensor3D, isTensor4D } from '../utils';
+import { awaitMediaLoaded } from './awaitMediaLoaded';
+import { isMediaElement } from './isMediaElement';
+import { NetInput } from './NetInput';
+import { resolveInput } from './resolveInput';
+import { TNetInput } from './types';
+
+/**
+ * Validates the input to make sure, they are valid net inputs and awaits all media elements
+ * to be finished loading.
+ *
+ * @param input The input, which can be a media element or an array of different media elements.
+ * @returns A NetInput instance, which can be passed into one of the neural networks.
+ */
+export async function toNetInput(inputs: TNetInput): Promise<NetInput> {
+  if (inputs instanceof NetInput) {
+    return inputs
+  }
+
+  let inputArgArray = Array.isArray(inputs)
+      ? inputs
+      : [inputs]
+
+  if (!inputArgArray.length) {
+    throw new Error('toNetInput - empty array passed as input')
+  }
+
+  const getIdxHint = (idx: number) => Array.isArray(inputs) ? ` at input index ${idx}:` : ''
+
+  const inputArray = inputArgArray.map(resolveInput)
+
+  inputArray.forEach((input, i) => {
+    if (!isMediaElement(input) && !isTensor3D(input) && !isTensor4D(input)) {
+
+      if (typeof inputArgArray[i] === 'string') {
+        throw new Error(`toNetInput -${getIdxHint(i)} string passed, but could not resolve HTMLElement for element id ${inputArgArray[i]}`)
+      }
+
+      throw new Error(`toNetInput -${getIdxHint(i)} expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id`)
+    }
+
+    if (isTensor4D(input)) {
+      // if tf.Tensor4D is passed in the input array, the batch size has to be 1
+      const batchSize = input.shape[0]
+      if (batchSize !== 1) {
+        throw new Error(`toNetInput -${getIdxHint(i)} tf.Tensor4D with batchSize ${batchSize} passed, but not supported in input array`)
+      }
+    }
+  })
+
+  // wait for all media elements being loaded
+  await Promise.all(
+    inputArray.map(input => isMediaElement(input) && awaitMediaLoaded(input))
+  )
+
+  return new NetInput(inputArray, Array.isArray(inputs))
+}
\ No newline at end of file

src/dom/types.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { NetInput } from './NetInput';
+
+export type TMediaElement = HTMLImageElement | HTMLVideoElement | HTMLCanvasElement
+
+export type TResolvedNetInput = TMediaElement | tf.Tensor3D | tf.Tensor4D
+
+export type TNetInputArg = string | TResolvedNetInput
+
+export type TNetInput = TNetInputArg | Array<TNetInputArg> | NetInput | tf.Tensor4D
\ No newline at end of file

src/draw/DrawBox.ts

+import { Box, IBoundingBox, IRect } from '../classes';
+import { getContext2dOrThrow } from '../dom/getContext2dOrThrow';
+import { AnchorPosition, DrawTextField, DrawTextFieldOptions, IDrawTextFieldOptions } from './DrawTextField';
+
+export interface IDrawBoxOptions {
+  boxColor?: string
+  lineWidth?: number
+  drawLabelOptions?: IDrawTextFieldOptions
+  label?: string
+}
+
+export class DrawBoxOptions {
+  public boxColor: string
+  public lineWidth: number
+  public drawLabelOptions: DrawTextFieldOptions
+  public label?: string
+
+  constructor(options: IDrawBoxOptions = {}) {
+    const { boxColor, lineWidth, label, drawLabelOptions } = options
+    this.boxColor = boxColor || 'rgba(0, 0, 255, 1)'
+    this.lineWidth = lineWidth || 2
+    this.label = label
+
+    const defaultDrawLabelOptions = {
+      anchorPosition: AnchorPosition.BOTTOM_LEFT,
+      backgroundColor: this.boxColor
+    }
+    this.drawLabelOptions = new DrawTextFieldOptions(Object.assign({}, defaultDrawLabelOptions, drawLabelOptions))
+  }
+}
+
+export class DrawBox {
+  public box: Box
+  public options: DrawBoxOptions
+
+  constructor(
+    box: IBoundingBox | IRect,
+    options: IDrawBoxOptions = {}
+  ) {
+    this.box = new Box(box)
+    this.options = new DrawBoxOptions(options)
+  }
+
+  draw(canvasArg: string | HTMLCanvasElement | CanvasRenderingContext2D) {
+    const ctx = getContext2dOrThrow(canvasArg)
+
+    const { boxColor, lineWidth } = this.options
+
+    const { x, y, width, height } = this.box
+    ctx.strokeStyle = boxColor
+    ctx.lineWidth = lineWidth
+    ctx.strokeRect(x, y, width, height)
+
+    const { label } = this.options
+    if (label) {
+      new DrawTextField([label], { x: x - (lineWidth / 2), y }, this.options.drawLabelOptions).draw(canvasArg)
+    }
+  }
+}
\ No newline at end of file

src/draw/DrawFaceLandmarks.ts

-import { getContext2dOrThrow, IPoint } from 'tfjs-image-recognition-base';
-
+import { IPoint } from '../classes';
 import { FaceLandmarks } from '../classes/FaceLandmarks';
 import { FaceLandmarks68 } from '../classes/FaceLandmarks68';
+import { getContext2dOrThrow } from '../dom/getContext2dOrThrow';
 import { WithFaceDetection } from '../factories/WithFaceDetection';
 import { isWithFaceLandmarks, WithFaceLandmarks } from '../factories/WithFaceLandmarks';
 import { drawContour } from './drawContour';

src/draw/DrawTextField.ts

+import { IDimensions, IPoint } from '../classes';
+import { getContext2dOrThrow } from '../dom/getContext2dOrThrow';
+import { resolveInput } from '../dom/resolveInput';
+
+export enum AnchorPosition {
+  TOP_LEFT = 'TOP_LEFT',
+  TOP_RIGHT = 'TOP_RIGHT',
+  BOTTOM_LEFT = 'BOTTOM_LEFT',
+  BOTTOM_RIGHT = 'BOTTOM_RIGHT'
+}
+
+export interface IDrawTextFieldOptions {
+  anchorPosition?: AnchorPosition
+  backgroundColor?: string
+  fontColor?: string
+  fontSize?: number
+  fontStyle?: string
+  padding?: number
+}
+
+export class DrawTextFieldOptions implements IDrawTextFieldOptions {
+  public anchorPosition: AnchorPosition
+  public backgroundColor: string
+  public fontColor: string
+  public fontSize: number
+  public fontStyle: string
+  public padding: number
+
+  constructor(options: IDrawTextFieldOptions = {}) {
+    const { anchorPosition, backgroundColor, fontColor, fontSize, fontStyle, padding } = options
+    this.anchorPosition = anchorPosition || AnchorPosition.TOP_LEFT
+    this.backgroundColor = backgroundColor || 'rgba(0, 0, 0, 0.5)'
+    this.fontColor = fontColor || 'rgba(255, 255, 255, 1)'
+    this.fontSize = fontSize || 14
+    this.fontStyle = fontStyle || 'Georgia'
+    this.padding = padding || 4
+  }
+}
+
+export class DrawTextField {
+  public text: string[]
+  public anchor : IPoint
+  public options: DrawTextFieldOptions
+
+  constructor(
+    text: string | string[] | DrawTextField,
+    anchor: IPoint,
+    options: IDrawTextFieldOptions = {}
+  ) {
+    this.text = typeof text === 'string'
+      ? [text]
+      : (text instanceof DrawTextField ? text.text : text)
+    this.anchor = anchor
+    this.options = new DrawTextFieldOptions(options)
+  }
+
+  measureWidth(ctx: CanvasRenderingContext2D): number {
+    const { padding } = this.options
+    return this.text.map(l => ctx.measureText(l).width).reduce((w0, w1) => w0 < w1 ? w1 : w0, 0) + (2 * padding)
+  }
+
+  measureHeight(): number {
+    const { fontSize, padding } = this.options
+    return this.text.length * fontSize + (2 * padding)
+  }
+
+  getUpperLeft(ctx: CanvasRenderingContext2D, canvasDims?: IDimensions): IPoint {
+    const { anchorPosition } = this.options
+    const isShiftLeft = anchorPosition === AnchorPosition.BOTTOM_RIGHT || anchorPosition === AnchorPosition.TOP_RIGHT
+    const isShiftTop = anchorPosition === AnchorPosition.BOTTOM_LEFT || anchorPosition === AnchorPosition.BOTTOM_RIGHT
+
+    const textFieldWidth = this.measureWidth(ctx)
+    const textFieldHeight = this.measureHeight()
+    const x = (isShiftLeft ? this.anchor.x - textFieldWidth : this.anchor.x)
+    const y = isShiftTop ? this.anchor.y - textFieldHeight : this.anchor.y
+
+    // adjust anchor if text box exceeds canvas borders
+    if (canvasDims) {
+      const { width, height } = canvasDims
+      const newX = Math.max(Math.min(x, width - textFieldWidth), 0)
+      const newY = Math.max(Math.min(y, height - textFieldHeight), 0)
+      return { x: newX, y: newY }
+    }
+    return { x, y }
+  }
+
+  draw(canvasArg: string | HTMLCanvasElement | CanvasRenderingContext2D) {
+    const canvas = resolveInput(canvasArg)
+    const ctx = getContext2dOrThrow(canvas)
+
+    const { backgroundColor, fontColor, fontSize, fontStyle, padding } = this.options
+
+    ctx.font = `${fontSize}px ${fontStyle}`
+    const maxTextWidth = this.measureWidth(ctx)
+    const textHeight = this.measureHeight()
+
+    ctx.fillStyle = backgroundColor
+    const upperLeft = this.getUpperLeft(ctx, canvas)
+    ctx.fillRect(upperLeft.x, upperLeft.y, maxTextWidth, textHeight)
+
+    ctx.fillStyle = fontColor;
+    this.text.forEach((textLine, i) => {
+      const x = padding + upperLeft.x
+      const y = padding + upperLeft.y + ((i + 1) * fontSize)
+      ctx.fillText(textLine, x, y)
+    })
+  }
+}
\ No newline at end of file

src/draw/drawContour.ts

-import { Point } from 'tfjs-image-recognition-base';
+import { Point } from '../classes';
 
 export function drawContour(
   ctx: CanvasRenderingContext2D,

src/draw/drawDetections.ts

-import { Box, draw, IBoundingBox, IRect, round } from 'tfjs-image-recognition-base';
-
+import { Box, IBoundingBox, IRect } from '../classes';
 import { FaceDetection } from '../classes/FaceDetection';
 import { isWithFaceDetection, WithFaceDetection } from '../factories/WithFaceDetection';
+import { round } from '../utils';
+import { DrawBox } from './DrawBox';
 
 export type TDrawDetectionsInput = IRect | IBoundingBox | FaceDetection | WithFaceDetection<{}>
 
@@ -21,6 +22,6 @@ export function drawDetections(
       : (isWithFaceDetection(det) ? det.detection.box : new Box(det))
 
     const label = score ? `${round(score)}` : undefined
-    new draw.DrawBox(box, { label }).draw(canvasArg)
+    new DrawBox(box, { label }).draw(canvasArg)
   })
 }
\ No newline at end of file

src/draw/drawFaceExpressions.ts

-import { draw, IPoint, Point, round } from 'tfjs-image-recognition-base';
-
+import { IPoint, Point } from '../classes';
 import { FaceExpressions } from '../faceExpressionNet';
 import { isWithFaceDetection } from '../factories/WithFaceDetection';
 import { isWithFaceExpressions, WithFaceExpressions } from '../factories/WithFaceExpressions';
+import { round } from '../utils';
+import { DrawTextField } from './DrawTextField';
 
 export type DrawFaceExpressionsInput = FaceExpressions | WithFaceExpressions<{}>
 
@@ -29,7 +30,7 @@ export function drawFaceExpressions(
       ? e.detection.box.bottomLeft
       : (textFieldAnchor || new Point(0, 0))
 
-    const drawTextField = new draw.DrawTextField(
+    const drawTextField = new DrawTextField(
       resultsToDisplay.map(expr => `${expr.expression} (${round(expr.probability)})`),
       anchor
     )

src/draw/index.ts

 export * from './drawContour'
 export * from './drawDetections'
 export * from './drawFaceExpressions'
+export * from './DrawBox'
 export * from './DrawFaceLandmarks'
+export * from './DrawTextField'
\ No newline at end of file

src/env/createBrowserEnv.ts

+import { Environment } from './types';
+
+export function createBrowserEnv(): Environment {
+
+  const fetch = window['fetch'] || function() {
+    throw new Error('fetch - missing fetch implementation for browser environment')
+  }
+
+  const readFile = function() {
+    throw new Error('readFile - filesystem not available for browser environment')
+  }
+
+  return {
+    Canvas: HTMLCanvasElement,
+    CanvasRenderingContext2D: CanvasRenderingContext2D,
+    Image: HTMLImageElement,
+    ImageData: ImageData,
+    Video: HTMLVideoElement,
+    createCanvasElement: () => document.createElement('canvas'),
+    createImageElement: () => document.createElement('img'),
+    fetch,
+    readFile
+  }
+}
\ No newline at end of file

src/env/createFileSystem.ts

+import { FileSystem } from './types';
+
+export function createFileSystem(fs?: any): FileSystem {
+
+  let requireFsError = ''
+
+  if (!fs) {
+    try {
+      fs = require('fs')
+    } catch (err) {
+      requireFsError = err.toString()
+    }
+  }
+
+  const readFile = fs
+    ? function(filePath: string) {
+      return new Promise<Buffer>((res, rej) => {
+        fs.readFile(filePath, function(err: any, buffer: Buffer) {
+          return err ? rej(err) : res(buffer)
+        })
+      })
+    }
+    : function() {
+      throw new Error(`readFile - failed to require fs in nodejs environment with error: ${requireFsError}`)
+    }
+
+  return {
+    readFile
+  }
+}
\ No newline at end of file

src/env/createNodejsEnv.ts

+import { createFileSystem } from './createFileSystem';
+import { Environment } from './types';
+
+export function createNodejsEnv(): Environment {
+
+  const Canvas = global['Canvas'] || global['HTMLCanvasElement']
+  const Image = global['Image'] || global['HTMLImageElement']
+
+  const createCanvasElement = function() {
+    if (Canvas) {
+      return new Canvas()
+    }
+    throw new Error('createCanvasElement - missing Canvas implementation for nodejs environment')
+  }
+
+  const createImageElement = function() {
+    if (Image) {
+      return new Image()
+    }
+    throw new Error('createImageElement - missing Image implementation for nodejs environment')
+  }
+
+  const fetch = global['fetch'] || function() {
+    throw new Error('fetch - missing fetch implementation for nodejs environment')
+  }
+
+  const fileSystem = createFileSystem()
+
+  return {
+    Canvas: Canvas || class {},
+    CanvasRenderingContext2D: global['CanvasRenderingContext2D'] || class {},
+    Image: Image || class {},
+    ImageData: global['ImageData'] || class {},
+    Video: global['HTMLVideoElement'] || class {},
+    createCanvasElement,
+    createImageElement,
+    fetch,
+    ...fileSystem
+  }
+}
\ No newline at end of file

src/env/index.ts

+import { createBrowserEnv } from './createBrowserEnv';
+import { createFileSystem } from './createFileSystem';
+import { createNodejsEnv } from './createNodejsEnv';
+import { isBrowser } from './isBrowser';
+import { isNodejs } from './isNodejs';
+import { Environment } from './types';
+
+let environment: Environment | null
+
+function getEnv(): Environment {
+  if (!environment) {
+    throw new Error('getEnv - environment is not defined, check isNodejs() and isBrowser()')
+  }
+  return environment
+}
+
+function setEnv(env: Environment) {
+  environment = env
+}
+
+function initialize() {
+  // check for isBrowser() first to prevent electron renderer process
+  // to be initialized with wrong environment due to isNodejs() returning true
+  if (isBrowser()) {
+    setEnv(createBrowserEnv())
+  }
+  if (isNodejs()) {
+    setEnv(createNodejsEnv())
+  }
+}
+
+function monkeyPatch(env: Partial<Environment>) {
+  if (!environment) {
+    initialize()
+  }
+
+  if (!environment) {
+    throw new Error('monkeyPatch - environment is not defined, check isNodejs() and isBrowser()')
+  }
+
+  const { Canvas = environment.Canvas, Image = environment.Image } = env
+  environment.Canvas = Canvas
+  environment.Image = Image
+  environment.createCanvasElement = env.createCanvasElement || (() => new Canvas())
+  environment.createImageElement = env.createImageElement || (() => new Image())
+
+  environment.ImageData = env.ImageData || environment.ImageData
+  environment.Video = env.Video || environment.Video
+  environment.fetch = env.fetch || environment.fetch
+  environment.readFile = env.readFile || environment.readFile
+}
+
+export const env = {
+  getEnv,
+  setEnv,
+  initialize,
+  createBrowserEnv,
+  createFileSystem,
+  createNodejsEnv,
+  monkeyPatch,
+  isBrowser,
+  isNodejs
+}
+
+initialize()
+
+export * from './types'

src/env/isBrowser.ts

+export function isBrowser(): boolean {
+  return typeof window === 'object'
+    && typeof document !== 'undefined'
+    && typeof HTMLImageElement !== 'undefined'
+    && typeof HTMLCanvasElement !== 'undefined'
+    && typeof HTMLVideoElement !== 'undefined'
+    && typeof ImageData !== 'undefined'
+    && typeof CanvasRenderingContext2D !== 'undefined'
+}
\ No newline at end of file

src/env/isNodejs.ts

+export function isNodejs(): boolean {
+  return typeof global === 'object'
+    && typeof require === 'function'
+    && typeof module !== 'undefined'
+    // issues with gatsby.js: module.exports is undefined
+    // && !!module.exports
+    && typeof process !== 'undefined' && !!process.version
+}
\ No newline at end of file

src/env/types.ts

+export type FileSystem = {
+  readFile: (filePath: string) => Promise<Buffer>
+}
+
+export type Environment = FileSystem & {
+  Canvas: typeof HTMLCanvasElement
+  CanvasRenderingContext2D: typeof CanvasRenderingContext2D
+  Image: typeof HTMLImageElement
+  ImageData: typeof ImageData
+  Video: typeof HTMLVideoElement
+  createCanvasElement: () => HTMLCanvasElement
+  createImageElement: () => HTMLImageElement
+  fetch: (url: string, init?: RequestInit) => Promise<Response>
+}

src/faceExpressionNet/FaceExpressionNet.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { NetInput, TNetInput, toNetInput } from '../dom';
 import { FaceFeatureExtractor } from '../faceFeatureExtractor/FaceFeatureExtractor';
 import { FaceFeatureExtractorParams } from '../faceFeatureExtractor/types';
 import { FaceProcessor } from '../faceProcessor/FaceProcessor';

src/faceFeatureExtractor/FaceFeatureExtractor.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, normalize, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { NetInput, TNetInput, toNetInput } from '../dom';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { normalize } from '../ops';
 import { denseBlock4 } from './denseBlock';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';

src/faceFeatureExtractor/TinyFaceFeatureExtractor.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, normalize, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { NetInput, TNetInput, toNetInput } from '../dom';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { normalize } from '../ops';
 import { denseBlock3 } from './denseBlock';
 import { extractParamsFromWeigthMapTiny } from './extractParamsFromWeigthMapTiny';
 import { extractParamsTiny } from './extractParamsTiny';

src/faceFeatureExtractor/denseBlock.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { ConvParams, SeparableConvParams } from '../common';
 import { depthwiseSeparableConv } from '../common/depthwiseSeparableConv';
 import { DenseBlock3Params, DenseBlock4Params } from './types';
 
@@ -13,10 +13,10 @@ export function denseBlock3(
     const out1 = tf.relu(
       isFirstLayer
         ? tf.add(
-          tf.conv2d(x, (denseBlockParams.conv0 as TfjsImageRecognitionBase.ConvParams).filters, [2, 2], 'same'),
+          tf.conv2d(x, (denseBlockParams.conv0 as ConvParams).filters, [2, 2], 'same'),
           denseBlockParams.conv0.bias
         )
-        : depthwiseSeparableConv(x, denseBlockParams.conv0 as TfjsImageRecognitionBase.SeparableConvParams, [2, 2])
+        : depthwiseSeparableConv(x, denseBlockParams.conv0 as SeparableConvParams, [2, 2])
     ) as tf.Tensor4D
     const out2 = depthwiseSeparableConv(out1, denseBlockParams.conv1, [1, 1])
 
