Upgrading tfjs

.gitignore

@@ -2,3 +2,5 @@ node_modules
 dist
 yarn-error.log
 .rpt2_cache
+.cache
+example/.cache

README.md

@@ -49,7 +49,7 @@ await mlClassifier.train(imageData, {
     onTrainBegin: () => {
       console.log('training begins');
     },
-    onBatchEnd: (batch: any,logs: any) => {
+    onBatchEnd: (batch, logs) => {
       console.log('Loss is: ' + logs.loss.toFixed(5));
     }
   },
@@ -232,7 +232,7 @@ Nothing.
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
 mlClassifier.train({
   callbacks: {
     onTrainBegin: () => {
@@ -259,9 +259,9 @@ mlClassifier.train({
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
 mlClassifier.train();
-mlClassifier.addData(evaluationImages, labels, DataType.EVALUATE);
+mlClassifier.addData(evaluationImages, labels, 'evaluate');
 mlClassifier.evaluate();
 ```
 
@@ -282,7 +282,7 @@ mlClassifier.evaluate();
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
 mlClassifier.train();
 mlClassifier.predict(imageToPredict);
 ```
@@ -304,7 +304,7 @@ mlClassifier.predict(imageToPredict);
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
 mlClassifier.train();
 mlClassifier.save(('path-to-save');
 ```
@@ -324,7 +324,7 @@ mlClassifier.save(('path-to-save');
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
 mlClassifier.train();
 mlClassifier.getModel();
 ```
@@ -345,13 +345,13 @@ The saved Tensorflow.js model.
 ```
 import MLClassifier from 'ml-classifier';
 const mlClassifier = new MLClassifier();
-mlClassifier.addData(images, labels, DataType.TRAIN);
-mlClassifier.clearData(DataType.TRAIN);
+mlClassifier.addData(images, labels, 'train');
+mlClassifier.clearData('train');
 ```
 
 #### Parameters
 
-* **dataType** (`DataType`) *Optional* - specifies which data to clear. If no argument is provided, all data will be cleared.
+* **dataType** (`string`) *Optional* - specifies which data to clear. If no argument is provided, all data will be cleared.
 
 #### Returns
 

example/.babelrc

@@ -0,0 +1,6 @@
+{
+  "plugins": [
+
+  "@babel/plugin-transform-runtime"
+  ]
+}

example/index.html

@@ -0,0 +1 @@
+<script src="./index.js"></script>

example/index.js

@@ -0,0 +1,76 @@
+/* globals Promise */
+import * as tf from '@tensorflow/tfjs';
+import MLClassifier from 'ml-classifier';
+import dog from './images/dog.jpg';
+import cat from './images/cat.png';
+
+const mlClassifier = new MLClassifier();
+
+const loadImage = (src) => new Promise((resolve, reject) => {
+  const image = new Image();
+  image.src = src;
+  image.crossOrigin = '';
+  image.onload = () => resolve(image);
+  image.onerror = (err) => reject(err);
+});
+
+function cropImage(img) {
+  const height = img.shape[0];
+  const width = img.shape[1];
+
+  // use the shorter side as the size to which we will crop
+  const shorterSide = Math.min(img.shape[0], img.shape[1]);
+
+  // calculate beginning and ending crop points
+  const startingWidth = Math.floor((width - shorterSide) / 2);
+  const startingHeight = Math.floor((height - shorterSide) / 2);
+  const endingWidth = Math.floor(startingWidth + shorterSide);
+  const endingHeight = Math.floor(startingHeight + shorterSide);
+
+  // return image data cropped to those points
+  return img.slice([startingHeight, startingWidth, 0], [endingHeight, endingWidth, 3]);
+}
+function resizeImage(image) {
+  return tf.image.resizeBilinear(image, [224, 224]);
+}
+function batchImage(image) {
+  // Expand our tensor to have an additional dimension, whose size is 1
+  const batchedImage = image.expandDims(0);
+
+  // Turn pixel data into a float between -1 and 1.
+  return batchedImage.toFloat().div(tf.scalar(127)).sub(tf.scalar(1));
+}
+function loadAndProcessImage(image) {
+  const croppedImage = cropImage(image);
+  const resizedImage = resizeImage(croppedImage);
+  const batchedImage = batchImage(resizedImage);
+  return batchedImage;
+}
+
+const parseImage = async (src) => {
+  const img = await loadImage(src);
+  const pixels = tf.browser.fromPixels(img);
+  const imageData = loadAndProcessImage(pixels);
+  return imageData;
+};
+
+(async function() {
+  // const dogPixels = await parseImage(dog);
+  // const catPixels = await parseImage(cat);
+  // const images = dogPixels.concat(catPixels);
+
+  const images = [dog, cat];
+  const labels = ['dog', 'cat'];
+  await mlClassifier.addData(images, labels, 'train');
+  mlClassifier.train({
+    callbacks: {
+      onTrainBegin: () => {
+        console.log('training begins');
+      },
+      onTrainEnd: () => {
+        console.log('training ends');
+      },
+    },
+  });
+})();
+

example/package.json

@@ -0,0 +1,30 @@
+{
+  "name": "ml-classifier-example",
+  "version": "0.1.0",
+  "description": "",
+  "main": "index.js",
+  "scripts": {
+    "develop": "parcel index.html"
+  },
+  "devDependencies": {
+    "@babel/core": "^7.0.0-beta.51",
+    "@babel/plugin-proposal-decorators": "^7.0.0-beta.51",
+    "@babel/plugin-transform-classes": "^7.0.0-beta.51",
+    "@babel/plugin-transform-regenerator": "^7.3.4",
+    "@babel/plugin-transform-runtime": "^7.3.4",
+    "@babel/preset-env": "^7.0.0-beta.51",
+    "@babel/preset-stage-0": "^7.0.0-beta.51",
+    "@babel/preset-typescript": "^7.0.0-beta.51",
+    "babel-core": "^7.0.0-0",
+    "babel-jest": "^23.2.0",
+    "babel-plugin-transform-class-properties": "^6.24.1",
+    "babel-plugin-transform-runtime": "^6.23.0",
+    "babel-polyfill": "^6.26.0",
+    "gh-pages": "^2.0.1",
+    "parcel": "^1.11.0"
+  },
+  "dependencies": {
+    "@babel/runtime-corejs2": "^7.3.4",
+    "@tensorflow/tfjs": "^1.0.0"
+  }
+}