Merge pull request #1045 from ml5js/joeyklee.637-mock-functions-in-tests

[testing]: mocks objectDetector functions

src/ObjectDetector/CocoSsd/index.js

@@ -17,7 +17,7 @@ const DEFAULTS = {
   modelUrl: undefined,
 };
 
-class CocoSsdBase {
+export class CocoSsdBase {
   /**
    * Create CocoSsd model. Works on video and images.
    * @param {function} constructorCallback - Optional. A callback function that is called once the model has loaded. If no callback is provided, it will return a promise
@@ -129,7 +129,7 @@ class CocoSsdBase {
   }
 }
 
-const CocoSsd = (videoOr, optionsOr, cb) => {
+export const CocoSsd = (videoOr, optionsOr, cb) => {
   let video = null;
   let options = {};
   let callback = cb;
@@ -153,4 +153,4 @@ const CocoSsd = (videoOr, optionsOr, cb) => {
   return new CocoSsdBase(video, options, callback);
 };
 
-export default CocoSsd;
+// export default CocoSsd;

src/ObjectDetector/YOLO/index.js

@@ -36,7 +36,7 @@ const DEFAULTS = {
 // Size of the video
 const imageSize = 416;
 
-class YOLOBase extends Video {
+export class YOLOBase extends Video {
   /**
    * @deprecated Please use ObjectDetector class instead
    */
@@ -236,7 +236,7 @@ class YOLOBase extends Video {
   }
 }
 
-const YOLO = (videoOr, optionsOr, cb) => {
+export const YOLO = (videoOr, optionsOr, cb) => {
   let video = null;
   let options = {};
   let callback = cb;
@@ -260,4 +260,4 @@ const YOLO = (videoOr, optionsOr, cb) => {
   return new YOLOBase(video, options, callback);
 };
 
-export default YOLO;
+// export default YOLO;

src/ObjectDetector/index.js

@@ -7,8 +7,8 @@
   ObjectDetection
 */
 
-import YOLO from "./YOLO/index";
-import CocoSsd from "./CocoSsd/index";
+import { YOLO } from "./YOLO/index";
+import {CocoSsd} from "./CocoSsd/index";
 
 class ObjectDetector {
   /**

src/ObjectDetector/index_test.js

@@ -3,6 +3,8 @@
 // This software is released under the MIT License.
 // https://opensource.org/licenses/MIT
 
+const { getImageData, getRobin } = ml5.testingUtils;
+
 const COCOSSD_DEFAULTS = {
   base: "lite_mobilenet_v2",
   modelUrl: undefined,
@@ -15,30 +17,31 @@ const YOLO_DEFAULTS = {
   size: 416,
 };
 
-async function getRobin() {
-  const img = new Image();
-  img.crossOrigin = "";
-  img.src = "https://cdn.jsdelivr.net/gh/ml5js/ml5-library@development/assets/bird.jpg";
-  await new Promise(resolve => {
-    img.onload = resolve;
-  });
-  return img;
-}
-
-async function getImageData() {
-  const arr = new Uint8ClampedArray(20000);
+const mockYoloObject = {
+  IOUThreshold: YOLO_DEFAULTS.IOUThreshold,
+  classProbThreshold: YOLO_DEFAULTS.classProbThreshold,
+  filterBoxesThreshold: YOLO_DEFAULTS.filterBoxesThreshold,
+  size: YOLO_DEFAULTS.size,
+  detect: () => {
+    return [{ label: "bird", confidence: 0.9 }];
+  },
+};
+const mockCocoObject = {
+  config: { ...COCOSSD_DEFAULTS },
+  detect: () => {
+    return [{ label: "bird", confidence: 0.9 }];
+  },
+};
 
-  // Iterate through every pixel
-  for (let i = 0; i < arr.length; i += 4) {
-    arr[i + 0] = 0; // R value
-    arr[i + 1] = 190; // G value
-    arr[i + 2] = 0; // B value
-    arr[i + 3] = 255; // A value
+function mockObjectDetector(modelName) {
+  switch (modelName) {
+    case "yolo":
+      return mockYoloObject;
+    case "cocossd":
+      return mockCocoObject;
+    default:
+      return mockCocoObject;
   }
-
-  // Initialize a new ImageData object
-  const img = new ImageData(arr, 200);
-  return img;
 }
 
 describe("objectDetector", () => {
@@ -47,31 +50,37 @@ describe("objectDetector", () => {
   /**
    * Test cocossd object detector
    */
+
   describe("objectDetector: cocossd", () => {
     let cocoDetector;
 
     beforeAll(async () => {
+      spyOn(ml5, "objectDetector").and.callFake(mockObjectDetector);
       cocoDetector = await ml5.objectDetector("cocossd");
     });
 