@@ -37,10 +37,10 @@ export function denseBlock4(
     const out1 = tf.relu(
       isFirstLayer
         ? tf.add(
-          tf.conv2d(x, (denseBlockParams.conv0 as TfjsImageRecognitionBase.ConvParams).filters, isScaleDown ? [2, 2] : [1, 1], 'same'),
+          tf.conv2d(x, (denseBlockParams.conv0 as ConvParams).filters, isScaleDown ? [2, 2] : [1, 1], 'same'),
           denseBlockParams.conv0.bias
         )
-        : depthwiseSeparableConv(x, denseBlockParams.conv0 as TfjsImageRecognitionBase.SeparableConvParams, isScaleDown ? [2, 2] : [1, 1])
+        : depthwiseSeparableConv(x, denseBlockParams.conv0 as SeparableConvParams, isScaleDown ? [2, 2] : [1, 1])
     ) as tf.Tensor4D
     const out2 = depthwiseSeparableConv(out1, denseBlockParams.conv1, [1, 1])
 

src/faceFeatureExtractor/extractParams.ts

-
+import { extractWeightsFactory, ParamMapping } from '../common';
 import { extractorsFactory } from './extractorsFactory';
 import { FaceFeatureExtractorParams } from './types';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
-export function extractParams(weights: Float32Array): { params: FaceFeatureExtractorParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+export function extractParams(weights: Float32Array): { params: FaceFeatureExtractorParams, paramMappings: ParamMapping[] } {
+
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
   const {
     extractDenseBlock4Params

src/faceFeatureExtractor/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { disposeUnusedWeightTensors, ParamMapping } from '../common';
 import { loadParamsFactory } from './loadParamsFactory';
 import { FaceFeatureExtractorParams } from './types';
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: FaceFeatureExtractorParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: FaceFeatureExtractorParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractDenseBlock4Params
@@ -21,7 +21,7 @@ export function extractParamsFromWeigthMap(
     dense3: extractDenseBlock4Params('dense3')
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/faceFeatureExtractor/extractParamsFromWeigthMapTiny.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { disposeUnusedWeightTensors, ParamMapping } from '../common';
 import { loadParamsFactory } from './loadParamsFactory';
 import { TinyFaceFeatureExtractorParams } from './types';
 
 export function extractParamsFromWeigthMapTiny(
   weightMap: tf.NamedTensorMap
-): { params: TinyFaceFeatureExtractorParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: TinyFaceFeatureExtractorParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractDenseBlock3Params
@@ -20,7 +20,7 @@ export function extractParamsFromWeigthMapTiny(
     dense2: extractDenseBlock3Params('dense2')
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/faceFeatureExtractor/extractParamsTiny.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import { extractWeightsFactory, ParamMapping } from '../common';
 import { extractorsFactory } from './extractorsFactory';
 import { TinyFaceFeatureExtractorParams } from './types';
 
 
-export function extractParamsTiny(weights: Float32Array): { params: TinyFaceFeatureExtractorParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+export function extractParamsTiny(weights: Float32Array): { params: TinyFaceFeatureExtractorParams, paramMappings: ParamMapping[] } {
+
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
   const {
     extractDenseBlock3Params

src/faceFeatureExtractor/extractorsFactory.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import {
+  extractConvParamsFactory,
+  extractSeparableConvParamsFactory,
+  ExtractWeightsFunction,
+  ParamMapping,
+} from '../common';
 import { DenseBlock3Params, DenseBlock4Params } from './types';
 
-export function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeightsFunction, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+export function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
 
-  const extractConvParams = TfjsImageRecognitionBase.extractConvParamsFactory(extractWeights, paramMappings)
-  const extractSeparableConvParams = TfjsImageRecognitionBase.extractSeparableConvParamsFactory(extractWeights, paramMappings)
+  const extractConvParams = extractConvParamsFactory(extractWeights, paramMappings)
+  const extractSeparableConvParams = extractSeparableConvParamsFactory(extractWeights, paramMappings)
 
   function extractDenseBlock3Params(channelsIn: number, channelsOut: number, mappedPrefix: string, isFirstLayer: boolean = false): DenseBlock3Params {
 

src/faceFeatureExtractor/loadParamsFactory.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import { extractWeightEntryFactory, loadSeparableConvParamsFactory, ParamMapping } from '../common';
 import { loadConvParamsFactory } from '../common/loadConvParamsFactory';
 import { DenseBlock3Params, DenseBlock4Params } from './types';
 
-export function loadParamsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+export function loadParamsFactory(weightMap: any, paramMappings: ParamMapping[]) {
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
   const extractConvParams = loadConvParamsFactory(extractWeightEntry)
-  const extractSeparableConvParams = TfjsImageRecognitionBase.loadSeparableConvParamsFactory(extractWeightEntry)
+  const extractSeparableConvParams = loadSeparableConvParamsFactory(extractWeightEntry)
 
   function extractDenseBlock3Params(prefix: string, isFirstLayer: boolean = false): DenseBlock3Params {
     const conv0 = isFirstLayer

src/faceFeatureExtractor/types.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, TNetInput, TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { NetInput, TNetInput } from '..';
+import { ConvParams, SeparableConvParams } from '../common';
+import { NeuralNetwork } from '../NeuralNetwork';
 
 export type ConvWithBatchNormParams = BatchNormParams & {
   filter: tf.Tensor4D
@@ -18,13 +21,13 @@ export type SeparableConvWithBatchNormParams = {
 }
 
 export type DenseBlock3Params = {
-  conv0: TfjsImageRecognitionBase.SeparableConvParams | TfjsImageRecognitionBase.ConvParams
-  conv1: TfjsImageRecognitionBase.SeparableConvParams
-  conv2: TfjsImageRecognitionBase.SeparableConvParams
+  conv0: SeparableConvParams | ConvParams
+  conv1: SeparableConvParams
+  conv2: SeparableConvParams
 }
 
 export type DenseBlock4Params = DenseBlock3Params & {
-  conv3: TfjsImageRecognitionBase.SeparableConvParams
+  conv3: SeparableConvParams
 }
 
 export type TinyFaceFeatureExtractorParams = {

src/faceLandmarkNet/FaceLandmark68NetBase.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { IDimensions, isEven, NetInput, Point, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { IDimensions, Point } from '../classes';
 import { FaceLandmarks68 } from '../classes/FaceLandmarks68';
+import { NetInput, TNetInput, toNetInput } from '../dom';
 import { FaceFeatureExtractorParams, TinyFaceFeatureExtractorParams } from '../faceFeatureExtractor/types';
 import { FaceProcessor } from '../faceProcessor/FaceProcessor';
+import { isEven } from '../utils';
 
 export abstract class FaceLandmark68NetBase<
   TExtractorParams extends FaceFeatureExtractorParams | TinyFaceFeatureExtractorParams

src/faceProcessor/FaceProcessor.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork } from 'tfjs-image-recognition-base';
 
 import { fullyConnectedLayer } from '../common/fullyConnectedLayer';
+import { NetInput } from '../dom';
 import {
   FaceFeatureExtractorParams,
   IFaceFeatureExtractor,
   TinyFaceFeatureExtractorParams,
 } from '../faceFeatureExtractor/types';
+import { NeuralNetwork } from '../NeuralNetwork';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';
 import { NetParams } from './types';

src/faceProcessor/extractParams.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import { extractFCParamsFactory, extractWeightsFactory, ParamMapping } from '../common';
 import { NetParams } from './types';
 
-export function extractParams(weights: Float32Array, channelsIn: number, channelsOut: number): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array, channelsIn: number, channelsOut: number): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
-  const extractFCParams = TfjsImageRecognitionBase.extractFCParamsFactory(extractWeights, paramMappings)
+  const extractFCParams = extractFCParamsFactory(extractWeights, paramMappings)
 
   const fc = extractFCParams(channelsIn, channelsOut, 'fc')
 

src/faceProcessor/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { disposeUnusedWeightTensors, extractWeightEntryFactory, FCParams, ParamMapping } from '../common';
 import { NetParams } from './types';
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
-  function extractFcParams(prefix: string): TfjsImageRecognitionBase.FCParams {
+  function extractFcParams(prefix: string): FCParams {
     const weights = extractWeightEntry<tf.Tensor2D>(`${prefix}/weights`, 2)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
     return { weights, bias }
@@ -21,7 +21,7 @@ export function extractParamsFromWeigthMap(
     fc: extractFcParams('fc')
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/faceProcessor/types.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+import { FCParams } from '../common';
 
 export type NetParams = {
-  fc: TfjsImageRecognitionBase.FCParams
+  fc: FCParams
 }
 

src/faceRecognitionNet/FaceRecognitionNet.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, normalize, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { NetInput, TNetInput, toNetInput } from '../dom';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { normalize } from '../ops';
 import { convDown } from './convLayer';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';

src/faceRecognitionNet/extractParams.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { isFloat, TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { ConvParams, extractWeightsFactory, ExtractWeightsFunction, ParamMapping } from '../common';
+import { isFloat } from '../utils';
 import { ConvLayerParams, NetParams, ResidualLayerParams, ScaleLayerParams } from './types';
 
-function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeightsFunction, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
 
   function extractFilterValues(numFilterValues: number, numFilters: number, filterSize: number): tf.Tensor4D {
     const weights = extractWeights(numFilterValues)
@@ -26,7 +27,7 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
     numFilters: number,
     filterSize: number,
     mappedPrefix: string
-  ): TfjsImageRecognitionBase.ConvParams {
+  ): ConvParams {
 
     const filters = extractFilterValues(numFilterValues, numFilters, filterSize)
     const bias = tf.tensor1d(extractWeights(numFilters))
@@ -89,14 +90,14 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
 
 }
 
-export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: ParamMapping[] } {
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractConvLayerParams,

src/faceRecognitionNet/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { isTensor2D, TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { disposeUnusedWeightTensors, extractWeightEntryFactory, ParamMapping } from '../common';
+import { isTensor2D } from '../utils';
 import { ConvLayerParams, NetParams, ResidualLayerParams, ScaleLayerParams } from './types';
 
-function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(weightMap: any, paramMappings: ParamMapping[]) {
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
   function extractScaleLayerParams(prefix: string): ScaleLayerParams {
 
@@ -40,9 +41,9 @@ function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBa
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractConvLayerParams,
@@ -94,7 +95,7 @@ export function extractParamsFromWeigthMap(
     fc
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/faceRecognitionNet/types.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { ConvParams } from '../common';
 
 export type ScaleLayerParams = {
   weights: tf.Tensor1D
@@ -11,7 +12,7 @@ export type ResidualLayerParams = {
 }
 
 export type ConvLayerParams = {
-  conv: TfjsImageRecognitionBase.ConvParams
+  conv: ConvParams
   scale: ScaleLayerParams
 }
 

src/factories/WithGender.ts

-import { isValidProbablitiy } from 'tfjs-image-recognition-base';
-
 import { Gender } from '../ageGenderNet/types';
+import { isValidProbablitiy } from '../utils';
 
 export type WithGender<TSource> = TSource & {
   gender: Gender

src/globalApi/ComputeFaceDescriptorsTasks.ts

-import { TNetInput } from 'tfjs-image-recognition-base';
-
+import { TNetInput } from '../dom';
 import { extendWithFaceDescriptor, WithFaceDescriptor } from '../factories/WithFaceDescriptor';
 import { WithFaceDetection } from '../factories/WithFaceDetection';
 import { WithFaceLandmarks } from '../factories/WithFaceLandmarks';

src/globalApi/DetectFaceLandmarksTasks.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TNetInput } from 'tfjs-image-recognition-base';
 
 import { FaceLandmarks68 } from '../classes/FaceLandmarks68';
-import { extractFaces, extractFaceTensors } from '../dom';
+import { extractFaces, extractFaceTensors, TNetInput } from '../dom';
 import { FaceLandmark68Net } from '../faceLandmarkNet/FaceLandmark68Net';
 import { FaceLandmark68TinyNet } from '../faceLandmarkNet/FaceLandmark68TinyNet';
 import { WithFaceDetection } from '../factories/WithFaceDetection';

src/globalApi/DetectFacesTasks.ts

-import { TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
-
 import { FaceDetection } from '../classes/FaceDetection';
+import { TNetInput } from '../dom';
 import { extendWithFaceDetection, WithFaceDetection } from '../factories/WithFaceDetection';
 import { MtcnnOptions } from '../mtcnn/MtcnnOptions';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1/SsdMobilenetv1Options';
 import { TinyFaceDetectorOptions } from '../tinyFaceDetector/TinyFaceDetectorOptions';
+import { TinyYolov2Options } from '../tinyYolov2';
 import { ComposableTask } from './ComposableTask';
 import { DetectAllFaceLandmarksTask, DetectSingleFaceLandmarksTask } from './DetectFaceLandmarksTasks';
 import { nets } from './nets';
@@ -38,7 +38,7 @@ export class DetectAllFacesTask extends DetectFacesTaskBase<FaceDetection[]> {
         options instanceof SsdMobilenetv1Options
           ? (input: TNetInput) => nets.ssdMobilenetv1.locateFaces(input, options)
           : (
-            options instanceof TfjsImageRecognitionBase.TinyYolov2Options
+            options instanceof TinyYolov2Options
               ? (input: TNetInput) => nets.tinyYolov2.locateFaces(input, options)
               : null
           )

src/globalApi/PredictAgeAndGenderTask.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TNetInput } from 'tfjs-image-recognition-base';
 
 import { AgeAndGenderPrediction } from '../ageGenderNet/types';
+import { TNetInput } from '../dom';
 import { extendWithAge, WithAge } from '../factories/WithAge';
 import { WithFaceDetection } from '../factories/WithFaceDetection';
 import { WithFaceLandmarks } from '../factories/WithFaceLandmarks';

src/globalApi/PredictFaceExpressionsTask.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TNetInput } from 'tfjs-image-recognition-base';
 
+import { TNetInput } from '../dom';
 import { FaceExpressions } from '../faceExpressionNet/FaceExpressions';
 import { WithFaceDetection } from '../factories/WithFaceDetection';
 import { extendWithFaceExpressions, WithFaceExpressions } from '../factories/WithFaceExpressions';

src/globalApi/allFaces.ts

-import { TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
-
+import { TNetInput } from '../dom';
 import { WithFaceDescriptor, WithFaceDetection, WithFaceLandmarks } from '../factories';
 import { IMtcnnOptions, MtcnnOptions } from '../mtcnn/MtcnnOptions';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1';
+import { ITinyYolov2Options, TinyYolov2Options } from '../tinyYolov2';
 import { detectAllFaces } from './detectFaces';
 
 // export allFaces API for backward compatibility
@@ -18,9 +18,9 @@ export async function allFacesSsdMobilenetv1(
 
 export async function allFacesTinyYolov2(
   input: TNetInput,
-  forwardParams: TfjsImageRecognitionBase.ITinyYolov2Options = {}
+  forwardParams: ITinyYolov2Options = {}
 ): Promise<WithFaceDescriptor<WithFaceLandmarks<WithFaceDetection<{}>>>[]> {
-  return await detectAllFaces(input, new TfjsImageRecognitionBase.TinyYolov2Options(forwardParams))
+  return await detectAllFaces(input, new TinyYolov2Options(forwardParams))
     .withFaceLandmarks()
     .withFaceDescriptors()
 }

src/globalApi/detectFaces.ts

-import { TNetInput } from 'tfjs-image-recognition-base';
-
+import { TNetInput } from '../dom';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1/SsdMobilenetv1Options';
 import { DetectAllFacesTask, DetectSingleFaceTask } from './DetectFacesTasks';
 import { FaceDetectionOptions } from './types';

src/globalApi/extractFacesAndComputeResults.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TNetInput } from 'tfjs-image-recognition-base';
 
 import { FaceDetection } from '../classes/FaceDetection';
-import { extractFaces, extractFaceTensors } from '../dom';
+import { extractFaces, extractFaceTensors, TNetInput } from '../dom';
 import { WithFaceDetection } from '../factories/WithFaceDetection';
 import { isWithFaceLandmarks, WithFaceLandmarks } from '../factories/WithFaceLandmarks';
 

src/globalApi/nets.ts

-import { TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
-
 import { AgeGenderNet } from '../ageGenderNet/AgeGenderNet';
 import { AgeAndGenderPrediction } from '../ageGenderNet/types';
 import { FaceDetection } from '../classes/FaceDetection';
 import { FaceLandmarks5 } from '../classes/FaceLandmarks5';
 import { FaceLandmarks68 } from '../classes/FaceLandmarks68';
+import { TNetInput } from '../dom';
 import { FaceExpressionNet } from '../faceExpressionNet/FaceExpressionNet';
 import { FaceExpressions } from '../faceExpressionNet/FaceExpressions';
 import { FaceLandmark68Net } from '../faceLandmarkNet/FaceLandmark68Net';
@@ -18,7 +17,7 @@ import { SsdMobilenetv1 } from '../ssdMobilenetv1/SsdMobilenetv1';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1/SsdMobilenetv1Options';
 import { TinyFaceDetector } from '../tinyFaceDetector/TinyFaceDetector';
 import { TinyFaceDetectorOptions } from '../tinyFaceDetector/TinyFaceDetectorOptions';
-import { TinyYolov2 } from '../tinyYolov2';
+import { ITinyYolov2Options, TinyYolov2 } from '../tinyYolov2';
 
 export const nets = {
   ssdMobilenetv1: new SsdMobilenetv1(),
@@ -59,7 +58,7 @@ export const tinyFaceDetector = (input: TNetInput, options: TinyFaceDetectorOpti
  * @param options (optional, default: see TinyYolov2Options constructor for default parameters).
  * @returns Bounding box of each face with score.
  */
-export const tinyYolov2 = (input: TNetInput, options: TfjsImageRecognitionBase.ITinyYolov2Options): Promise<FaceDetection[]> =>
+export const tinyYolov2 = (input: TNetInput, options: ITinyYolov2Options): Promise<FaceDetection[]> =>
   nets.tinyYolov2.locateFaces(input, options)
 
 /**

src/globalApi/types.ts

-import { TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
-
 import { FaceDetection } from '../classes/FaceDetection';
+import { TNetInput } from '../dom';
 import { MtcnnOptions } from '../mtcnn/MtcnnOptions';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1/SsdMobilenetv1Options';
 import { TinyFaceDetectorOptions } from '../tinyFaceDetector/TinyFaceDetectorOptions';
+import { TinyYolov2Options } from '../tinyYolov2';
 
-export type FaceDetectionOptions = TinyFaceDetectorOptions | SsdMobilenetv1Options | MtcnnOptions | TfjsImageRecognitionBase.TinyYolov2Options
+export type FaceDetectionOptions = TinyFaceDetectorOptions | SsdMobilenetv1Options | MtcnnOptions | TinyYolov2Options
 
 export type FaceDetectionFunction = (input: TNetInput) => Promise<FaceDetection[]>
\ No newline at end of file

src/index.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { draw as drawBase } from 'tfjs-image-recognition-base';
 
-import * as drawExtended from './draw';
+import * as draw from './draw';
+import * as utils from './utils';
 
 export {
+  draw,
+  utils,
   tf
 }
 
-export * from 'tfjs-image-recognition-base';
-
 export * from './ageGenderNet/index';
-const draw = {...drawBase, ...drawExtended }
-export { draw }
-
 export * from './classes/index';
 export * from './dom/index'
+export * from './env/index';
 export * from './faceExpressionNet/index';
 export * from './faceLandmarkNet/index';
 export * from './faceRecognitionNet/index';
 export * from './factories/index';
 export * from './globalApi/index';
 export * from './mtcnn/index';
+export * from './ops/index';
 export * from './ssdMobilenetv1/index';
 export * from './tinyFaceDetector/index';
 export * from './tinyYolov2/index';
 
 export * from './euclideanDistance';
+export * from './NeuralNetwork';
 export * from './resizeResults';
\ No newline at end of file

src/mtcnn/Mtcnn.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, Point, Rect, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { Point, Rect } from '../classes';
 import { FaceDetection } from '../classes/FaceDetection';
 import { FaceLandmarks5 } from '../classes/FaceLandmarks5';
+import { NetInput, TNetInput, toNetInput } from '../dom';
 import { extendWithFaceDetection, extendWithFaceLandmarks } from '../factories';
+import { NeuralNetwork } from '../NeuralNetwork';
 import { bgrToRgbTensor } from './bgrToRgbTensor';
 import { CELL_SIZE } from './config';
 import { extractParams } from './extractParams';

src/mtcnn/MtcnnBox.ts

-import { Box } from 'tfjs-image-recognition-base';
+import { Box } from '../classes';
 
 export class MtcnnBox extends Box<MtcnnBox> {
   constructor(left: number, top: number, right: number, bottom: number) {

src/mtcnn/ONet.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { convLayer } from '../common';
 import { fullyConnectedLayer } from '../common/fullyConnectedLayer';
 import { prelu } from './prelu';
 import { sharedLayer } from './sharedLayers';
@@ -11,7 +11,7 @@ export function ONet(x: tf.Tensor4D, params: ONetParams): { scores: tf.Tensor1D,
 
     let out = sharedLayer(x, params)
     out = tf.maxPool(out, [2, 2], [2, 2], 'same')
-    out = TfjsImageRecognitionBase.convLayer(out, params.conv4, 'valid')
+    out = convLayer(out, params.conv4, 'valid')
     out = prelu<tf.Tensor4D>(out, params.prelu4_alpha)
 
     const vectorized = tf.reshape(out, [out.shape[0], params.fc1.weights.shape[0]]) as tf.Tensor2D

src/mtcnn/PNet.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { convLayer } from '../common';
 import { sharedLayer } from './sharedLayers';
 import { PNetParams } from './types';
 