     it("Should instantiate with the following defaults", () => {
-      expect(cocoDetector.config.base).toBe(COCOSSD_DEFAULTS.base);
-      expect(cocoDetector.config.modelUrl).toBe(COCOSSD_DEFAULTS.modelUrl);
+      expect(ml5.objectDetector).toHaveBeenCalled();
+      expect(cocoDetector.config.toString()).toBe(COCOSSD_DEFAULTS.toString());
     });
 
     it("detects a robin", async () => {
+      spyOn(cocoDetector, "detect").and.returnValue([{ label: "bird", confidence: 0.9 }]);
+
       const robin = await getRobin();
       const detection = await cocoDetector.detect(robin);
       expect(detection[0].label).toBe("bird");
     });
 
     it("detects takes ImageData", async () => {
+      spyOn(cocoDetector, "detect").and.returnValue([]);
       const img = await getImageData();
       const detection = await cocoDetector.detect(img);
       expect(detection).toEqual([]);
     });
 
     it("throws error when a non image is trying to be detected", async () => {
+      spyOn(cocoDetector, "detect").and.throwError("Detection subject not supported");
       const notAnImage = "not_an_image";
       try {
         await cocoDetector.detect(notAnImage);
@@ -88,23 +97,27 @@ describe("objectDetector", () => {
   describe("objectDetector: yolo", () => {
     let yolo;
     beforeAll(async () => {
+      spyOn(ml5, "objectDetector").and.callFake(mockObjectDetector);
       yolo = await ml5.objectDetector("yolo", { disableDeprecationNotice: true, ...YOLO_DEFAULTS });
     });
 
     it("instantiates the YOLO classifier with defaults", () => {
+      expect(ml5.objectDetector).toHaveBeenCalled();
       expect(yolo.IOUThreshold).toBe(YOLO_DEFAULTS.IOUThreshold);
       expect(yolo.classProbThreshold).toBe(YOLO_DEFAULTS.classProbThreshold);
       expect(yolo.filterBoxesThreshold).toBe(YOLO_DEFAULTS.filterBoxesThreshold);
       expect(yolo.size).toBe(YOLO_DEFAULTS.size);
     });
 
     it("detects a robin", async () => {
+      spyOn(yolo, "detect").and.returnValue([{ label: "bird", confidence: 0.9 }]);
       const robin = await getRobin();
       const detection = await yolo.detect(robin);
       expect(detection[0].label).toBe("bird");
     });
 
     it("detects takes ImageData", async () => {
+      spyOn(yolo, "detect").and.returnValue([]);
       const img = await getImageData();
       const detection = await yolo.detect(img);
       expect(detection).toEqual([]);

src/index.js

@@ -11,7 +11,8 @@ import soundClassifier from "./SoundClassifier/";
 import KNNClassifier from "./KNNClassifier/";
 import featureExtractor from "./FeatureExtractor/";
 import word2vec from "./Word2vec/";
-import YOLO from "./ObjectDetector/YOLO";
+import {YOLO} from "./ObjectDetector/YOLO";
+import {CocoSsd} from "./ObjectDetector/CocoSsd";
 import objectDetector from "./ObjectDetector";
 import poseNet from "./PoseNet";
 import * as imageUtils from "./utils/imageUtilities";
@@ -35,6 +36,7 @@ import facemesh from "./Facemesh";
 import handpose from './Handpose';
 import p5Utils from "./utils/p5Utils";
 import communityStatement from "./utils/community";
+import * as testingUtils from "./utils/testingUtils";
 
 const withPreload = {
   charRNN,
@@ -52,6 +54,7 @@ const withPreload = {
   styleTransfer,
   word2vec,
   YOLO,
+  CocoSsd,
   objectDetector,
   uNet,
   sentiment,
@@ -75,4 +78,5 @@ module.exports = Object.assign({ p5Utils }, preloadRegister(withPreload), {
   tfvis,
   version,
   neuralNetwork,
+  testingUtils
 });

src/utils/testingUtils/index.js

@@ -0,0 +1,26 @@
+
+export const getRobin = async () => {
+  const img = new Image();
+  img.crossOrigin = "";
+  img.src = "https://cdn.jsdelivr.net/gh/ml5js/ml5-library@development/assets/bird.jpg";
+  await new Promise(resolve => {
+    img.onload = resolve;
+  });
+  return img;
+}
+
+export const getImageData = async () => {
+  const arr = new Uint8ClampedArray(20000);
+
+  // Iterate through every pixel
+  for (let i = 0; i < arr.length; i += 4) {
+    arr[i + 0] = 0; // R value
+    arr[i + 1] = 190; // G value
+    arr[i + 2] = 0; // B value
+    arr[i + 3] = 255; // A value
+  }
+
+  // Initialize a new ImageData object
+  const img = new ImageData(arr, 200);
+  return img;
+}