@@ -8,10 +8,10 @@ export function PNet(x: tf.Tensor4D, params: PNetParams): { prob: tf.Tensor4D, r
   return tf.tidy(() => {
 
     let out = sharedLayer(x, params, true)
-    const conv = TfjsImageRecognitionBase.convLayer(out, params.conv4_1, 'valid')
+    const conv = convLayer(out, params.conv4_1, 'valid')
     const max = tf.expandDims(tf.max(conv, 3), 3)
     const prob = tf.softmax(tf.sub(conv, max), 3) as tf.Tensor4D
-    const regions = TfjsImageRecognitionBase.convLayer(out, params.conv4_2, 'valid')
+    const regions = convLayer(out, params.conv4_2, 'valid')
 
     return { prob, regions }
   })

src/mtcnn/extractImagePatches.ts

 import * as tf from '@tensorflow/tfjs-core';
-import {
-  Box,
-  createCanvas,
-  createCanvasFromMedia,
-  env,
-  getContext2dOrThrow,
-  IDimensions,
-} from 'tfjs-image-recognition-base';
 
+import { Box, IDimensions } from '../classes';
+import { createCanvas, createCanvasFromMedia, getContext2dOrThrow } from '../dom';
+import { env } from '../env';
 import { normalize } from './normalize';
 
 export async function extractImagePatches(

src/mtcnn/extractParams.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import {
+  extractConvParamsFactory,
+  extractFCParamsFactory,
+  extractWeightsFactory,
+  ExtractWeightsFunction,
+  ParamMapping,
+} from '../common';
 import { NetParams, ONetParams, PNetParams, RNetParams, SharedParams } from './types';
 
-function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeightsFunction, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
 
-  const extractConvParams = TfjsImageRecognitionBase.extractConvParamsFactory(extractWeights, paramMappings)
-  const extractFCParams = TfjsImageRecognitionBase.extractFCParamsFactory(extractWeights, paramMappings)
+  const extractConvParams = extractConvParamsFactory(extractWeights, paramMappings)
+  const extractFCParams = extractFCParamsFactory(extractWeights, paramMappings)
 
   function extractPReluParams(size: number, paramPath: string): tf.Tensor1D {
     const alpha = tf.tensor1d(extractWeights(size))
@@ -68,14 +74,14 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
 
 }
 
-export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: ParamMapping[] } {
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractPNetParams,

src/mtcnn/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { ConvParams, disposeUnusedWeightTensors, extractWeightEntryFactory, FCParams, ParamMapping } from '../common';
 import { NetParams, ONetParams, PNetParams, RNetParams, SharedParams } from './types';
 
-function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(weightMap: any, paramMappings: ParamMapping[]) {
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
-  function extractConvParams(prefix: string): TfjsImageRecognitionBase.ConvParams {
+  function extractConvParams(prefix: string): ConvParams {
     const filters = extractWeightEntry<tf.Tensor4D>(`${prefix}/weights`, 4, `${prefix}/filters`)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
 
     return { filters, bias }
   }
 
-  function extractFCParams(prefix: string): TfjsImageRecognitionBase.FCParams {
+  function extractFCParams(prefix: string): FCParams {
     const weights = extractWeightEntry<tf.Tensor2D>(`${prefix}/weights`, 2)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
 
@@ -81,9 +81,9 @@ function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBa
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractPNetParams,
@@ -95,7 +95,7 @@ export function extractParamsFromWeigthMap(
   const rnet = extractRNetParams()
   const onet = extractONetParams()
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params: { pnet, rnet, onet }, paramMappings }
 }
\ No newline at end of file

src/mtcnn/pyramidDown.ts

-import { CELL_SIZE } from './config';
+import { CELL_SIZE } from './config'
 
 export function pyramidDown(minFaceSize: number, scaleFactor: number, dims: number[]): number[] {
 

src/mtcnn/sharedLayers.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { convLayer } from '../common';
 import { prelu } from './prelu';
 import { SharedParams } from './types';
 
 export function sharedLayer(x: tf.Tensor4D, params: SharedParams, isPnet: boolean = false) {
   return tf.tidy(() => {
 
-    let out = TfjsImageRecognitionBase.convLayer(x, params.conv1, 'valid')
+    let out = convLayer(x, params.conv1, 'valid')
     out = prelu<tf.Tensor4D>(out, params.prelu1_alpha)
     out = tf.maxPool(out, isPnet ? [2, 2]: [3, 3], [2, 2], 'same')
-    out = TfjsImageRecognitionBase.convLayer(out, params.conv2, 'valid')
+    out = convLayer(out, params.conv2, 'valid')
     out = prelu<tf.Tensor4D>(out, params.prelu2_alpha)
     out = isPnet ? out : tf.maxPool(out, [3, 3], [2, 2], 'valid')
-    out = TfjsImageRecognitionBase.convLayer(out, params.conv3, 'valid')
+    out = convLayer(out, params.conv3, 'valid')
     out = prelu<tf.Tensor4D>(out, params.prelu3_alpha)
 
     return out

src/mtcnn/stage1.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { BoundingBox, nonMaxSuppression, Point } from 'tfjs-image-recognition-base';
 
+import { BoundingBox, Point } from '../classes';
+import { nonMaxSuppression } from '../ops';
 import { CELL_SIZE, CELL_STRIDE } from './config';
 import { getSizesForScale } from './getSizesForScale';
 import { MtcnnBox } from './MtcnnBox';

src/mtcnn/stage2.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { Box, nonMaxSuppression } from 'tfjs-image-recognition-base';
 
+import { Box } from '../classes';
+import { nonMaxSuppression } from '../ops';
 import { extractImagePatches } from './extractImagePatches';
 import { MtcnnBox } from './MtcnnBox';
 import { RNet } from './RNet';

src/mtcnn/stage3.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { BoundingBox, Box, nonMaxSuppression, Point } from 'tfjs-image-recognition-base';
 
+import { BoundingBox, Box, Point } from '../classes';
+import { nonMaxSuppression } from '../ops';
 import { extractImagePatches } from './extractImagePatches';
 import { MtcnnBox } from './MtcnnBox';
 import { ONet } from './ONet';

src/mtcnn/types.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
 import { FaceLandmarks5 } from '../classes/FaceLandmarks5';
+import { ConvParams, FCParams } from '../common';
 import { WithFaceDetection, WithFaceLandmarks } from '../factories';
 
 export type SharedParams = {
-  conv1: TfjsImageRecognitionBase.ConvParams
+  conv1: ConvParams
   prelu1_alpha: tf.Tensor1D
-  conv2: TfjsImageRecognitionBase.ConvParams
+  conv2: ConvParams
   prelu2_alpha: tf.Tensor1D
-  conv3: TfjsImageRecognitionBase.ConvParams
+  conv3: ConvParams
   prelu3_alpha: tf.Tensor1D
 }
 
 export type PNetParams = SharedParams & {
-  conv4_1: TfjsImageRecognitionBase.ConvParams
-  conv4_2: TfjsImageRecognitionBase.ConvParams
+  conv4_1: ConvParams
+  conv4_2: ConvParams
 }
 
 export type RNetParams = SharedParams & {
-  fc1: TfjsImageRecognitionBase.FCParams
+  fc1: FCParams
   prelu4_alpha: tf.Tensor1D
-  fc2_1: TfjsImageRecognitionBase.FCParams
-  fc2_2: TfjsImageRecognitionBase.FCParams
+  fc2_1: FCParams
+  fc2_2: FCParams
 }
 
 export type ONetParams = SharedParams & {
-  conv4: TfjsImageRecognitionBase.ConvParams
+  conv4: ConvParams
   prelu4_alpha: tf.Tensor1D
-  fc1: TfjsImageRecognitionBase.FCParams
+  fc1: FCParams
   prelu5_alpha: tf.Tensor1D
-  fc2_1: TfjsImageRecognitionBase.FCParams
-  fc2_2: TfjsImageRecognitionBase.FCParams
-  fc2_3: TfjsImageRecognitionBase.FCParams
+  fc2_1: FCParams
+  fc2_2: FCParams
+  fc2_3: FCParams
 }
 
 export type NetParams = {

src/ops/index.ts

+export * from './iou'
+export * from './minBbox'
+export * from './nonMaxSuppression'
+export * from './normalize'
+export * from './padToSquare'
+export * from './shuffleArray'
+
+export function sigmoid(x: number) {
+  return 1 / (1 + Math.exp(-x))
+}
+
+export function inverseSigmoid(x: number) {
+  return Math.log(x / (1 - x))
+}
\ No newline at end of file

src/ops/iou.ts

+import { Box } from '../classes/Box';
+
+export function iou(box1: Box, box2: Box, isIOU: boolean = true) {
+  const width = Math.max(0.0, Math.min(box1.right, box2.right) - Math.max(box1.left, box2.left))
+  const height = Math.max(0.0, Math.min(box1.bottom, box2.bottom) - Math.max(box1.top, box2.top))
+  const interSection = width * height
+
+  return isIOU
+    ? interSection / (box1.area + box2.area - interSection)
+    : interSection / Math.min(box1.area, box2.area)
+}
\ No newline at end of file

src/minBbox.ts → src/ops/minBbox.ts

-import { BoundingBox, IPoint } from 'tfjs-image-recognition-base';
+import { BoundingBox, IPoint } from '../classes';
 
 export function minBbox(pts: IPoint[]): BoundingBox {
   const xs = pts.map(pt => pt.x)

src/ops/nonMaxSuppression.ts

+import { Box } from '../classes/Box';
+import { iou } from './iou';
+
+export function nonMaxSuppression(
+  boxes: Box[],
+  scores: number[],
+  iouThreshold: number,
+  isIOU: boolean = true
+): number[] {
+
+  let indicesSortedByScore = scores
+    .map((score, boxIndex) => ({ score, boxIndex }))
+    .sort((c1, c2) => c1.score - c2.score)
+    .map(c => c.boxIndex)
+
+  const pick: number[] = []
+
+  while(indicesSortedByScore.length > 0) {
+    const curr = indicesSortedByScore.pop() as number
+    pick.push(curr)
+
+    const indices = indicesSortedByScore
+
+    const outputs: number[] = []
+    for (let i = 0; i < indices.length; i++) {
+      const idx = indices[i]
+
+      const currBox = boxes[curr]
+      const idxBox = boxes[idx]
+
+      outputs.push(iou(currBox, idxBox, isIOU))
+    }
+
+    indicesSortedByScore = indicesSortedByScore.filter(
+      (_, j) => outputs[j] <= iouThreshold
+    )
+  }
+
+  return pick
+
+}
\ No newline at end of file

src/ops/normalize.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+export function normalize(x: tf.Tensor4D, meanRgb: number[]): tf.Tensor4D {
+  return tf.tidy(() => {
+    const [r, g, b] = meanRgb
+    const avg_r = tf.fill([...x.shape.slice(0, 3), 1], r)
+    const avg_g = tf.fill([...x.shape.slice(0, 3), 1], g)
+    const avg_b = tf.fill([...x.shape.slice(0, 3), 1], b)
+    const avg_rgb = tf.concat([avg_r, avg_g, avg_b], 3)
+
+    return tf.sub(x, avg_rgb)
+  })
+}
\ No newline at end of file

src/ops/padToSquare.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+/**
+ * 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 an equal amount of padding on
+ * both sides of the minor dimension oof the image.
+ * @returns The padded tensor with width === height.
+ */
+export function padToSquare(
+  imgTensor: tf.Tensor4D,
+  isCenterImage: boolean = false
+): tf.Tensor4D {
+  return tf.tidy(() => {
+
+    const [height, width] = imgTensor.shape.slice(1)
+    if (height === width) {
+      return imgTensor
+    }
+
+    const dimDiff = Math.abs(height - width)
+    const paddingAmount = Math.round(dimDiff * (isCenterImage ? 0.5 : 1))
+    const paddingAxis = height > width ? 2 : 1
+
+    const createPaddingTensor = (paddingAmount: number): tf.Tensor => {
+      const paddingTensorShape = imgTensor.shape.slice()
+      paddingTensorShape[paddingAxis] = paddingAmount
+      return tf.fill(paddingTensorShape, 0)
+    }
+
+    const paddingTensorAppend = createPaddingTensor(paddingAmount)
+    const remainingPaddingAmount = dimDiff - (paddingTensorAppend.shape[paddingAxis] as number)
+
+    const paddingTensorPrepend = isCenterImage && remainingPaddingAmount
+      ? createPaddingTensor(remainingPaddingAmount)
+      : null
+
+    const tensorsToStack = [
+      paddingTensorPrepend,
+      imgTensor,
+      paddingTensorAppend
+    ]
+      .filter(t => !!t)
+      .map((t: tf.Tensor) => t.toFloat()) as tf.Tensor4D[]
+    return tf.concat(tensorsToStack, paddingAxis)
+  })
+}
\ No newline at end of file

src/ops/shuffleArray.ts

+export function shuffleArray(inputArray: any[]) {
+  const array = inputArray.slice()
+  for (let i = array.length - 1; i > 0; i--) {
+      const j = Math.floor(Math.random() * (i + 1))
+      const x = array[i]
+      array[i] = array[j]
+      array[j] = x
+  }
+  return array
+}
\ No newline at end of file

src/resizeResults.ts

-import { Dimensions, IDimensions } from 'tfjs-image-recognition-base';
-
+import { Dimensions, IDimensions } from './classes';
 import { FaceDetection } from './classes/FaceDetection';
 import { FaceLandmarks } from './classes/FaceLandmarks';
 import { extendWithFaceDetection, isWithFaceDetection } from './factories/WithFaceDetection';

src/ssdMobilenetv1/SsdMobilenetv1.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { NetInput, NeuralNetwork, Rect, TNetInput, toNetInput } from 'tfjs-image-recognition-base';
 
+import { Rect } from '../classes';
 import { FaceDetection } from '../classes/FaceDetection';
+import { NetInput, TNetInput, toNetInput } from '../dom';
+import { NeuralNetwork } from '../NeuralNetwork';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';
 import { mobileNetV1 } from './mobileNetV1';

src/ssdMobilenetv1/boxPredictionLayer.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { convLayer } from '../common';
 import { BoxPredictionParams } from './types';
 
 
@@ -13,11 +13,11 @@ export function boxPredictionLayer(
     const batchSize = x.shape[0]
 
     const boxPredictionEncoding = tf.reshape(
-      TfjsImageRecognitionBase.convLayer(x, params.box_encoding_predictor),
+      convLayer(x, params.box_encoding_predictor),
       [batchSize, -1, 1, 4]
     )
     const classPrediction = tf.reshape(
-      TfjsImageRecognitionBase.convLayer(x, params.class_predictor),
+      convLayer(x, params.class_predictor),
       [batchSize, -1, 3]
     )
 

src/ssdMobilenetv1/extractParams.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { ExtractWeightsFunction, ParamMapping, ConvParams, extractWeightsFactory } from '../common';
 import { MobileNetV1, NetParams, PointwiseConvParams, PredictionLayerParams } from './types';
 
-function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeightsFunction, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
 
   function extractDepthwiseConvParams(numChannels: number, mappedPrefix: string): MobileNetV1.DepthwiseConvParams {
 
@@ -36,7 +36,7 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
     filterSize: number,
     mappedPrefix: string,
     isPointwiseConv?: boolean
-  ): TfjsImageRecognitionBase.ConvParams {
+  ): ConvParams {
 
     const filters = tf.tensor4d(
       extractWeights(channelsIn * channelsOut * filterSize * filterSize),
@@ -191,14 +191,14 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
 
 }
 
-export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
   const {
     extractMobilenetV1Params,

src/ssdMobilenetv1/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { isTensor3D, TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
 
+import { ConvParams, disposeUnusedWeightTensors, extractWeightEntryFactory, ParamMapping } from '../common';
+import { isTensor3D } from '../utils';
 import { BoxPredictionParams, MobileNetV1, NetParams, PointwiseConvParams, PredictionLayerParams } from './types';
 
-function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(weightMap: any, paramMappings: ParamMapping[]) {
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
   function extractPointwiseConvParams(prefix: string, idx: number, mappedPrefix: string): PointwiseConvParams {
 
@@ -59,7 +60,7 @@ function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBa
     }
   }
 
-  function extractConvParams(prefix: string, mappedPrefix: string): TfjsImageRecognitionBase.ConvParams {
+  function extractConvParams(prefix: string, mappedPrefix: string): ConvParams {
     const filters = extractWeightEntry<tf.Tensor4D>(`${prefix}/weights`, 4, `${mappedPrefix}/filters`)
     const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/biases`, 1, `${mappedPrefix}/bias`)
 
@@ -107,9 +108,9 @@ function extractorsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBa
 
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap
-): { params: NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: NetParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractMobilenetV1Params,
@@ -131,7 +132,7 @@ export function extractParamsFromWeigthMap(
     }
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params, paramMappings }
 }
\ No newline at end of file

src/ssdMobilenetv1/types.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+
+import { ConvParams } from '../common';
 
 export type PointwiseConvParams = {
   filters: tf.Tensor4D
@@ -41,8 +42,8 @@ export namespace MobileNetV1 {
 }
 
 export type BoxPredictionParams = {
-  box_encoding_predictor: TfjsImageRecognitionBase.ConvParams
-  class_predictor: TfjsImageRecognitionBase.ConvParams
+  box_encoding_predictor: ConvParams
+  class_predictor: ConvParams
 }
 
 export type PredictionLayerParams = {

src/tinyFaceDetector/TinyFaceDetector.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { Point, TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
 
-import { FaceDetection } from '../classes';
+import { FaceDetection, Point } from '../classes';
+import { ParamMapping } from '../common';
+import { TNetInput } from '../dom';
+import { ITinyYolov2Options } from '../tinyYolov2';
+import { TinyYolov2Base } from '../tinyYolov2/TinyYolov2Base';
+import { TinyYolov2NetParams } from '../tinyYolov2/types';
 import { BOX_ANCHORS, IOU_THRESHOLD, MEAN_RGB } from './const';
 
-export class TinyFaceDetector extends TfjsImageRecognitionBase.TinyYolov2 {
+export class TinyFaceDetector extends TinyYolov2Base {
 
   constructor() {
     const config = {
@@ -24,7 +28,7 @@ export class TinyFaceDetector extends TfjsImageRecognitionBase.TinyYolov2 {
     return this.config.anchors
   }
 
-  public async locateFaces(input: TNetInput, forwardParams: TfjsImageRecognitionBase.ITinyYolov2Options): Promise<FaceDetection[]> {
+  public async locateFaces(input: TNetInput, forwardParams: ITinyYolov2Options): Promise<FaceDetection[]> {
     const objectDetections = await this.detect(input, forwardParams)
     return objectDetections.map(det => new FaceDetection(det.score, det.relativeBox, { width: det.imageWidth, height: det.imageHeight }))
   }
@@ -33,7 +37,7 @@ export class TinyFaceDetector extends TfjsImageRecognitionBase.TinyYolov2 {
     return 'tiny_face_detector_model'
   }
 
-  protected extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap): { params: TfjsImageRecognitionBase.TinyYolov2NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+  protected extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap): { params: TinyYolov2NetParams, paramMappings: ParamMapping[] } {
     return super.extractParamsFromWeigthMap(weightMap)
   }
 }
\ No newline at end of file

src/tinyFaceDetector/TinyFaceDetectorOptions.ts

-import { TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
+import { ITinyYolov2Options, TinyYolov2Options } from '../tinyYolov2';
 
-export interface ITinyFaceDetectorOptions extends TfjsImageRecognitionBase.ITinyYolov2Options {}
+export interface ITinyFaceDetectorOptions extends ITinyYolov2Options {}
 
-export class TinyFaceDetectorOptions extends TfjsImageRecognitionBase.TinyYolov2Options {
+export class TinyFaceDetectorOptions extends TinyYolov2Options {
   protected _name: string = 'TinyFaceDetectorOptions'
 }
\ No newline at end of file

src/tinyFaceDetector/const.ts

-import { Point } from 'tfjs-image-recognition-base';
+import { Point } from '../classes'
 
 export const IOU_THRESHOLD = 0.4
 

src/tinyYolov2/TinyYolov2.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { Point, TfjsImageRecognitionBase, TNetInput } from 'tfjs-image-recognition-base';
 
-import { FaceDetection } from '../classes';
+import { FaceDetection, Point } from '../classes';
+import { ParamMapping } from '../common/types';
+import { TNetInput } from '../dom/types';
 import {
   BOX_ANCHORS,
   BOX_ANCHORS_SEPARABLE,
@@ -10,8 +11,11 @@ import {
   IOU_THRESHOLD,
   MEAN_RGB_SEPARABLE,
 } from './const';
+import { TinyYolov2Base } from './TinyYolov2Base';
+import { ITinyYolov2Options } from './TinyYolov2Options';
+import { TinyYolov2NetParams } from './types';
 
-export class TinyYolov2 extends TfjsImageRecognitionBase.TinyYolov2 {
+export class TinyYolov2 extends TinyYolov2Base {
 
   constructor(withSeparableConvs: boolean = true) {
     const config = Object.assign({}, {
@@ -41,7 +45,7 @@ export class TinyYolov2 extends TfjsImageRecognitionBase.TinyYolov2 {
     return this.config.anchors
   }
 
-  public async locateFaces(input: TNetInput, forwardParams: TfjsImageRecognitionBase.ITinyYolov2Options): Promise<FaceDetection[]> {
+  public async locateFaces(input: TNetInput, forwardParams: ITinyYolov2Options): Promise<FaceDetection[]> {
     const objectDetections = await this.detect(input, forwardParams)
     return objectDetections.map(det => new FaceDetection(det.score, det.relativeBox, { width: det.imageWidth, height: det.imageHeight }))
   }
@@ -50,7 +54,7 @@ export class TinyYolov2 extends TfjsImageRecognitionBase.TinyYolov2 {
     return this.withSeparableConvs ? DEFAULT_MODEL_NAME_SEPARABLE_CONV : DEFAULT_MODEL_NAME
   }
 
-  protected extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap): { params: TfjsImageRecognitionBase.TinyYolov2NetParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+  protected extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap): { params: TinyYolov2NetParams, paramMappings: ParamMapping[] } {
     return super.extractParamsFromWeigthMap(weightMap)
   }
 }
\ No newline at end of file

src/tinyYolov2/TinyYolov2Base.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { BoundingBox } from '../classes/BoundingBox';
+import { Dimensions } from '../classes/Dimensions';
+import { ObjectDetection } from '../classes/ObjectDetection';
+import { convLayer } from '../common';
+import { ConvParams, SeparableConvParams } from '../common/types';
+import { toNetInput } from '../dom';
+import { NetInput } from '../dom/NetInput';
+import { TNetInput } from '../dom/types';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { sigmoid } from '../ops';
+import { nonMaxSuppression } from '../ops/nonMaxSuppression';
+import { normalize } from '../ops/normalize';
+import { TinyYolov2Config, validateConfig } from './config';
+import { convWithBatchNorm } from './convWithBatchNorm';
+import { depthwiseSeparableConv } from './depthwiseSeparableConv';
+import { extractParams } from './extractParams';
+import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';
+import { leaky } from './leaky';
+import { ITinyYolov2Options, TinyYolov2Options } from './TinyYolov2Options';
+import { DefaultTinyYolov2NetParams, MobilenetParams, TinyYolov2NetParams } from './types';
+
+export class TinyYolov2Base extends NeuralNetwork<TinyYolov2NetParams> {
+
+  public static DEFAULT_FILTER_SIZES = [
+    3, 16, 32, 64, 128, 256, 512, 1024, 1024
+  ]
+
+  private _config: TinyYolov2Config
+
+  constructor(config: TinyYolov2Config) {
+    super('TinyYolov2')
+    validateConfig(config)
+    this._config = config
+  }
+
+  public get config(): TinyYolov2Config {
+    return this._config
+  }
+
+  public get withClassScores(): boolean {
+    return this.config.withClassScores || this.config.classes.length > 1
+  }
+
+  public get boxEncodingSize(): number {
+    return 5 + (this.withClassScores ? this.config.classes.length : 0)
+  }
+
+  public runTinyYolov2(x: tf.Tensor4D, params: DefaultTinyYolov2NetParams): tf.Tensor4D {
+
+    let out = convWithBatchNorm(x, params.conv0)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = convWithBatchNorm(out, params.conv1)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = convWithBatchNorm(out, params.conv2)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = convWithBatchNorm(out, params.conv3)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = convWithBatchNorm(out, params.conv4)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = convWithBatchNorm(out, params.conv5)
+    out = tf.maxPool(out, [2, 2], [1, 1], 'same')
+    out = convWithBatchNorm(out, params.conv6)
+    out = convWithBatchNorm(out, params.conv7)
+
+    return convLayer(out, params.conv8, 'valid', false)
+  }
+
+  public runMobilenet(x: tf.Tensor4D, params: MobilenetParams): tf.Tensor4D {
+
+    let out = this.config.isFirstLayerConv2d
+      ? leaky(convLayer(x, params.conv0 as ConvParams, 'valid', false))
+      : depthwiseSeparableConv(x, params.conv0 as SeparableConvParams)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = depthwiseSeparableConv(out, params.conv1)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = depthwiseSeparableConv(out, params.conv2)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = depthwiseSeparableConv(out, params.conv3)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = depthwiseSeparableConv(out, params.conv4)
+    out = tf.maxPool(out, [2, 2], [2, 2], 'same')
+    out = depthwiseSeparableConv(out, params.conv5)
+    out = tf.maxPool(out, [2, 2], [1, 1], 'same')
+    out = params.conv6 ? depthwiseSeparableConv(out, params.conv6) : out
+    out = params.conv7 ? depthwiseSeparableConv(out, params.conv7) : out
+
+    return convLayer(out, params.conv8, 'valid', false)
+  }
+
+  public forwardInput(input: NetInput, inputSize: number): tf.Tensor4D {
+
+    const { params } = this
+
+    if (!params) {
+      throw new Error('TinyYolov2 - load model before inference')
+    }
+
+    return tf.tidy(() => {
+
+      let batchTensor = input.toBatchTensor(inputSize, false).toFloat()
+      batchTensor = this.config.meanRgb
+        ? normalize(batchTensor, this.config.meanRgb)
+        : batchTensor
+      batchTensor = batchTensor.div(tf.scalar(256)) as tf.Tensor4D
+
+      return this.config.withSeparableConvs
+        ? this.runMobilenet(batchTensor, params as MobilenetParams)
+        : this.runTinyYolov2(batchTensor, params as DefaultTinyYolov2NetParams)
+    })
+  }
+
+  public async forward(input: TNetInput, inputSize: number): Promise<tf.Tensor4D> {
+    return await this.forwardInput(await toNetInput(input), inputSize)
+  }
+
+  public async detect(input: TNetInput, forwardParams: ITinyYolov2Options = {}): Promise<ObjectDetection[]> {
+
+    const { inputSize, scoreThreshold } = new TinyYolov2Options(forwardParams)
+
+    const netInput = await toNetInput(input)
+    const out = await this.forwardInput(netInput, inputSize)
+    const out0 = tf.tidy(() => tf.unstack(out)[0].expandDims()) as tf.Tensor4D
+
+    const inputDimensions = {
+      width: netInput.getInputWidth(0),
+      height: netInput.getInputHeight(0)
+    }
+
+    const results = await this.extractBoxes(out0, netInput.getReshapedInputDimensions(0), scoreThreshold)
+    out.dispose()
+    out0.dispose()
+
+    const boxes = results.map(res => res.box)
+    const scores = results.map(res => res.score)
+    const classScores = results.map(res => res.classScore)
+    const classNames = results.map(res => this.config.classes[res.label])
+
+    const indices = nonMaxSuppression(
+      boxes.map(box => box.rescale(inputSize)),
+      scores,
+      this.config.iouThreshold,
+      true
+    )
+
+    const detections = indices.map(idx =>
+      new ObjectDetection(
+        scores[idx],
+        classScores[idx],
+        classNames[idx],
+        boxes[idx],
+        inputDimensions
+      )
+    )
+
+    return detections
+  }
+
+  protected getDefaultModelName(): string {
+    return ''
+  }
+
+  protected extractParamsFromWeigthMap(weightMap: tf.NamedTensorMap) {
+    return extractParamsFromWeigthMap(weightMap, this.config)
+  }
+
+  protected extractParams(weights: Float32Array) {
+    const filterSizes = this.config.filterSizes || TinyYolov2Base.DEFAULT_FILTER_SIZES
+
+    const numFilters = filterSizes ? filterSizes.length : undefined
+    if (numFilters !== 7 && numFilters !== 8 && numFilters !== 9) {
+      throw new Error(`TinyYolov2 - expected 7 | 8 | 9 convolutional filters, but found ${numFilters} filterSizes in config`)
+    }
+    return extractParams(weights, this.config, this.boxEncodingSize, filterSizes)
+  }
+
+  protected async extractBoxes(
+    outputTensor: tf.Tensor4D,
+    inputBlobDimensions: Dimensions,
+    scoreThreshold?: number
+  ) {
+
+    const { width, height } = inputBlobDimensions
+    const inputSize = Math.max(width, height)
+    const correctionFactorX = inputSize / width
+    const correctionFactorY = inputSize / height
+
+    const numCells = outputTensor.shape[1]
+    const numBoxes = this.config.anchors.length
+
+    const [boxesTensor, scoresTensor, classScoresTensor] = tf.tidy(() => {
+      const reshaped = outputTensor.reshape([numCells, numCells, numBoxes, this.boxEncodingSize])
+
+      const boxes = reshaped.slice([0, 0, 0, 0], [numCells, numCells, numBoxes, 4])
+      const scores = reshaped.slice([0, 0, 0, 4], [numCells, numCells, numBoxes, 1])
+      const classScores = this.withClassScores
+        ? tf.softmax(reshaped.slice([0, 0, 0, 5], [numCells, numCells, numBoxes, this.config.classes.length]), 3)
+        : tf.scalar(0)
+      return [boxes, scores, classScores]
+    })
+
+    const results = []
+
+    const scoresData = await scoresTensor.array()
+    const boxesData = await boxesTensor.array()
+    for (let row = 0; row < numCells; row ++) {
+      for (let col = 0; col < numCells; col ++) {
+        for (let anchor = 0; anchor < numBoxes; anchor ++) {
+
+          const score = sigmoid(scoresData[row][col][anchor][0]);
+          if (!scoreThreshold || score > scoreThreshold) {
+            const ctX = ((col + sigmoid(boxesData[row][col][anchor][0])) / numCells) * correctionFactorX
+            const ctY = ((row + sigmoid(boxesData[row][col][anchor][1])) / numCells) * correctionFactorY
+            const width = ((Math.exp(boxesData[row][col][anchor][2]) * this.config.anchors[anchor].x) / numCells) * correctionFactorX
+            const height = ((Math.exp(boxesData[row][col][anchor][3]) * this.config.anchors[anchor].y) / numCells) * correctionFactorY
+
+            const x = (ctX - (width / 2))
+            const y = (ctY - (height / 2))
+
+            const pos = { row, col, anchor }
+            const { classScore, label } = this.withClassScores
+              ? await this.extractPredictedClass(classScoresTensor as tf.Tensor4D, pos)
+              : { classScore: 1, label: 0 }
+
+            results.push({
+              box: new BoundingBox(x, y, x + width, y + height),
+              score: score,
+              classScore: score * classScore,
+              label,
+              ...pos
+            })
+          }
+        }
+      }
+    }
+
+    boxesTensor.dispose()
+    scoresTensor.dispose()
+    classScoresTensor.dispose()
+
+    return results
+  }
+
+  private async extractPredictedClass(classesTensor: tf.Tensor4D, pos: { row: number, col: number, anchor: number },) {
+    const { row, col, anchor } = pos
+    const classesData = await classesTensor.array()
+    return Array(this.config.classes.length).fill(0)
+      .map((_, i) => classesData[row][col][anchor][i])
+      .map((classScore, label) => ({
+        classScore,
+        label
+      }))
+      .reduce((max, curr) => max.classScore > curr.classScore ? max : curr)
+  }
+}
\ No newline at end of file

src/tinyYolov2/TinyYolov2Options.ts

+export enum TinyYolov2SizeType {
+  XS = 224,
+  SM = 320,
+  MD = 416,
+  LG = 608
+}
+
+export interface ITinyYolov2Options {
+  inputSize?: number
+  scoreThreshold?: number
+}
+
+export class TinyYolov2Options {
+  protected _name: string = 'TinyYolov2Options'
+
+  private _inputSize: number
+  private _scoreThreshold: number
+
+  constructor({ inputSize, scoreThreshold }: ITinyYolov2Options = {}) {
+    this._inputSize = inputSize || 416
+    this._scoreThreshold = scoreThreshold || 0.5
+
+    if (typeof this._inputSize !== 'number' || this._inputSize % 32 !== 0) {
+      throw new Error(`${this._name} - expected inputSize to be a number divisible by 32`)
+    }
+
+    if (typeof this._scoreThreshold !== 'number' || this._scoreThreshold <= 0 || this._scoreThreshold >= 1) {
+      throw new Error(`${this._name} - expected scoreThreshold to be a number between 0 and 1`)
+    }
+  }
+
+  get inputSize(): number { return this._inputSize }
+  get scoreThreshold(): number { return this._scoreThreshold }
+}
\ No newline at end of file

src/tinyYolov2/config.ts

+import { Point } from '../classes/Point';
+
+export type TinyYolov2Config = {
+  withSeparableConvs: boolean
+  iouThreshold: number
+  anchors: Point[]
+  classes: string[]
+  meanRgb?: [number, number, number]
+  withClassScores?: boolean,
+  filterSizes?: number[]
+  isFirstLayerConv2d?: boolean
+}
+
+const isNumber = (arg: any) => typeof arg === 'number'
+
+export function validateConfig(config: any) {
+  if (!config) {
+    throw new Error(`invalid config: ${config}`)
+  }
+
+  if (typeof config.withSeparableConvs !== 'boolean') {
+    throw new Error(`config.withSeparableConvs has to be a boolean, have: ${config.withSeparableConvs}`)
+  }
+
+  if (!isNumber(config.iouThreshold) || config.iouThreshold < 0 || config.iouThreshold > 1.0) {
+    throw new Error(`config.iouThreshold has to be a number between [0, 1], have: ${config.iouThreshold}`)
+  }
+
+  if (
+    !Array.isArray(config.classes)
+    || !config.classes.length
+    || !config.classes.every((c: any) => typeof c === 'string')
+  ) {
+
+    throw new Error(`config.classes has to be an array class names: string[], have: ${JSON.stringify(config.classes)}`)
+  }
+
+  if (
+    !Array.isArray(config.anchors)
+    || !config.anchors.length
+    || !config.anchors.map((a: any) => a || {}).every((a: any) => isNumber(a.x) && isNumber(a.y))
+  ) {
+
+    throw new Error(`config.anchors has to be an array of { x: number, y: number }, have: ${JSON.stringify(config.anchors)}`)
+  }
+
+  if (config.meanRgb && (
+    !Array.isArray(config.meanRgb)
+    || config.meanRgb.length !== 3
+    || !config.meanRgb.every(isNumber)
+  )) {
+
+    throw new Error(`config.meanRgb has to be an array of shape [number, number, number], have: ${JSON.stringify(config.meanRgb)}`)
+  }
+}
\ No newline at end of file

src/tinyYolov2/const.ts

-import { Point } from 'tfjs-image-recognition-base';
+import { Point } from '../classes';
 
 export const IOU_THRESHOLD = 0.4
 

src/tinyYolov2/convWithBatchNorm.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { leaky } from './leaky';
+import { ConvWithBatchNorm } from './types';
+
+export function convWithBatchNorm(x: tf.Tensor4D, params: ConvWithBatchNorm): tf.Tensor4D  {
+  return tf.tidy(() => {
+    let out = tf.pad(x, [[0, 0], [1, 1], [1, 1], [0, 0]]) as tf.Tensor4D
+
+    out = tf.conv2d(out, params.conv.filters, [1, 1], 'valid')
+    out = tf.sub(out, params.bn.sub)
+    out = tf.mul(out, params.bn.truediv)
+    out = tf.add(out, params.conv.bias)
+
+    return leaky(out)
+  })
+}
\ No newline at end of file

src/tinyYolov2/depthwiseSeparableConv.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { SeparableConvParams } from '../common/types';
+import { leaky } from './leaky';
+
+export function depthwiseSeparableConv(x: tf.Tensor4D, params: SeparableConvParams): tf.Tensor4D  {
+  return tf.tidy(() => {
+    let out = tf.pad(x, [[0, 0], [1, 1], [1, 1], [0, 0]]) as tf.Tensor4D
+
+    out = tf.separableConv2d(out, params.depthwise_filter, params.pointwise_filter, [1, 1], 'valid')
+    out = tf.add(out, params.bias)
+
+    return leaky(out)
+  })
+}
\ No newline at end of file

src/tinyYolov2/extractParams.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { extractConvParamsFactory } from '../common';
+import { extractSeparableConvParamsFactory } from '../common/extractSeparableConvParamsFactory';
+import { extractWeightsFactory } from '../common/extractWeightsFactory';
+import { ExtractWeightsFunction, ParamMapping } from '../common/types';
+import { TinyYolov2Config } from './config';
+import { BatchNorm, ConvWithBatchNorm, TinyYolov2NetParams } from './types';
+
+function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
+
+  const extractConvParams = extractConvParamsFactory(extractWeights, paramMappings)
+
+  function extractBatchNormParams(size: number, mappedPrefix: string): BatchNorm {
+
+    const sub = tf.tensor1d(extractWeights(size))
+    const truediv = tf.tensor1d(extractWeights(size))
+
+    paramMappings.push(
+      { paramPath: `${mappedPrefix}/sub` },
+      { paramPath: `${mappedPrefix}/truediv` }
+    )
+
+    return { sub, truediv }
+  }
+
+  function extractConvWithBatchNormParams(channelsIn: number, channelsOut: number, mappedPrefix: string): ConvWithBatchNorm {
+
+    const conv = extractConvParams(channelsIn, channelsOut, 3, `${mappedPrefix}/conv`)
+    const bn = extractBatchNormParams(channelsOut, `${mappedPrefix}/bn`)
+
+    return { conv, bn }
+  }
+  const extractSeparableConvParams = extractSeparableConvParamsFactory(extractWeights, paramMappings)
+
+  return {
+    extractConvParams,
+    extractConvWithBatchNormParams,
+    extractSeparableConvParams
+  }
+
+}
+
+export function extractParams(
+  weights: Float32Array,
+  config: TinyYolov2Config,
+  boxEncodingSize: number,
+  filterSizes: number[]
+): { params: TinyYolov2NetParams, paramMappings: ParamMapping[] } {
+
+  const {
+    extractWeights,
+    getRemainingWeights
+  } = extractWeightsFactory(weights)
+
+  const paramMappings: ParamMapping[] = []
+
+  const {
+    extractConvParams,
+    extractConvWithBatchNormParams,
+    extractSeparableConvParams
+  } = extractorsFactory(extractWeights, paramMappings)
+
+  let params: TinyYolov2NetParams
+
+  if (config.withSeparableConvs) {
+    const [s0, s1, s2, s3, s4, s5, s6, s7, s8] = filterSizes
+
+    const conv0 = config.isFirstLayerConv2d
+      ? extractConvParams(s0, s1, 3, 'conv0')
+      : extractSeparableConvParams(s0, s1, 'conv0')
+    const conv1 = extractSeparableConvParams(s1, s2, 'conv1')
+    const conv2 = extractSeparableConvParams(s2, s3, 'conv2')
+    const conv3 = extractSeparableConvParams(s3, s4, 'conv3')
+    const conv4 = extractSeparableConvParams(s4, s5, 'conv4')
+    const conv5 = extractSeparableConvParams(s5, s6, 'conv5')
+    const conv6 = s7 ? extractSeparableConvParams(s6, s7, 'conv6') : undefined
+    const conv7 = s8 ? extractSeparableConvParams(s7, s8, 'conv7') : undefined
+    const conv8 = extractConvParams(s8 || s7 || s6, 5 * boxEncodingSize, 1, 'conv8')
+    params = { conv0, conv1, conv2, conv3, conv4, conv5, conv6, conv7, conv8 }
+  } else {
+    const [s0, s1, s2, s3, s4, s5, s6, s7, s8] = filterSizes
+    const conv0 = extractConvWithBatchNormParams(s0, s1, 'conv0',)
+    const conv1 = extractConvWithBatchNormParams(s1, s2, 'conv1')
+    const conv2 = extractConvWithBatchNormParams(s2, s3, 'conv2')
+    const conv3 = extractConvWithBatchNormParams(s3, s4, 'conv3')
+    const conv4 = extractConvWithBatchNormParams(s4, s5, 'conv4')
+    const conv5 = extractConvWithBatchNormParams(s5, s6, 'conv5')
+    const conv6 = extractConvWithBatchNormParams(s6, s7, 'conv6')
+    const conv7 = extractConvWithBatchNormParams(s7, s8, 'conv7')
+    const conv8 = extractConvParams(s8, 5 * boxEncodingSize, 1, 'conv8')
+    params = { conv0, conv1, conv2, conv3, conv4, conv5, conv6, conv7, conv8 }
+  }
+
+  if (getRemainingWeights().length !== 0) {
+    throw new Error(`weights remaing after extract: ${getRemainingWeights().length}`)
+  }
+
+
+  return { params, paramMappings }
+}
\ No newline at end of file

src/tinyYolov2/extractParamsFromWeigthMap.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ConvParams } from '../common';
+import { disposeUnusedWeightTensors } from '../common/disposeUnusedWeightTensors';
+import { loadSeparableConvParamsFactory } from '../common/extractSeparableConvParamsFactory';
+import { extractWeightEntryFactory } from '../common/extractWeightEntryFactory';
+import { ParamMapping } from '../common/types';
+import { TinyYolov2Config } from './config';
+import { BatchNorm, ConvWithBatchNorm, TinyYolov2NetParams } from './types';
+
+function extractorsFactory(weightMap: any, paramMappings: ParamMapping[]) {
+
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
+
+  function extractBatchNormParams(prefix: string): BatchNorm {
+    const sub = extractWeightEntry<tf.Tensor1D>(`${prefix}/sub`, 1)
+    const truediv = extractWeightEntry<tf.Tensor1D>(`${prefix}/truediv`, 1)
+    return { sub, truediv }
+  }
+
+  function extractConvParams(prefix: string): ConvParams {
+    const filters = extractWeightEntry<tf.Tensor4D>(`${prefix}/filters`, 4)
+    const bias = extractWeightEntry<tf.Tensor1D>(`${prefix}/bias`, 1)
+    return { filters, bias }
+  }
+
+  function extractConvWithBatchNormParams(prefix: string): ConvWithBatchNorm {
+    const conv = extractConvParams(`${prefix}/conv`)
+    const bn = extractBatchNormParams(`${prefix}/bn`)
+    return { conv, bn }
+  }
+
+  const extractSeparableConvParams = loadSeparableConvParamsFactory(extractWeightEntry)
+
+  return {
+    extractConvParams,
+    extractConvWithBatchNormParams,
+    extractSeparableConvParams
+  }
+
+}
+
+export function extractParamsFromWeigthMap(
+  weightMap: tf.NamedTensorMap,
+  config: TinyYolov2Config
+): { params: TinyYolov2NetParams, paramMappings: ParamMapping[] } {
+
+  const paramMappings: ParamMapping[] = []
+
+  const {
+    extractConvParams,
+    extractConvWithBatchNormParams,
+    extractSeparableConvParams
+  } = extractorsFactory(weightMap, paramMappings)
+
+  let params: TinyYolov2NetParams
+
+  if (config.withSeparableConvs) {
+    const numFilters = (config.filterSizes && config.filterSizes.length || 9)
+    params = {
+      conv0: config.isFirstLayerConv2d ? extractConvParams('conv0') : extractSeparableConvParams('conv0'),
+      conv1: extractSeparableConvParams('conv1'),
+      conv2: extractSeparableConvParams('conv2'),
+      conv3: extractSeparableConvParams('conv3'),
+      conv4: extractSeparableConvParams('conv4'),
+      conv5: extractSeparableConvParams('conv5'),
+      conv6: numFilters > 7 ? extractSeparableConvParams('conv6') : undefined,
+      conv7: numFilters > 8 ? extractSeparableConvParams('conv7') : undefined,
+      conv8: extractConvParams('conv8')
+    }
+  } else {
+    params = {
+      conv0: extractConvWithBatchNormParams('conv0'),
+      conv1: extractConvWithBatchNormParams('conv1'),
+      conv2: extractConvWithBatchNormParams('conv2'),
+      conv3: extractConvWithBatchNormParams('conv3'),
+      conv4: extractConvWithBatchNormParams('conv4'),
+      conv5: extractConvWithBatchNormParams('conv5'),
+      conv6: extractConvWithBatchNormParams('conv6'),
+      conv7: extractConvWithBatchNormParams('conv7'),
+      conv8: extractConvParams('conv8')
+    }
+  }
+
+  disposeUnusedWeightTensors(weightMap, paramMappings)
+
+  return { params, paramMappings }
+}
\ No newline at end of file

src/tinyYolov2/index.ts

 import { TinyYolov2 } from './TinyYolov2';
 
+export * from './TinyYolov2Options';
+
+export * from './config'
+export * from './types'
+
 export { TinyYolov2 }
 
 export function createTinyYolov2(weights: Float32Array, withSeparableConvs: boolean = true) {

src/tinyYolov2/leaky.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+export function leaky(x: tf.Tensor4D): tf.Tensor4D {
+  return tf.tidy(() => {
+    const min = tf.mul(x, tf.scalar(0.10000000149011612))
+    return tf.add(tf.relu(tf.sub(x, min)), min)
+    //return tf.maximum(x, min)
+  })
+}
\ No newline at end of file

src/tinyYolov2/types.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { ConvParams } from '../common';
+import { SeparableConvParams } from '../common/types';
+
+export type BatchNorm = {
+  sub: tf.Tensor1D
+  truediv: tf.Tensor1D
+}
+
+export type ConvWithBatchNorm = {
+  conv: ConvParams
+  bn: BatchNorm
+}
+
+export type MobilenetParams = {
+  conv0: SeparableConvParams | ConvParams
+  conv1: SeparableConvParams
+  conv2: SeparableConvParams
+  conv3: SeparableConvParams
+  conv4: SeparableConvParams
+  conv5: SeparableConvParams
+  conv6?: SeparableConvParams
+  conv7?: SeparableConvParams
+  conv8: ConvParams
+}
+
+export type DefaultTinyYolov2NetParams = {
+  conv0: ConvWithBatchNorm
+  conv1: ConvWithBatchNorm
+  conv2: ConvWithBatchNorm
+  conv3: ConvWithBatchNorm
+  conv4: ConvWithBatchNorm
+  conv5: ConvWithBatchNorm
+  conv6: ConvWithBatchNorm
+  conv7: ConvWithBatchNorm
+  conv8: ConvParams
+}
+
+export type TinyYolov2NetParams = DefaultTinyYolov2NetParams | MobilenetParams
\ No newline at end of file

src/utils/index.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { Point } from '../classes';
+import { Dimensions, IDimensions } from '../classes/Dimensions';
+
+export function isTensor(tensor: any, dim: number) {
+  return tensor instanceof tf.Tensor && tensor.shape.length === dim
+}
+
+export function isTensor1D(tensor: any): tensor is tf.Tensor1D {
+  return isTensor(tensor, 1)
+}
+
+export function isTensor2D(tensor: any): tensor is tf.Tensor2D {
+  return isTensor(tensor, 2)
+}
+
+export function isTensor3D(tensor: any): tensor is tf.Tensor3D {
+  return isTensor(tensor, 3)
+}
+
+export function isTensor4D(tensor: any): tensor is tf.Tensor4D {
+  return isTensor(tensor, 4)
+}
+
+export function isFloat(num: number) {
+  return num % 1 !== 0
+}
+
+export function isEven(num: number) {
+  return num % 2 === 0
+}
+
+export function round(num: number, prec: number = 2) {
+  const f = Math.pow(10, prec)
+  return Math.floor(num * f) / f
+}
+
+export function isDimensions(obj: any): boolean {
+  return obj && obj.width && obj.height
+}
+
+export function computeReshapedDimensions({ width, height }: IDimensions, inputSize: number) {
+  const scale = inputSize / Math.max(height, width)
+  return new Dimensions(Math.round(width * scale), Math.round(height * scale))
+}
+
+export function getCenterPoint(pts: Point[]): Point {
+  return pts.reduce((sum, pt) => sum.add(pt), new Point(0, 0))
+    .div(new Point(pts.length, pts.length))
+}
+
+export function range(num: number, start: number, step: number): number[] {
+  return Array(num).fill(0).map((_, i) => start + (i * step))
+}
+
+export function isValidNumber(num: any) {
+  return !!num && num !== Infinity && num !== -Infinity && !isNaN(num) || num === 0
+}
+
+export function isValidProbablitiy(num: any) {
+  return isValidNumber(num) && 0 <= num && num <= 1.0
+}
\ No newline at end of file

src/xception/TinyXception.ts

 import * as tf from '@tensorflow/tfjs-core';
-import {
-  NetInput,
-  NeuralNetwork,
-  normalize,
-  range,
-  TfjsImageRecognitionBase,
-  TNetInput,
-  toNetInput,
-} from 'tfjs-image-recognition-base';
-
-import { depthwiseSeparableConv } from '../common/depthwiseSeparableConv';
-import { bgrToRgbTensor } from '../mtcnn/bgrToRgbTensor';
+
+import { ConvParams, depthwiseSeparableConv } from '../common';
+import { NetInput, TNetInput, toNetInput } from '../dom';
+import { NeuralNetwork } from '../NeuralNetwork';
+import { normalize } from '../ops';
+import { range } from '../utils';
 import { extractParams } from './extractParams';
 import { extractParamsFromWeigthMap } from './extractParamsFromWeigthMap';
 import { MainBlockParams, ReductionBlockParams, TinyXceptionParams } from './types';
 
-function conv(x: tf.Tensor4D, params: TfjsImageRecognitionBase.ConvParams, stride: [number, number]): tf.Tensor4D {
+function conv(x: tf.Tensor4D, params: ConvParams, stride: [number, number]): tf.Tensor4D {
   return tf.add(tf.conv2d(x, params.filters, stride, 'same'), params.bias)
 }
 

src/xception/extractParams.ts

-import { range, TfjsImageRecognitionBase } from 'tfjs-image-recognition-base';
-
+import { extractConvParamsFactory, extractSeparableConvParamsFactory, extractWeightsFactory } from '../common';
+import { ExtractWeightsFunction, ParamMapping } from '../common/types';
+import { range } from '../utils';
 import { MainBlockParams, ReductionBlockParams, TinyXceptionParams } from './types';
 
-function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeightsFunction, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function extractorsFactory(extractWeights: ExtractWeightsFunction, paramMappings: ParamMapping[]) {
 
-  const extractConvParams = TfjsImageRecognitionBase.extractConvParamsFactory(extractWeights, paramMappings)
-  const extractSeparableConvParams = TfjsImageRecognitionBase.extractSeparableConvParamsFactory(extractWeights, paramMappings)
+  const extractConvParams = extractConvParamsFactory(extractWeights, paramMappings)
+  const extractSeparableConvParams = extractSeparableConvParamsFactory(extractWeights, paramMappings)
 
   function extractReductionBlockParams(channelsIn: number, channelsOut: number, mappedPrefix: string): ReductionBlockParams {
 
@@ -34,14 +35,14 @@ function extractorsFactory(extractWeights: TfjsImageRecognitionBase.ExtractWeigh
 
 }
 
-export function extractParams(weights: Float32Array, numMainBlocks: number): { params: TinyXceptionParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+export function extractParams(weights: Float32Array, numMainBlocks: number): { params: TinyXceptionParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractWeights,
     getRemainingWeights
-  } = TfjsImageRecognitionBase.extractWeightsFactory(weights)
+  } = extractWeightsFactory(weights)
 
   const {
     extractConvParams,

src/xception/extractParamsFromWeigthMap.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { TfjsImageRecognitionBase, range } from 'tfjs-image-recognition-base';
 
+import {
+  disposeUnusedWeightTensors,
+  extractWeightEntryFactory,
+  loadSeparableConvParamsFactory,
+  ParamMapping,
+} from '../common';
 import { loadConvParamsFactory } from '../common/loadConvParamsFactory';
+import { range } from '../utils';
 import { MainBlockParams, ReductionBlockParams, TinyXceptionParams } from './types';
 
-function loadParamsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBase.ParamMapping[]) {
+function loadParamsFactory(weightMap: any, paramMappings: ParamMapping[]) {
 
-  const extractWeightEntry = TfjsImageRecognitionBase.extractWeightEntryFactory(weightMap, paramMappings)
+  const extractWeightEntry = extractWeightEntryFactory(weightMap, paramMappings)
 
   const extractConvParams = loadConvParamsFactory(extractWeightEntry)
-  const extractSeparableConvParams = TfjsImageRecognitionBase.loadSeparableConvParamsFactory(extractWeightEntry)
+  const extractSeparableConvParams = loadSeparableConvParamsFactory(extractWeightEntry)
 
   function extractReductionBlockParams(mappedPrefix: string): ReductionBlockParams {
 
@@ -40,9 +46,9 @@ function loadParamsFactory(weightMap: any, paramMappings: TfjsImageRecognitionBa
 export function extractParamsFromWeigthMap(
   weightMap: tf.NamedTensorMap,
   numMainBlocks: number
-): { params: TinyXceptionParams, paramMappings: TfjsImageRecognitionBase.ParamMapping[] } {
+): { params: TinyXceptionParams, paramMappings: ParamMapping[] } {
 
-  const paramMappings: TfjsImageRecognitionBase.ParamMapping[] = []
+  const paramMappings: ParamMapping[] = []
 
   const {
     extractConvParams,
@@ -74,7 +80,7 @@ export function extractParamsFromWeigthMap(
     separable_conv: exit_flow_separable_conv
   }
 
-  TfjsImageRecognitionBase.disposeUnusedWeightTensors(weightMap, paramMappings)
+  disposeUnusedWeightTensors(weightMap, paramMappings)
 
   return { params: { entry_flow, middle_flow, exit_flow }, paramMappings }
 }
\ No newline at end of file

src/xception/types.ts

-import { TfjsImageRecognitionBase } from "tfjs-image-recognition-base";
+import { ConvParams, SeparableConvParams } from '../common';
 
 export type ReductionBlockParams = {
-  separable_conv0: TfjsImageRecognitionBase.SeparableConvParams
-  separable_conv1: TfjsImageRecognitionBase.SeparableConvParams
-  expansion_conv: TfjsImageRecognitionBase.ConvParams
+  separable_conv0: SeparableConvParams
+  separable_conv1: SeparableConvParams
+  expansion_conv: ConvParams
 }
 
 export type MainBlockParams = {
-  separable_conv0: TfjsImageRecognitionBase.SeparableConvParams
-  separable_conv1: TfjsImageRecognitionBase.SeparableConvParams
-  separable_conv2: TfjsImageRecognitionBase.SeparableConvParams
+  separable_conv0: SeparableConvParams
+  separable_conv1: SeparableConvParams
+  separable_conv2: SeparableConvParams
 }
 
 export type TinyXceptionParams = {
   entry_flow: {
-    conv_in: TfjsImageRecognitionBase.ConvParams
+    conv_in: ConvParams
     reduction_block_0: ReductionBlockParams
     reduction_block_1: ReductionBlockParams
   }
   middle_flow: any,
   exit_flow: {
     reduction_block: ReductionBlockParams
-    separable_conv: TfjsImageRecognitionBase.SeparableConvParams
+    separable_conv: SeparableConvParams
   }
 }
\ No newline at end of file

test/Environment.ts

-import { NeuralNetwork } from 'tfjs-image-recognition-base';
+import { NeuralNetwork } from '../src';
 
 export type TestEnv = {
   loadImage: (uri: string) => Promise<HTMLImageElement>

test/NeuralNetwork.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { NeuralNetwork } from '../src';
+
+class FakeNeuralNetwork extends NeuralNetwork<any> {
+  constructor(
+    convFilter: tf.Tensor = tf.tensor(0),
+    convBias: tf.Tensor = tf.tensor(0),
+    fcWeights: tf.Tensor = tf.tensor(0)
+  ) {
+    super('FakeNeuralNetwork')
+    this._params = {
+      conv: {
+        filter: convFilter,
+        bias: convBias,
+      },
+      fc: fcWeights
+    }
+
+    this._paramMappings = [
+      { originalPath: 'conv2d/filter', paramPath: 'conv/filter' },
+      { originalPath: 'conv2d/bias', paramPath: 'conv/bias' },
+      { originalPath: 'dense/weights', paramPath: 'fc' }
+    ]
+  }
+
+  protected getDefaultModelName(): any {
+    throw new Error('FakeNeuralNetwork - getDefaultModelName not implemented')
+  }
+
+  protected extractParams(_: any): any {
+    throw new Error('FakeNeuralNetwork - extractParams not implemented')
+  }
+
+  protected extractParamsFromWeigthMap(_: any): any {
+    throw new Error('FakeNeuralNetwork - extractParamsFromWeigthMap not implemented')
+  }
+}
+
+describe('NeuralNetwork', () => {
+
+  describe('getParamFromPath', () => {
+
+    it('returns correct params', () => tf.tidy(() => {
+      const convFilter = tf.tensor(0)
+      const convBias = tf.tensor(0)
+      const fcWeights = tf.tensor(0)
+      const net = new FakeNeuralNetwork(convFilter, convBias, fcWeights)
+
+      expect(net.getParamFromPath('conv/filter')).toEqual(convFilter)
+      expect(net.getParamFromPath('conv/bias')).toEqual(convBias)
+      expect(net.getParamFromPath('fc')).toEqual(fcWeights)
+    }))
+
+    it('throws if param is not a tensor', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork(null as any)
+      const fakePath = 'conv/filter'
+
+      expect(
+        () => net.getParamFromPath(fakePath)
+      ).toThrowError(`traversePropertyPath - parameter is not a tensor, for path ${fakePath}`)
+    }))
+
+    it('throws if key path invalid', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork()
+      const fakePath = 'conv2d/foo'
+
+      expect(
+        () => net.getParamFromPath(fakePath)
+      ).toThrowError(`traversePropertyPath - object does not have property conv2d, for path ${fakePath}`)
+    }))
+
+  })
+
+  describe('reassignParamFromPath', () => {
+
+    it('sets correct params', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork()
+
+      const convFilter = tf.tensor(0)
+      const convBias = tf.tensor(0)
+      const fcWeights = tf.tensor(0)
+      net.reassignParamFromPath('conv/filter', convFilter)
+      net.reassignParamFromPath('conv/bias', convBias)
+      net.reassignParamFromPath('fc', fcWeights)
+
+      expect(net.params.conv.filter).toEqual(convFilter)
+      expect(net.params.conv.bias).toEqual(convBias)
+      expect(net.params.fc).toEqual(fcWeights)
+    }))
+
+    it('throws if param is not a tensor', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork(null as any)
+      const fakePath = 'conv/filter'
+
+      expect(
+        () => net.reassignParamFromPath(fakePath, tf.tensor(0))
+      ).toThrowError(`traversePropertyPath - parameter is not a tensor, for path ${fakePath}`)
+    }))
+
+    it('throws if key path invalid', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork()
+      const fakePath = 'conv2d/foo'
+
+      expect(
+        () => net.reassignParamFromPath(fakePath, tf.tensor(0))
+      ).toThrowError(`traversePropertyPath - object does not have property conv2d, for path ${fakePath}`)
+    }))
+
+  })
+
+  describe('getParamList', () => {
+
+    it('returns param tensors with path', () => tf.tidy(() => {
+      const convFilter = tf.tensor(0)
+      const convBias = tf.tensor(0)
+      const fcWeights = tf.tensor(0)
+      const net = new FakeNeuralNetwork(convFilter, convBias, fcWeights)
+
+      const paramList = net.getParamList()
+
+      expect(paramList.length).toEqual(3)
+      expect(paramList[0].path).toEqual('conv/filter')
+      expect(paramList[1].path).toEqual('conv/bias')
+      expect(paramList[2].path).toEqual('fc')
+      expect(paramList[0].tensor).toEqual(convFilter)
+      expect(paramList[1].tensor).toEqual(convBias)
+      expect(paramList[2].tensor).toEqual(fcWeights)
+    }))
+
+  })
+
+  describe('getFrozenParams', () => {
+
+    it('returns all frozen params', () => tf.tidy(() => {
+      const convFilter = tf.tensor(0)
+      const convBias = tf.tensor(0)
+      const fcWeights = tf.variable(tf.scalar(0))
+      const net = new FakeNeuralNetwork(convFilter, convBias, fcWeights)
+
+      const frozenParams = net.getFrozenParams()
+
+      expect(frozenParams.length).toEqual(2)
+      expect(frozenParams[0].path).toEqual('conv/filter')
+      expect(frozenParams[1].path).toEqual('conv/bias')
+      expect(frozenParams[0].tensor).toEqual(convFilter)
+      expect(frozenParams[1].tensor).toEqual(convBias)
+    }))
+
+  })
+
+  describe('getTrainableParams', () => {
+
+    it('returns all trainable params', () => tf.tidy(() => {
+      const convFilter = tf.variable(tf.scalar(0))
+      const convBias = tf.variable(tf.scalar(0))
+      const fcWeights = tf.tensor(0)
+      const net = new FakeNeuralNetwork(convFilter, convBias, fcWeights)
+
+      const trainableParams = net.getTrainableParams()
+
+      expect(trainableParams.length).toEqual(2)
+      expect(trainableParams[0].path).toEqual('conv/filter')
+      expect(trainableParams[1].path).toEqual('conv/bias')
+      expect(trainableParams[0].tensor).toEqual(convFilter)
+      expect(trainableParams[1].tensor).toEqual(convBias)
+    }))
+
+  })
+
+  describe('dispose', () => {
+
+    it('disposes all param tensors', () => tf.tidy(() => {
+      const numTensors = tf.memory().numTensors
+      const net = new FakeNeuralNetwork()
+
+      net.dispose()
+
+      expect(net.params).toBe(undefined)
+      expect(tf.memory().numTensors - numTensors).toEqual(0)
+    }))
+
+  })
+
+  describe('variable', () => {
+
+    it('make all param tensors trainable', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork()
+
+      net.variable()
+
+      expect(net.params.conv.filter instanceof tf.Variable).toBe(true)
+      expect(net.params.conv.bias instanceof tf.Variable).toBe(true)
+      expect(net.params.fc instanceof tf.Variable).toBe(true)
+    }))
+
+    it('disposes old tensors', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork()
+      const numTensors = tf.memory().numTensors
+
+      net.variable()
+
+      expect(tf.memory().numTensors - numTensors).toEqual(0)
+    }))
+
+  })
+
+  describe('freeze', () => {
+
+    it('freezes all param variables', () => tf.tidy(() => {
+      const net = new FakeNeuralNetwork(
+        tf.variable(tf.scalar(0)),
+        tf.variable(tf.scalar(0)),
+        tf.variable(tf.scalar(0))
+      )
+
+      net.freeze()
+
+      expect(net.params.conv.filter instanceof tf.Variable).toBe(false)
+      expect(net.params.conv.bias instanceof tf.Variable).toBe(false)
+      expect(net.params.fc instanceof tf.Variable).toBe(false)
+    }))
+
+    it('disposes old tensors', () => () => {
+      const net = new FakeNeuralNetwork(
+        tf.variable(tf.scalar(0)),
+        tf.variable(tf.scalar(0)),
+        tf.variable(tf.scalar(0))
+      )
+      const numTensors = tf.memory().numTensors
+
+      net.freeze()
+
+      expect(tf.memory().numTensors - numTensors).toEqual(0)
+    })
+
+  })
+
+})

test/data/boxes.json

+[{"predictions":[],"groundTruth":[{"label":14,"x":174,"y":101,"width":175,"height":250}]},{"predictions":[{"score":0.7691803822132828,"classScore":0.7687403453616095,"label":14,"x":174.80739067078562,"y":101.24652901136822,"width":107.21253846292248,"height":130.36960304401345},{"score":0.6650128354611389,"classScore":0.6648250709475522,"label":0,"x":88.00403034948206,"y":57.28796517004654,"width":346.8964870013924,"height":154.64824963698928}],"groundTruth":[{"label":0,"x":104,"y":78,"width":271,"height":105},{"label":0,"x":133,"y":88,"width":64,"height":35},{"label":14,"x":195,"y":180,"width":18,"height":49},{"label":14,"x":26,"y":189.00000000000003,"width":18,"height":49}]},{"predictions":[{"score":0.9796686439333566,"classScore":0.9796404986030244,"label":0,"x":0.7092259472062379,"y":112.93828229104882,"width":503.36772784059957,"height":125.40083611003898}],"groundTruth":[{"label":0,"x":9,"y":107,"width":490,"height":156},{"label":0,"x":421,"y":200.00000000000003,"width":61,"height":26},{"label":0,"x":325,"y":188,"width":86,"height":35}]},{"predictions":[{"score":0.6555995462593681,"classScore":0.6555995462593681,"label":19,"x":168.43528196184994,"y":103.87677334009256,"width":142.1917672089718,"height":152.21249718337248}],"groundTruth":[{"label":19,"x":156,"y":89,"width":188,"height":190.00000000000003}]},{"predictions":[{"score":0.7654244759562995,"classScore":0.7653893919815798,"label":18,"x":4.657064294604191,"y":19.396089415586324,"width":244.37305501026242,"height":295.7467446833704}],"groundTruth":[{"label":18,"x":263,"y":32,"width":237,"height":263},{"label":18,"x":1,"y":36,"width":234,"height":263}]},{"predictions":[],"groundTruth":[{"label":3,"x":274,"y":11,"width":163,"height":268},{"label":3,"x":184,"y":213.99999999999997,"width":97,"height":38}]},{"predictions":[{"score":0.6628550520040408,"classScore":0.6247242104538263,"label":11,"x":72.80442223810171,"y":96.69588943582755,"width":348.97090277779495,"height":201.39196990473874}],"groundTruth":[{"label":11,"x":123,"y":114.99999999999999,"width":256,"height":160},{"label":8,"x":75,"y":1,"width":353,"height":374}]},{"predictions":[{"score":0.9813790697369239,"classScore":0.9785855359209782,"label":2,"x":27.170210865162982,"y":71.72488948863362,"width":236.38816975654763,"height":288.48520937333933}],"groundTruth":[{"label":2,"x":27,"y":45,"width":239,"height":330}]},{"predictions":[{"score":0.994037798212397,"classScore":0.994037798212397,"label":19,"x":221.0525346736894,"y":17.11757711236289,"width":270.7030429591586,"height":238.46462518346138},{"score":0.9744214784574556,"classScore":0.9744214784574556,"label":19,"x":-8.843471823254735,"y":26.29916942997549,"width":300.1775416643035,"height":221.028915747058}],"groundTruth":[{"label":19,"x":251,"y":28.000000000000004,"width":224,"height":239},{"label":19,"x":22,"y":28.000000000000004,"width":229,"height":245}]},{"predictions":[{"score":0.9887495371422556,"classScore":0.6017762085053843,"label":16,"x":-19.909558311815623,"y":21.542763412106453,"width":414.6293719402734,"height":308.439681183784}],"groundTruth":[{"label":18,"x":1,"y":26,"width":357,"height":314}]},{"predictions":[{"score":0.9098243761845755,"classScore":0.9012128531842766,"label":14,"x":71.38750749942446,"y":54.7591392923992,"width":223.77602020255526,"height":366.18752833394205}],"groundTruth":[{"label":1,"x":70,"y":202,"width":185,"height":298},{"label":1,"x":251,"y":242,"width":83,"height":258},{"label":1,"x":1,"y":144,"width":66,"height":292},{"label":14,"x":1,"y":1,"width":65,"height":362},{"label":14,"x":74,"y":1,"width":198,"height":461},{"label":14,"x":251.99999999999997,"y":19,"width":82,"height":468}]},{"predictions":[],"groundTruth":[{"label":4,"x":87,"y":100,"width":22,"height":65},{"label":4,"x":41,"y":114,"width":32,"height":67},{"label":4,"x":324,"y":148,"width":28,"height":58},{"label":4,"x":426,"y":157,"width":17,"height":38},{"label":14,"x":3,"y":91,"width":40,"height":114.99999999999999},{"label":14,"x":4,"y":28.000000000000004,"width":457,"height":344}]},{"predictions":[{"score":0.7958273091892446,"classScore":0.7818440392104067,"label":16,"x":56.79348195528594,"y":31.70232255780979,"width":313.6508630022377,"height":233.41861230559402}],"groundTruth":[{"label":16,"x":25,"y":34,"width":394,"height":237}]},{"predictions":[{"score":0.9201249561107422,"classScore":0.9190199099731305,"label":19,"x":-26.03511346499099,"y":86.38443500732924,"width":297.6374106029245,"height":254.9645026012354}],"groundTruth":[{"label":19,"x":1,"y":95.00000000000001,"width":239,"height":241.00000000000003}]},{"predictions":[{"score":0.9161253701392962,"classScore":0.9058719094189636,"label":3,"x":32.61301936266575,"y":115.64641272085254,"width":124.58349953920403,"height":101.73433847921152},{"score":0.8166243906870652,"classScore":0.8145709063786849,"label":3,"x":377.0253934850286,"y":187.73553354763786,"width":117.8902182556456,"height":98.97330181906452},{"score":0.682690236232842,"classScore":0.6699430513504846,"label":3,"x":235.7334992854692,"y":132.159729119493,"width":107.03585321829007,"height":91.19282688695355}],"groundTruth":[{"label":3,"x":356,"y":183,"width":144,"height":97},{"label":3,"x":60,"y":109,"width":82,"height":104},{"label":3,"x":246,"y":134,"width":102,"height":83}]},{"predictions":[],"groundTruth":[{"label":0,"x":181,"y":127,"width":93,"height":66}]},{"predictions":[{"score":0.7002533672629911,"classScore":0.6986950241075993,"label":10,"x":15.059672065713514,"y":183.7041276920172,"width":442.0153237833172,"height":187.85830177400535}],"groundTruth":[{"label":10,"x":1,"y":170,"width":473,"height":205},{"label":4,"x":97,"y":124,"width":53,"height":173}]},{"predictions":[{"score":0.8785236648425084,"classScore":0.8785169098747734,"label":0,"x":5.851446853403164,"y":97.19943552245653,"width":498.761858292814,"height":123.84420987488453},{"score":0.6115907530866461,"classScore":0.50117202877969,"label":13,"x":53.19739845463284,"y":133.81894255992148,"width":226.98909377605403,"height":110.1798238654668}],"groundTruth":[{"label":0,"x":8,"y":96,"width":483,"height":136}]},{"predictions":[{"score":0.944397664196853,"classScore":0.944397664196853,"label":14,"x":-4.598343170909137,"y":82.8255836803386,"width":335.55829964238654,"height":426.5030936299306}],"groundTruth":[{"label":14,"x":25,"y":71,"width":279,"height":429}]},{"predictions":[{"score":0.8280960338115692,"classScore":0.8280960338115692,"label":14,"x":-1.5531990164389664,"y":-25.619537343989954,"width":343.5896574164529,"height":418.021221131633}],"groundTruth":[{"label":14,"x":277,"y":3,"width":223,"height":372},{"label":14,"x":12,"y":3,"width":293,"height":372}]},{"predictions":[],"groundTruth":[{"label":12,"x":54,"y":50,"width":231,"height":212}]},{"predictions":[],"groundTruth":[{"label":18,"x":1,"y":39,"width":366,"height":231}]},{"predictions":[{"score":0.9923042836360139,"classScore":0.9923042836360139,"label":14,"x":142.7915727970977,"y":103.88205108871067,"width":355.23004763212964,"height":246.3305211907232}],"groundTruth":[{"label":4,"x":124,"y":107,"width":106,"height":236},{"label":14,"x":137,"y":78,"width":360,"height":297},{"label":14,"x":89,"y":201.99999999999997,"width":40,"height":45},{"label":14,"x":72,"y":209,"width":39,"height":50}]},{"predictions":[{"score":0.7524108551868512,"classScore":0.7511045464768792,"label":2,"x":118.02073153439419,"y":100.075300105594,"width":422.1093645713623,"height":236.96557132518132},{"score":0.6675556654678318,"classScore":0.6547138694454316,"label":2,"x":73.15242855329525,"y":27.560020743027827,"width":138.40089468837368,"height":233.12522715598948}],"groundTruth":[{"label":2,"x":59,"y":15,"width":161,"height":224.00000000000003},{"label":2,"x":161,"y":122,"width":339,"height":211}]},{"predictions":[{"score":0.8762162584051255,"classScore":0.8753776565500138,"label":13,"x":316.9200075142487,"y":15.978265485680515,"width":176.55547521701214,"height":162.79950103493826},{"score":0.7377645066274695,"classScore":0.7377447182413641,"label":13,"x":4.093778050646285,"y":53.134905261691884,"width":484.3047072124648,"height":242.59882876631897}],"groundTruth":[{"label":13,"x":54,"y":25,"width":400,"height":290},{"label":13,"x":318,"y":37,"width":171,"height":124},{"label":14,"x":369,"y":1,"width":89,"height":129}]},{"predictions":[{"score":0.6987292311899743,"classScore":0.6981430425202332,"label":14,"x":190.34686802768005,"y":49.30728986502912,"width":125.24766886822394,"height":156.9966166652352}],"groundTruth":[{"label":12,"x":100,"y":96,"width":255,"height":228},{"label":14,"x":198,"y":58,"width":88,"height":139}]},{"predictions":[{"score":0.7222071563927183,"classScore":0.722190583335829,"label":14,"x":334.27473386380956,"y":84.92576867213481,"width":130.96630892426847,"height":266.5457370602929}],"groundTruth":[{"label":14,"x":51,"y":80,"width":71,"height":186},{"label":14,"x":367,"y":92,"width":94,"height":223}]},{"predictions":[],"groundTruth":[{"label":17,"x":50,"y":121,"width":245,"height":217},{"label":8,"x":474,"y":140,"width":26,"height":144}]},{"predictions":[{"score":0.9539370238688553,"classScore":0.9261299783258337,"label":9,"x":53.1153804004625,"y":182.89761062792732,"width":186.2471532372765,"height":135.7749268658459}],"groundTruth":[{"label":9,"x":71,"y":252,"width":145,"height":62},{"label":9,"x":58,"y":202,"width":183,"height":93}]},{"predictions":[{"score":0.9517353634593023,"classScore":0.9517353634593023,"label":14,"x":295.17106719541056,"y":154.82150461830355,"width":110.50320081452242,"height":211.90106600291168},{"score":0.8956736871398788,"classScore":0.858500424420289,"label":0,"x":37.63960073319886,"y":77.60772153913085,"width":428.89316111622963,"height":269.4121395866779},{"score":0.8835170300270249,"classScore":0.815483197617464,"label":14,"x":2.93943475182442,"y":180.41620151640234,"width":112.98117869748658,"height":203.70600878430784}],"groundTruth":[{"label":14,"x":293,"y":162,"width":126,"height":212.99999999999997},{"label":14,"x":114,"y":165,"width":114,"height":208},{"label":14,"x":5,"y":172,"width":111,"height":201},{"label":0,"x":2,"y":68,"width":484,"height":257}]},{"predictions":[{"score":0.9625422337065817,"classScore":0.5203173443928317,"label":5,"x":-21.917914358316192,"y":20.243043364143507,"width":496.5538125661707,"height":282.63866737628933}],"groundTruth":[{"label":9,"x":2,"y":2,"width":457,"height":332}]},{"predictions":[],"groundTruth":[{"label":9,"x":341,"y":102,"width":159,"height":304},{"label":14,"x":36,"y":34,"width":43,"height":89},{"label":14,"x":2,"y":44,"width":34,"height":76},{"label":14,"x":29,"y":42,"width":12,"height":22}]},{"predictions":[],"groundTruth":[{"label":1,"x":1,"y":198.99999999999997,"width":39,"height":94},{"label":14,"x":1,"y":85,"width":28,"height":194},{"label":6,"x":64,"y":164,"width":87,"height":46},{"label":6,"x":147,"y":167,"width":39,"height":33}]},{"predictions":[],"groundTruth":[{"label":7,"x":124,"y":68,"width":195,"height":242}]},{"predictions":[],"groundTruth":[{"label":3,"x":290,"y":143,"width":210,"height":107},{"label":6,"x":422,"y":237,"width":39,"height":18}]},{"predictions":[{"score":0.8962182918520621,"classScore":0.8930069154800067,"label":7,"x":-4.008327108320088,"y":70.37482224480168,"width":352.62230878734033,"height":434.67563823380846}],"groundTruth":[{"label":7,"x":1,"y":49,"width":340,"height":450}]},{"predictions":[{"score":0.8076964825223881,"classScore":0.8067911153833874,"label":19,"x":94.5857889149978,"y":15.249397515597346,"width":312.17182522265784,"height":232.76671025556075}],"groundTruth":[{"label":4,"x":36,"y":250,"width":43,"height":104},{"label":19,"x":160,"y":26,"width":211,"height":214.99999999999997}]},{"predictions":[],"groundTruth":[{"label":1,"x":1,"y":178,"width":234,"height":153},{"label":1,"x":8,"y":134,"width":218,"height":163}]},{"predictions":[{"score":0.8167527436184676,"classScore":0.8150299759023477,"label":17,"x":324.627645002174,"y":209.56248991054514,"width":178.6013940037733,"height":160.7920031534356}],"groundTruth":[{"label":1,"x":31,"y":74,"width":338,"height":272},{"label":17,"x":344,"y":207.00000000000003,"width":156,"height":167}]},{"predictions":[{"score":0.9407357523667703,"classScore":0.8181894101476949,"label":3,"x":-12.598615939355684,"y":77.28791949345168,"width":451.4040682919014,"height":104.75131312062308}],"groundTruth":[{"label":18,"x":34,"y":84,"width":466,"height":100},{"label":14,"x":87,"y":100,"width":20,"height":24}]},{"predictions":[{"score":0.752703742634889,"classScore":0.7477982878493062,"label":18,"x":81.62897083839832,"y":109.91140563880278,"width":329.63940263951787,"height":207.33306830787868}],"groundTruth":[{"label":18,"x":79,"y":96,"width":313,"height":204}]},{"predictions":[{"score":0.8644956045627069,"classScore":0.8612763441448485,"label":2,"x":132.49570312762185,"y":53.63054832818552,"width":361.1467077508502,"height":308.41826405754813}],"groundTruth":[{"label":2,"x":135,"y":46,"width":365,"height":328},{"label":2,"x":124,"y":146,"width":241,"height":229}]},{"predictions":[{"score":0.7878781057167802,"classScore":0.7875985926864165,"label":5,"x":-52.21908643538992,"y":93.02825869092997,"width":459.81048644960373,"height":189.74556017463232}],"groundTruth":[{"label":14,"x":394,"y":199,"width":10,"height":24},{"label":14,"x":424,"y":199,"width":12,"height":20.999999999999996},{"label":14,"x":434,"y":195.99999999999997,"width":10,"height":24},{"label":14,"x":443,"y":195,"width":9,"height":25},{"label":5,"x":29,"y":113,"width":324,"height":153},{"label":0,"x":328,"y":86,"width":146,"height":106},{"label":6,"x":2,"y":213,"width":26,"height":22}]},{"predictions":[{"score":0.743069405764908,"classScore":0.6373241114093381,"label":15,"x":78.30857626719687,"y":92.88842351228324,"width":90.06631442691248,"height":108.2613080721939},{"score":0.6118246612799929,"classScore":0.5526098988725323,"label":17,"x":123.27749832194552,"y":160.7244957117589,"width":181.02068890434398,"height":101.80487324073387}],"groundTruth":[{"label":17,"x":119,"y":177,"width":212,"height":100.99999999999999},{"label":15,"x":65,"y":90,"width":105,"height":107},{"label":8,"x":133,"y":159,"width":46,"height":38}]},{"predictions":[{"score":0.8145260519407391,"classScore":0.81435355543938,"label":5,"x":7.672527282127808,"y":32.34257681806332,"width":382.86829973169773,"height":419.57299155829344}],"groundTruth":[{"label":5,"x":12,"y":2,"width":405,"height":465},{"label":6,"x":7.000000000000001,"y":39,"width":56.00000000000001,"height":55},{"label":6,"x":362,"y":24,"width":60,"height":29},{"label":6,"x":376,"y":36,"width":46,"height":45},{"label":6,"x":373,"y":68,"width":49,"height":40},{"label":6,"x":376,"y":98,"width":46,"height":112}]},{"predictions":[{"score":0.9424702777106074,"classScore":0.9424702777106074,"label":14,"x":36.45312644547885,"y":6.678077418736085,"width":290.71862354973774,"height":363.4338941971187}],"groundTruth":[{"label":14,"x":51,"y":2,"width":274,"height":373}]},{"predictions":[],"groundTruth":[{"label":16,"x":59,"y":94,"width":413,"height":176}]},{"predictions":[{"score":0.573654529177009,"classScore":0.3838500178452209,"label":3,"x":69.91134561978498,"y":181.02671387178248,"width":207.20328569794563,"height":137.88072756263017}],"groundTruth":[{"label":3,"x":80,"y":168,"width":174,"height":144},{"label":3,"x":184,"y":160,"width":288,"height":109}]},{"predictions":[{"score":0.5588748494638404,"classScore":0.5467246329459337,"label":11,"x":201.5844166952682,"y":114.26291552404027,"width":127.83844565704325,"height":148.51310204029576}],"groundTruth":[{"label":11,"x":217,"y":118,"width":88,"height":145}]},{"predictions":[{"score":0.7826680552768471,"classScore":0.3132984973801781,"label":8,"x":110.13095572354015,"y":296.0810272956989,"width":136.69938049361562,"height":149.26902152141102}],"groundTruth":[{"label":5,"x":89,"y":323,"width":184,"height":113},{"label":6,"x":272,"y":369,"width":51.00000000000001,"height":19},{"label":6,"x":2,"y":380,"width":87,"height":55},{"label":14,"x":333,"y":356,"width":25,"height":82},{"label":14,"x":319,"y":357,"width":22,"height":67},{"label":14,"x":354,"y":363,"width":21,"height":51},{"label":6,"x":48.99999999999999,"y":367,"width":41,"height":22},{"label":6,"x":113,"y":138,"width":47,"height":26}]}]
\ No newline at end of file

test/env.node.ts

 import * as fs from 'fs';
 import * as path from 'path';
-import { NeuralNetwork } from 'tfjs-image-recognition-base';
 
-import { env } from '../src';
+import { env, NeuralNetwork } from '../src';
 import { TestEnv } from './Environment';
 
 require('@tensorflow/tfjs-node')

test/env.ts

 import * as tf from '@tensorflow/tfjs-core';
-import { fetchNetWeights, NeuralNetwork } from 'tfjs-image-recognition-base';
 
-import { fetchImage, fetchJson } from '../src';
+import { fetchImage, fetchJson, fetchNetWeights, NeuralNetwork } from '../src';
 import { TestEnv } from './Environment';
 
 jasmine.DEFAULT_TIMEOUT_INTERVAL = 60000

test/tests/faceLandmarkNet/faceLandmark68Net.uncompressed.test.ts → test/tests-legacy/faceLandmark68Net.uncompressed.test.ts

-import { Point } from '../../../src';
-import { FaceLandmarks68 } from '../../../src/classes/FaceLandmarks68';
-import { getTestEnv } from '../../env';
-import { describeWithBackend, describeWithNets, expectPointClose } from '../../utils';
+import { FaceLandmarks68, Point } from '../../src';
+import { getTestEnv } from '../env';
+import { describeWithBackend, describeWithNets, expectPointClose } from '../utils';
 
 describeWithBackend('faceLandmark68Net, uncompressed', () => {
 

test/tests/faceLandmarkNet/faceLandmark68TinyNet.uncompressed.test.ts → test/tests-legacy/faceLandmark68TinyNet.uncompressed.test.ts

-import { Point } from '../../../src';
-import { FaceLandmarks68 } from '../../../src/classes/FaceLandmarks68';
-import { getTestEnv } from '../../env';
-import { describeWithBackend, describeWithNets, expectPointClose } from '../../utils';
+import { Point } from '../../src';
+import { FaceLandmarks68 } from '../../src/classes/FaceLandmarks68';
+import { getTestEnv } from '../env';
+import { describeWithBackend, describeWithNets, expectPointClose } from '../utils';
 
 describeWithBackend('faceLandmark68TinyNet, uncompressed', () => {
 

test/tests/faceRecognitionNet/faceRecognitionNet.uncompressed.test.ts → test/tests-legacy/faceRecognitionNet.uncompressed.test.ts

-import { createCanvasFromMedia } from '../../../src';
-import { euclideanDistance } from '../../../src/euclideanDistance';
-import { getTestEnv } from '../../env';
-import { describeWithBackend, describeWithNets } from '../../utils';
+import { createCanvasFromMedia, euclideanDistance } from '../../src';
+import { getTestEnv } from '../env';
+import { describeWithBackend, describeWithNets } from '../utils';
 
 describeWithBackend('faceRecognitionNet, uncompressed', () => {
 

test/tests/ssdMobilenetv1/ssdMobilenetv1.locateFaces.uncompressed.test.ts → test/tests-legacy/ssdMobilenetv1.locateFaces.uncompressed.test.ts

-import * as faceapi from '../../../src';
-import { getTestEnv } from '../../env';
-import { expectFaceDetections } from '../../expectFaceDetections';
-import { describeWithBackend, describeWithNets } from '../../utils';
-import { expectedSsdBoxes } from './expectedBoxes';
+import * as faceapi from '../../src';
+import { getTestEnv } from '../env';
+import { expectFaceDetections } from '../expectFaceDetections';
+import { describeWithBackend, describeWithNets } from '../utils';
+import { expectedSsdBoxes } from '../tests/ssdMobilenetv1/expectedBoxes';
 
 describeWithBackend('ssdMobilenetv1.locateFaces, uncompressed', () => {
 

test/tests/classes/BoundingBox.test.ts

+import { BoundingBox } from '../../../src/classes/BoundingBox';
+
+describe('BoundingBox', () => {
+
+  describe('constructor', () => {
+
+    it('properties', () => {
+      const box = new BoundingBox(5, 10, 15, 20)
+      expect(box.left).toEqual(5)
+      expect(box.x).toEqual(5)
+      expect(box.top).toEqual(10)
+      expect(box.y).toEqual(10)
+      expect(box.right).toEqual(15)
+      expect(box.bottom).toEqual(20)
+      expect(box.width).toEqual(10)
+      expect(box.height).toEqual(10)
+      expect(box.area).toEqual(100)
+    })
+
+  })
+
+})
\ No newline at end of file

test/tests/classes/Box.test.ts

+import { Box } from '../../../src/classes/Box';
+
+describe('BoundingBox', () => {
+
+  describe('constructor', () => {
+
+    describe('from IBoundingBox', () => {
+
+      it('properties', () => {
+
+        const box = new Box({ left: 5, top: 10, right: 15, bottom: 20 })
+
+        expect(box.left).toEqual(5)
+        expect(box.x).toEqual(5)
+        expect(box.top).toEqual(10)
+        expect(box.y).toEqual(10)
+        expect(box.right).toEqual(15)
+        expect(box.bottom).toEqual(20)
+        expect(box.width).toEqual(10)
+        expect(box.height).toEqual(10)
+        expect(box.area).toEqual(100)
+      })
+
+    })
+
+    describe('from IRect', () => {
+
+      it('properties', () => {
+
+        const box = new Box({ x: 5, y: 10, width: 15, height: 20 })
+
+        expect(box.left).toEqual(5)
+        expect(box.x).toEqual(5)
+        expect(box.top).toEqual(10)
+        expect(box.y).toEqual(10)
+        expect(box.right).toEqual(20)
+        expect(box.bottom).toEqual(30)
+        expect(box.width).toEqual(15)
+        expect(box.height).toEqual(20)
+        expect(box.area).toEqual(300)
+      })
+
+    })
+
+  })
+
+  describe('rescale', () => {
+
+    it('scale down by factor 0.5', () => {
+
+      const box = new Box({ x: 10, y: 20, width: 20, height: 40 })
+
+      const rescaled = box.rescale(0.5)
+
+      expect(rescaled.x).toEqual(5)
+      expect(rescaled.y).toEqual(10)
+      expect(rescaled.width).toEqual(10)
+      expect(rescaled.height).toEqual(20)
+
+    })
+
+    it('scale up by factor 2', () => {
+
+      const box = new Box({ x: 10, y: 20, width: 20, height: 40 })
+
+      const rescaled = box.rescale(2)
+
+      expect(rescaled.x).toEqual(20)
+      expect(rescaled.y).toEqual(40)
+      expect(rescaled.width).toEqual(40)
+      expect(rescaled.height).toEqual(80)
+
+    })
+
+    it('scale to dimensions ', () => {
+
+      const box = new Box({ x: 0.1, y: 0.2, width: 0.2, height: 0.4 })
+
+      const rescaled = box.rescale({ width: 100, height: 200 })
+
+      expect(rescaled.x).toEqual(10)
+      expect(rescaled.y).toEqual(40)
+      expect(rescaled.width).toEqual(20)
+      expect(rescaled.height).toEqual(80)
+
+    })
+
+  })
+
+  describe('shift', () => {
+
+    it('should shift box by x, y', () => {
+
+      const box = new Box({ x: 10, y: 20, width: 20, height: 40 })
+
+      const shifted = box.shift(20, 40)
+
+      expect(shifted.x).toEqual(30)
+      expect(shifted.y).toEqual(60)
+      expect(shifted.width).toEqual(20)
+      expect(shifted.height).toEqual(40)
+
+    })
+
+  })
+
+})
\ No newline at end of file

test/tests/classes/Rect.test.ts

+import { Box } from '../../../src/classes/Box';
+import { Rect } from '../../../src/classes/Rect';
+
+describe('Rect', () => {
+
+  describe('constructor', () => {
+
+    it('can be created', () => {
+      const rect = new Rect(0, 10, 20, 30)
+      expect(rect instanceof Rect).toBe(true)
+      expect(rect instanceof Box).toBe(true)
+      expect(rect.x).toEqual(0)
+      expect(rect.y).toEqual(10)
+      expect(rect.width).toEqual(20)
+      expect(rect.height).toEqual(30)
+    })
+
+    it('throws if coordinates are invalid', () => {
+
+      const expectConstructorToThrow = (x: any, y: any, width: any, height: any) => {
+        expect(() => new Rect(x, y, width, height)).toThrowError(`Box.constructor - expected box to be IBoundingBox | IRect, instead have ${JSON.stringify({ x, y, width, height })}`)
+      }
+
+      expectConstructorToThrow(NaN, 10, 20, 30)
+      expectConstructorToThrow(0, Infinity, 20, 30)
+      expectConstructorToThrow(0, 10, -Infinity, 30)
+      expectConstructorToThrow(0, 10, 20, null)
+      expectConstructorToThrow(NaN, Infinity, undefined, null)
+      expectConstructorToThrow(undefined, undefined, undefined, undefined)
+    })
+
+    it('throws if height or width invalid', () => {
+      expect(() => new Rect(0, 10, -20, 30, false)).toThrowError('Box.constructor - width (-20) and height (30) must be positive numbers')
+      expect(() => new Rect(0, 10, 20, -30, false)).toThrowError('Box.constructor - width (20) and height (-30) must be positive numbers')
+    })
+
+    it('properties', () => {
+      const rect = new Rect(5, 10, 15, 20)
+      expect(rect.left).toEqual(5)
+      expect(rect.x).toEqual(5)
+      expect(rect.top).toEqual(10)
+      expect(rect.y).toEqual(10)
+      expect(rect.right).toEqual(20)
+      expect(rect.bottom).toEqual(30)
+      expect(rect.width).toEqual(15)
+      expect(rect.height).toEqual(20)
+      expect(rect.area).toEqual(300)
+    })
+
+  })
+
+})
\ No newline at end of file

test/tests/common/getModelUris.test.ts

+import { getModelUris } from '../../../src/common/getModelUris';
+
+const FAKE_DEFAULT_MODEL_NAME = 'fake_model_name'
+
+describe('getModelUris', () => {
+
+  it('returns uris from relative url if no argument passed', () => {
+    const result = getModelUris(undefined, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual('')
+  })
+
+  it('returns uris from relative url for empty string', () => {
+    const result = getModelUris('', FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual('')
+  })
+
+  it('returns uris for top level url, leading slash preserved', () => {
+    const result = getModelUris('/', FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`/${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual('/')
+  })
+
+  it('returns uris, given url path', () => {
+    const uri = 'path/to/modelfiles'
+    const result = getModelUris(uri, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`${uri}/${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual(uri)
+  })
+
+  it('returns uris, given url path, leading slash preserved', () => {
+    const uri = '/path/to/modelfiles'
+    const result = getModelUris(`/${uri}`, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`${uri}/${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual(uri)
+  })
+
+  it('returns uris, given manifest uri', () => {
+    const uri = 'path/to/modelfiles/model-weights_manifest.json'
+    const result = getModelUris(uri, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(uri)
+    expect(result.modelBaseUri).toEqual('path/to/modelfiles')
+  })
+
+  it('returns uris, given manifest uri, leading slash preserved', () => {
+    const uri = '/path/to/modelfiles/model-weights_manifest.json'
+    const result = getModelUris(uri, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(uri)
+    expect(result.modelBaseUri).toEqual('/path/to/modelfiles')
+  })
+
+  it('returns correct uris, given external path', () => {
+    const uri = 'https://example.com/path/to/modelfiles';
+    const result = getModelUris(uri, FAKE_DEFAULT_MODEL_NAME)
+
+    expect(result.manifestUri).toEqual(`${uri}/${FAKE_DEFAULT_MODEL_NAME}-weights_manifest.json`)
+    expect(result.modelBaseUri).toEqual(uri)
+  })
+
+})
\ No newline at end of file

test/tests/dom/NetInput.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { NetInput } from '../../../src';
+import { getTestEnv } from '../../env';
+import { expectAllTensorsReleased, fakeTensor3d } from '../../utils';
+
+describe('NetInput', () => {
+
+  let imgEl: HTMLImageElement
+
+  beforeAll(async () => {
+    imgEl = await getTestEnv().loadImage('test/images/white.png')
+  })
+
+  describe('toBatchTensor', () => {
+
+    it('HTMLImageElement, batchSize === 1', () => tf.tidy(() => {
+      const netInput = new NetInput([imgEl])
+      const batchTensor = netInput.toBatchTensor(100)
+      expect(batchTensor.shape).toEqual([1, 100, 100, 3])
+    }))
+
+    it('tf.Tensor3D, batchSize === 1', () => tf.tidy(() => {
+      const tensor = tf.zeros<tf.Rank.R3>([200, 200, 3], 'int32')
+      const netInput = new NetInput([tensor])
+      const batchTensor = netInput.toBatchTensor(100)
+      expect(batchTensor.shape).toEqual([1, 100, 100, 3])
+    }))
+
+    it('HTMLImageElements, batchSize === 4', () => tf.tidy(() => {
+      const netInput = new NetInput([imgEl, imgEl, imgEl, imgEl])
+      const batchTensor = netInput.toBatchTensor(100)
+      expect(batchTensor.shape).toEqual([4, 100, 100, 3])
+    }))
+
+    it('tf.Tensor3Ds, batchSize === 4', () => tf.tidy(() => {
+      const tensor = tf.zeros<tf.Rank.R3>([200, 200, 3], 'int32')
+      const netInput = new NetInput([tensor, tensor, tensor, tensor])
+      const batchTensor = netInput.toBatchTensor(100)
+      expect(batchTensor.shape).toEqual([4, 100, 100, 3])
+    }))
+
+    it('tf.Tensor3Ds and HTMLImageElements, batchSize === 4', () => tf.tidy(() => {
+      const tensor = tf.zeros<tf.Rank.R3>([200, 200, 3], 'int32')
+      const netInput = new NetInput([tensor, tensor, imgEl, imgEl])
+      const batchTensor = netInput.toBatchTensor(100)
+      expect(batchTensor.shape).toEqual([4, 100, 100, 3])
+    }))
+
+  })
+
+  describe('no memory leaks', () => {
+
+    it('constructor', async () => {
+      const tensors = [fakeTensor3d(), fakeTensor3d(), fakeTensor3d()]
+
+      await expectAllTensorsReleased(() => {
+        new NetInput([imgEl])
+        new NetInput([imgEl, imgEl, imgEl])
+        new NetInput([tensors[0]])
+        new NetInput(tensors)
+      })
+
+      tensors.forEach(t => t.dispose())
+    })
+
+    describe('toBatchTensor', () => {
+
+      it('single image element', async () => {
+        await expectAllTensorsReleased(() => {
+          const batchTensor = new NetInput([imgEl]).toBatchTensor(100, false)
+          batchTensor.dispose()
+        })
+      })
+
+      it('multiple image elements', async () => {
+        await expectAllTensorsReleased(() => {
+          const batchTensor = new NetInput([imgEl, imgEl, imgEl]).toBatchTensor(100, false)
+          batchTensor.dispose()
+        })
+      })
+
+    })
+
+  })
+
+})

test/tests/dom/fetchImage.browser.test.ts

+import { fetchImage } from '../../../src';
+
+describe('fetchImage', () => {
+
+  it('invalid mime type', async () => {
+    const url = 'base/test/data/boxes.json'
+
+    let err = ''
+    try {
+      await fetchImage(url)
+    } catch (e) {
+      err = e.toString()
+    }
+
+    expect(err).toContain('fetchImage - expected blob type to be of type image/*, instead have: application/json')
+    expect(err).toContain(url)
+  })
+
+  it('fetches image', async () => {
+    const url = 'base/test/images/white.png'
+    const img = await fetchImage(url)
+    expect(img instanceof HTMLImageElement).toBe(true)
+  })
+
+})

test/tests/dom/fetchJson.browser.test.ts

+import { fetchJson } from '../../../src';
+
+describe('fetchJson', () => {
+
+  it('fetches json', async () => {
+    const url = 'test/data/boxes.json'
+    expect(async () => await fetchJson(url)).not.toThrow()
+  })
+
+})

test/tests/dom/fetchNetWeights.browser.test.ts

+import { fetchNetWeights } from '../../../src';
+
+describe('fetchNetWeights', () => {
+
+  it('fetches .weights file', async () => {
+    const url = 'base/test/data/dummy.weights'
+    const weights = await fetchNetWeights(url)
+    expect(weights instanceof Float32Array).toBe(true)
+  })
+
+})

test/tests/dom/fetchOrThrow.browser.test.ts

+import { fetchOrThrow } from '../../../src';
+
+describe('fetchOrThrow', () => {
+
+  it('404, throws', async () => {
+    const url = '/does/not/exist'
+
+    let err = ''
+    try {
+      await fetchOrThrow(url)
+    } catch (e) {
+      err = e.toString()
+    }
+
+    expect(err).toContain('failed to fetch: (404)')
+    expect(err).toContain(url)
+  })
+
+})

test/tests/dom/toNetInput.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { createCanvasFromMedia, env, NetInput, toNetInput } from '../../../src';
+import { getTestEnv } from '../../env';
+import { expectAllTensorsReleased } from '../../utils';
+
+
+describe('toNetInput', () => {
+
+  let imgEl: HTMLImageElement, canvasEl: HTMLCanvasElement
+
+  beforeAll(async () => {
+    imgEl = await getTestEnv().loadImage('test/images/white.png')
+    canvasEl = createCanvasFromMedia(imgEl)
+  })
+
+  describe('valid args', () => {
+
+    it('from HTMLImageElement', async () => {
+      const netInput = await toNetInput(imgEl)
+      expect(netInput instanceof NetInput).toBe(true)
+      expect(netInput.batchSize).toEqual(1)
+    })
+
+    it('from HTMLCanvasElement', async () => {
+      const netInput = await toNetInput(canvasEl)
+      expect(netInput instanceof NetInput).toBe(true)
+      expect(netInput.batchSize).toEqual(1)
+    })
+
+    it('from HTMLImageElement array', async () => {
+      const netInput = await toNetInput([
+        imgEl,
+        imgEl
+      ])
+      expect(netInput instanceof NetInput).toBe(true)
+      expect(netInput.batchSize).toEqual(2)
+    })
+
+    it('from HTMLCanvasElement array', async () => {
+      const netInput = await toNetInput([
+        canvasEl,
+        canvasEl
+      ])
+      expect(netInput instanceof NetInput).toBe(true)
+      expect(netInput.batchSize).toEqual(2)
+    })
+
+    it('from mixed media array', async () => {
+      const netInput = await toNetInput([
+        imgEl,
+        canvasEl,
+        canvasEl
+      ])
+      expect(netInput instanceof NetInput).toBe(true)
+      expect(netInput.batchSize).toEqual(3)
+    })
+
+  })
+
+  describe('invalid args', () => {
+    it('undefined', async () => {
+      let errorMessage
+      try {
+        await toNetInput(undefined as any)
+      } catch (error) {
+          errorMessage = error.message;
+      }
+      expect(errorMessage).toBe('toNetInput - expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id')
+    })
+
+    it('empty array', async () => {
+      let errorMessage
+      try {
+        await toNetInput([])
+      } catch (error) {
+          errorMessage = error.message;
+      }
+      expect(errorMessage).toBe('toNetInput - empty array passed as input')
+    })
+
+    it('undefined at input index 1', async () => {
+      let errorMessage
+      try {
+        await toNetInput([env.getEnv().createImageElement(), undefined] as any)
+      } catch (error) {
+          errorMessage = error.message;
+      }
+      expect(errorMessage).toBe('toNetInput - at input index 1: expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id')
+    })
+
+  })
+
+  describe('no memory leaks', () => {
+
+    it('constructor', async () => {
+      const tensors = [imgEl, imgEl, imgEl].map(el => tf.browser.fromPixels(createCanvasFromMedia(el)))
+      const tensor4ds = tensors.map(t => t.expandDims<tf.Rank.R4>())
+
+      await expectAllTensorsReleased(async () => {
+        await toNetInput(imgEl)
+        await toNetInput([imgEl, imgEl, imgEl])
+        await toNetInput(tensors[0])
+        await toNetInput(tensors)
+        await toNetInput(tensor4ds[0])
+        await toNetInput(tensor4ds)
+      })
+
+      tensors.forEach(t => t.dispose())
+      tensor4ds.forEach(t => t.dispose())
+    })
+
+  })
+
+})

test/tests/faceLandmarkNet/faceLandmark68Net.test.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { createCanvasFromMedia, IDimensions, isTensor3D, NetInput, Point, TMediaElement, toNetInput } from '../../../src';
-import { FaceLandmarks68 } from '../../../src/classes/FaceLandmarks68';
+import { createCanvasFromMedia, IDimensions, FaceLandmarks68, utils, NetInput, Point, TMediaElement, toNetInput } from '../../../src';
 import { getTestEnv } from '../../env';
 import {
   describeWithBackend,
@@ -13,7 +12,7 @@ import {
 
 function getInputDims (input: tf.Tensor | TMediaElement): IDimensions {
   if (input instanceof tf.Tensor) {
-    const [height, width] = input.shape.slice(isTensor3D(input) ? 0 : 1)
+    const [height, width] = input.shape.slice(utils.isTensor3D(input) ? 0 : 1)
     return { width, height }
   }
   return input

test/tests/faceLandmarkNet/faceLandmark68TinyNet.test.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { createCanvasFromMedia, IDimensions, isTensor3D, NetInput, Point, TMediaElement, toNetInput } from '../../../src';
-import { FaceLandmarks68 } from '../../../src/classes/FaceLandmarks68';
+import { createCanvasFromMedia, IDimensions, utils, NetInput, Point, TMediaElement, toNetInput, FaceLandmarks68 } from '../../../src';
 import { getTestEnv } from '../../env';
 import { describeWithBackend, describeWithNets, expectAllTensorsReleased, expectPointClose } from '../../utils';
 
 function getInputDims (input: tf.Tensor | TMediaElement): IDimensions {
   if (input instanceof tf.Tensor) {
-    const [height, width] = input.shape.slice(isTensor3D(input) ? 0 : 1)
+    const [height, width] = input.shape.slice(utils.isTensor3D(input) ? 0 : 1)
     return { width, height }
   }
   return input

test/tests/faceRecognitionNet/faceRecognitionNet.test.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { createCanvasFromMedia, NetInput, toNetInput } from '../../../src';
-import { euclideanDistance } from '../../../src/euclideanDistance';
+import { createCanvasFromMedia, euclideanDistance, NetInput, toNetInput } from '../../../src';
 import { getTestEnv } from '../../env';
 import { describeWithBackend, describeWithNets, expectAllTensorsReleased } from '../../utils';
 

test/tests/factories/WithFaceDetection.test.ts

-import { Rect } from '../../../src';
-import { FaceDetection } from '../../../src/classes/FaceDetection';
-import { extendWithFaceDetection } from '../../../src/factories/WithFaceDetection';
+import { extendWithFaceDetection, FaceDetection, Rect } from '../../../src';
 
 const detection = new FaceDetection(1.0, new Rect(0, 0, 0.5, 0.5), { width: 100, height: 100 })
 

test/tests/factories/WithFaceLandmarks.test.ts

-import { Point, Rect } from '../../../src';
-import { FaceDetection } from '../../../src/classes/FaceDetection';
-import { FaceLandmarks68 } from '../../../src/classes/FaceLandmarks68';
-import { extendWithFaceDetection } from '../../../src/factories/WithFaceDetection';
-import { extendWithFaceLandmarks } from '../../../src/factories/WithFaceLandmarks';
+import { extendWithFaceDetection, extendWithFaceLandmarks, FaceDetection, FaceLandmarks68, Point, Rect } from '../../../src';
 
 const detection = new FaceDetection(1.0, new Rect(0.5, 0.5, 0.5, 0.5), { width: 100, height: 100 })
 const unshiftedLandmarks = new FaceLandmarks68(Array(68).fill(0).map((_, i) => new Point(i / 100, i / 100)), { width: 100, height: 100 })

test/tests/globalApi/FaceMatcher.test.ts

-import { FaceMatcher } from '../../../src/globalApi/FaceMatcher';
+import { FaceMatcher } from '../../../src';
 import { LabeledFaceDescriptors } from '../../../src';
 
 describe('globalApi', () => {

test/tests/globalApi/detectAllFaces.test.ts

-import * as faceapi from '../../../src';
-import { WithAge } from '../../../src/factories/WithAge';
-import { WithFaceDetection } from '../../../src/factories/WithFaceDetection';
-import { WithFaceExpressions } from '../../../src/factories/WithFaceExpressions';
-import { WithGender } from '../../../src/factories/WithGender';
+import { WithAge, detectAllFaces, WithFaceExpressions, WithFaceDetection, WithGender } from '../../../src';
 import { getTestEnv } from '../../env';
 import { expectedTinyFaceDetectorBoxes } from '../../expectedTinyFaceDetectorBoxes';
 import { expectFaceDetectionsWithLandmarks } from '../../expectFaceDetectionsWithLandmarks';
@@ -58,8 +54,7 @@ describeWithBackend('globalApi', () => {
     describe('without face alignment', () => {
 
       it('detectAllFaces.withFaceExpressions()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceExpressions()
 
         expect(results.length).toEqual(6)
@@ -67,8 +62,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withAgeAndGender()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withAgeAndGender()
 
         expect(results.length).toEqual(6)
@@ -76,8 +70,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceExpressions().withAgeAndGender()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceExpressions()
           .withAgeAndGender()
 
@@ -87,8 +80,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withAgeAndGender().withFaceExpressions()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withAgeAndGender()
           .withFaceExpressions()
 
@@ -102,8 +94,7 @@ describeWithBackend('globalApi', () => {
     describe('with face alignment', () => {
 
       it('detectAllFaces.withFaceLandmarks().withFaceExpressions()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
 
@@ -113,8 +104,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withAgeAndGender()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
 
@@ -124,8 +114,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withFaceDescriptors()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceDescriptors()
 
@@ -134,8 +123,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withFaceExpressions().withAgeAndGender()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withAgeAndGender()
@@ -147,8 +135,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withAgeAndGender().withFaceExpressions()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
           .withFaceExpressions()
@@ -160,8 +147,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withFaceExpressions().withFaceDescriptors()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withFaceDescriptors()
@@ -172,8 +158,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withAgeAndGender().withFaceDescriptors()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
           .withFaceDescriptors()
@@ -184,8 +169,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectAllFaces.withFaceLandmarks().withFaceExpressions().withAgeAndGender().withFaceDescriptors()', async () => {
-        const results = await faceapi
-          .detectAllFaces(imgEl, faceDetectorOptions)
+        const results = await detectAllFaces(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withAgeAndGender()
@@ -203,8 +187,7 @@ describeWithBackend('globalApi', () => {
 
       it('detectAllFaces.withFaceLandmarks().withFaceDescriptors()', async () => {
         await expectAllTensorsReleased(async () => {
-          await faceapi
-            .detectAllFaces(imgEl, faceDetectorOptions)
+          await detectAllFaces(imgEl, faceDetectorOptions)
             .withFaceLandmarks()
             .withFaceDescriptors()
         })
@@ -212,8 +195,7 @@ describeWithBackend('globalApi', () => {
 
       it('detectAllFaces.withFaceLandmarks().withFaceExpressions().withAgeAndGender().withFaceDescriptors()', async () => {
         await expectAllTensorsReleased(async () => {
-          await faceapi
-            .detectAllFaces(imgEl, faceDetectorOptions)
+          await detectAllFaces(imgEl, faceDetectorOptions)
             .withFaceLandmarks()
             .withFaceExpressions()
             .withAgeAndGender()

test/tests/globalApi/detectSingleFace.test.ts

-import * as faceapi from '../../../src';
-import { WithAge } from '../../../src/factories/WithAge';
-import { WithFaceExpressions } from '../../../src/factories/WithFaceExpressions';
-import { WithGender } from '../../../src/factories/WithGender';
-import { getTestEnv } from '../../env';
+import {
+  detectSingleFace,
+  WithAge,
+  WithFaceDescriptor,
+  WithFaceDetection,
+  WithFaceExpressions,
+  WithFaceLandmarks,
+  WithGender,
+} from '../../../src';
 import { expectedTinyFaceDetectorBoxes } from '../../expectedTinyFaceDetectorBoxes';
 import { expectFaceDetectionsWithLandmarks } from '../../expectFaceDetectionsWithLandmarks';
 import { expectFullFaceDescriptions } from '../../expectFullFaceDescriptions';
@@ -14,6 +18,7 @@ import {
   ExpectedFullFaceDescription,
 } from '../../utils';
 import { deltas, expectedScores, faceDetectorOptions, withNetArgs } from './consts';
+import { getTestEnv } from '../../env';
 
 
 function expectFaceExpressions(result: WithFaceExpressions<{}> | undefined) {
@@ -44,7 +49,7 @@ describeWithBackend('globalApi', () => {
     expectedFullFaceDescriptions = await assembleExpectedFullFaceDescriptions(expectedTinyFaceDetectorBoxes)
   })
 
-  function expectFaceDetectionWithLandmarks(result: faceapi.WithFaceLandmarks<faceapi.WithFaceDetection<{}>> | undefined) {
+  function expectFaceDetectionWithLandmarks(result: WithFaceLandmarks<WithFaceDetection<{}>> | undefined) {
     expect(!!result).toBeTruthy()
     if (result) {
       expectFaceDetectionsWithLandmarks(
@@ -56,7 +61,7 @@ describeWithBackend('globalApi', () => {
     }
   }
 
-  function expectFullFaceDescription(result: faceapi.WithFaceDescriptor<faceapi.WithFaceLandmarks<faceapi.WithFaceDetection<{}>>> | undefined) {
+  function expectFullFaceDescription(result: WithFaceDescriptor<WithFaceLandmarks<WithFaceDetection<{}>>> | undefined) {
     expect(!!result).toBeTruthy()
     if (result) {
       expectFullFaceDescriptions(
@@ -73,24 +78,21 @@ describeWithBackend('globalApi', () => {
     describe('without face alignment', () => {
 
       it('detectSingleFace.withFaceExpressions()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceExpressions()
 
         expectFaceExpressions(result)
       })
 
       it('detectSingleFace.withAgeAndGender()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withAgeAndGender()
 
         expectAgeAndGender(result, false)
       })
 
       it('detectSingleFace.withFaceExpressions().withAgeAndGender()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceExpressions()
           .withAgeAndGender()
 
@@ -99,8 +101,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withAgeAndGender().withFaceExpressions()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withAgeAndGender()
           .withFaceExpressions()
 
@@ -113,8 +114,7 @@ describeWithBackend('globalApi', () => {
     describe('with face alignment', () => {
 
       it('detectSingleFace.withFaceLandmarks().withFaceExpressions()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
 
@@ -123,8 +123,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withAgeAndGender()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
 
@@ -133,8 +132,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withFaceDescriptor()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceDescriptor()
 
@@ -142,8 +140,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withFaceExpressions().withAgeAndGender()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withAgeAndGender()
@@ -154,8 +151,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withAgeAndGender().withFaceExpressions()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
           .withFaceExpressions()
@@ -166,8 +162,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withFaceExpressions().withFaceDescriptor()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withFaceDescriptor()
@@ -177,8 +172,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withAgeAndGender().withFaceDescriptor()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withAgeAndGender()
           .withFaceDescriptor()
@@ -188,8 +182,7 @@ describeWithBackend('globalApi', () => {
       })
 
       it('detectSingleFace.withFaceLandmarks().withFaceExpressions().withAgeAndGender().withFaceDescriptor()', async () => {
-        const result = await faceapi
-          .detectSingleFace(imgEl, faceDetectorOptions)
+        const result = await detectSingleFace(imgEl, faceDetectorOptions)
           .withFaceLandmarks()
           .withFaceExpressions()
           .withAgeAndGender()
@@ -206,8 +199,7 @@ describeWithBackend('globalApi', () => {
 
       it('detectSingleFace.withFaceLandmarks().withFaceDescriptor()', async () => {
         await expectAllTensorsReleased(async () => {
-          await faceapi
-            .detectSingleFace(imgEl, faceDetectorOptions)
+          await detectSingleFace(imgEl, faceDetectorOptions)
             .withFaceLandmarks()
             .withFaceDescriptor()
         })
@@ -215,8 +207,7 @@ describeWithBackend('globalApi', () => {
 
       it('detectSingleFace.withFaceLandmarks().withFaceExpressions().withAgeAndGender().withFaceDescriptor()', async () => {
         await expectAllTensorsReleased(async () => {
-          await faceapi
-            .detectSingleFace(imgEl, faceDetectorOptions)
+          await detectSingleFace(imgEl, faceDetectorOptions)
             .withFaceLandmarks()
             .withFaceExpressions()
             .withAgeAndGender()

test/tests/ops/iou.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { iou, Rect } from '../../../src';
+
+describe('iou', () => {
+
+  it('should be 1.0', () => tf.tidy(() => {
+
+    const box = new Rect(0, 0, 20, 20)
+
+    expect(iou(box, box)).toEqual(1)
+
+  }))
+
+  it('should be 0', () => tf.tidy(() => {
+
+    const box1 = new Rect(0, 0, 20, 20)
+    const box2 = new Rect(20, 20, 20, 20)
+
+    expect(iou(box1, box2)).toEqual(0)
+
+  }))
+
+  it('should be 0.5', () => tf.tidy(() => {
+
+    const box1 = new Rect(0, 0, 20, 20)
+    const box2 = new Rect(0, 0, 10, 20)
+
+    expect(iou(box1, box2)).toEqual(0.5)
+
+  }))
+
+  it('should be 0.5', () => tf.tidy(() => {
+
+    const box1 = new Rect(0, 0, 20, 20)
+    const box2 = new Rect(0, 10, 20, 10)
+
+    expect(iou(box1, box2)).toEqual(0.5)
+
+  }))
+
+})
\ No newline at end of file

test/tests/ops/padToSquare.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { padToSquare } from '../../../src';
+import { ones, zeros } from '../../utils';
+
+describe('padToSquare', () => {
+
+  describe('even size', () => {
+
+    it('is padded to square by 2 columns', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(24).fill(1), [1, 4, 2, 3])
+      const result = padToSquare(imgTensor)
+
+      expect(result.shape).toEqual([1, 4, 4, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(4)
+      expect(paddedCols[0].dataSync()).toEqual(ones(12))
+      expect(paddedCols[1].dataSync()).toEqual(ones(12))
+      expect(paddedCols[2].dataSync()).toEqual(zeros(12))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(12))
+    }))
+
+    it('is padded to square by 2 columns and centered', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(24).fill(1), [1, 4, 2, 3])
+      const result = padToSquare(imgTensor, true)
+
+      expect(result.shape).toEqual([1, 4, 4, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(4)
+      expect(paddedCols[0].dataSync()).toEqual(zeros(12))
+      expect(paddedCols[1].dataSync()).toEqual(ones(12))
+      expect(paddedCols[2].dataSync()).toEqual(ones(12))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(12))
+    }))
+
+    it('is padded to square by 1 column', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(36).fill(1), [1, 4, 3, 3])
+      const result = padToSquare(imgTensor)
+
+      expect(result.shape).toEqual([1, 4, 4, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(4)
+      expect(paddedCols[0].dataSync()).toEqual(ones(12))
+      expect(paddedCols[1].dataSync()).toEqual(ones(12))
+      expect(paddedCols[2].dataSync()).toEqual(ones(12))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(12))
+    }))
+
+    it('is padded to square by 1 column and centered', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(36).fill(1), [1, 4, 3, 3])
+      const result = padToSquare(imgTensor, true)
+
+      expect(result.shape).toEqual([1, 4, 4, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(4)
+      expect(paddedCols[0].dataSync()).toEqual(ones(12))
+      expect(paddedCols[1].dataSync()).toEqual(ones(12))
+      expect(paddedCols[2].dataSync()).toEqual(ones(12))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(12))
+    }))
+
+  })
+
+  describe('uneven size', () => {
+
+    it('is padded to square by 3 columns', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(30).fill(1), [1, 5, 2, 3])
+      const result = padToSquare(imgTensor)
+
+      expect(result.shape).toEqual([1, 5, 5, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(5)
+      expect(paddedCols[0].dataSync()).toEqual(ones(15))
+      expect(paddedCols[1].dataSync()).toEqual(ones(15))
+      expect(paddedCols[2].dataSync()).toEqual(zeros(15))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(15))
+      expect(paddedCols[4].dataSync()).toEqual(zeros(15))
+    }))
+
+    it('is padded to square by 3 columns and centered', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(30).fill(1), [1, 5, 2, 3])
+      const result = padToSquare(imgTensor, true)
+
+      expect(result.shape).toEqual([1, 5, 5, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(5)
+      expect(paddedCols[0].dataSync()).toEqual(zeros(15))
+      expect(paddedCols[1].dataSync()).toEqual(ones(15))
+      expect(paddedCols[2].dataSync()).toEqual(ones(15))
+      expect(paddedCols[3].dataSync()).toEqual(zeros(15))
+      expect(paddedCols[4].dataSync()).toEqual(zeros(15))
+    }))
+
+    it('is padded to square by 1 column', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(60).fill(1), [1, 5, 4, 3])
+      const result = padToSquare(imgTensor)
+
+      expect(result.shape).toEqual([1, 5, 5, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(5)
+      expect(paddedCols[0].dataSync()).toEqual(ones(15))
+      expect(paddedCols[1].dataSync()).toEqual(ones(15))
+      expect(paddedCols[2].dataSync()).toEqual(ones(15))
+      expect(paddedCols[3].dataSync()).toEqual(ones(15))
+      expect(paddedCols[4].dataSync()).toEqual(zeros(15))
+    }))
+
+    it('is padded to square by 1 column and centered', () => tf.tidy(() => {
+      const imgTensor = tf.tensor4d(Array(60).fill(1), [1, 5, 4, 3])
+      const result = padToSquare(imgTensor, true)
+
+      expect(result.shape).toEqual([1, 5, 5, 3])
+
+      const paddedCols = tf.unstack(result, 2)
+      expect(paddedCols.length).toEqual(5)
+      expect(paddedCols[0].dataSync()).toEqual(ones(15))
+      expect(paddedCols[1].dataSync()).toEqual(ones(15))
+      expect(paddedCols[2].dataSync()).toEqual(ones(15))
+      expect(paddedCols[3].dataSync()).toEqual(ones(15))
+      expect(paddedCols[4].dataSync()).toEqual(zeros(15))
+    }))
+
+  })
+})

test/tests/resizeResults.test.ts

-import { Point } from 'tfjs-image-recognition-base';
-
-import { Rect } from '../../src';
-import { FaceDetection } from '../../src/classes/FaceDetection';
-import { FaceLandmarks68 } from '../../src/classes/FaceLandmarks68';
-import { extendWithFaceDetection } from '../../src/factories/WithFaceDetection';
-import { extendWithFaceLandmarks } from '../../src/factories/WithFaceLandmarks';
-import { resizeResults } from '../../src/resizeResults';
+import {
+  extendWithFaceDetection,
+  extendWithFaceLandmarks,
+  FaceDetection,
+  FaceLandmarks68,
+  Point,
+  Rect,
+  resizeResults,
+} from '../../src';
 import { expectPointsClose, expectRectClose } from '../utils';
 
 const detection = new FaceDetection(1.0, new Rect(0, 0, 0.5, 0.5), { width: 100, height: 100 })

test/utils.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import * as faceapi from '../src';
-import { FaceRecognitionNet, IPoint, IRect, Mtcnn, TinyYolov2 } from '../src/';
-import { AgeGenderNet } from '../src/ageGenderNet/AgeGenderNet';
-import { FaceDetection } from '../src/classes/FaceDetection';
-import { FaceLandmarks } from '../src/classes/FaceLandmarks';
-import { FaceExpressionNet } from '../src/faceExpressionNet/FaceExpressionNet';
-import { FaceLandmark68Net } from '../src/faceLandmarkNet/FaceLandmark68Net';
-import { FaceLandmark68TinyNet } from '../src/faceLandmarkNet/FaceLandmark68TinyNet';
-import { SsdMobilenetv1 } from '../src/ssdMobilenetv1/SsdMobilenetv1';
-import { TinyFaceDetector } from '../src/tinyFaceDetector/TinyFaceDetector';
+import {
+  AgeGenderNet,
+  FaceDetection,
+  FaceExpressionNet,
+  FaceLandmark68Net,
+  FaceLandmark68TinyNet,
+  FaceLandmarks,
+  FaceRecognitionNet,
+  IPoint,
+  IRect,
+  LabeledBox,
+  Mtcnn,
+  nets,
+  PredictedBox,
+  SsdMobilenetv1,
+  TinyFaceDetector,
+  TinyYolov2,
+} from '../src';
 import { getTestEnv } from './env';
 
 export function expectMaxDelta(val1: number, val2: number, maxDelta: number) {
@@ -80,6 +88,41 @@ export function sortByFaceDetection<T extends { detection: FaceDetection }>(objs
   return sortByDistanceToOrigin(objs, d => d.detection.box)
 }
 
+export function fakeTensor3d(dtype: tf.DataType = 'int32') {
+  return tf.tensor3d([0], [1, 1, 1], dtype)
+}
+
+export function zeros(length: number): Float32Array {
+  return new Float32Array(length)
+}
+
+export function ones(length: number): Float32Array {
+  return new Float32Array(length).fill(1)
+}
+
+export function createLabeledBox(
+  x: number,
+  y: number,
+  width: number,
+  height: number,
+  classLabel: number = 0
+): LabeledBox {
+  return new LabeledBox({ x, y, width, height }, classLabel)
+}
+
+export function createPredictedBox(
+  x: number,
+  y: number,
+  width: number,
+  height: number,
+  classLabel: number = 0,
+  score: number = 1.0,
+  classScore: number = 1.0
+): PredictedBox {
+  return new PredictedBox({ x, y, width, height }, classLabel, score, classScore)
+}
+
+
 export type ExpectedFaceDetectionWithLandmarks = {
   detection: IRect
   landmarks: IPoint[]
@@ -191,7 +234,7 @@ export function describeWithNets(
       faceExpressionNet,
       ageGenderNet,
       tinyYolov2
-    } = faceapi.nets
+    } = nets
 
     beforeAll(async () => {
       const {

test/utils/index.test.ts

+import { utils } from '../../src';
+
+describe('utils', () => {
+
+  describe('isValidNumber', () => {
+
+    it('0 is valid', () => {
+      expect(utils.isValidNumber(0)).toBe(true)
+    })
+
+    it('1 is valid', () => {
+      expect(utils.isValidNumber(1)).toBe(true)
+    })
+
+    it('-1 is valid', () => {
+      expect(utils.isValidNumber(-1)).toBe(true)
+    })
+
+    it('NaN is invalid', () => {
+      expect(utils.isValidNumber(NaN)).toBe(false)
+    })
+
+    it('Infinity is invalid', () => {
+      expect(utils.isValidNumber(Infinity)).toBe(false)
+    })
+
+    it('-Infinity is invalid', () => {
+      expect(utils.isValidNumber(-Infinity)).toBe(false)
+    })
+
+    it('null is invalid', () => {
+      expect(utils.isValidNumber(null)).toBe(false)
+    })
+
+    it('undefined is invalid', () => {
+      expect(utils.isValidNumber(undefined)).toBe(false)
+    })
+
+  })
+})