bug fixes, release 1.1.4

bonastreyair · bonastreyair · commit dfa182cfaeb7 · 2020-06-05T17:22:01.000+02:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,6 +1,14 @@
 # Changelog
 All notable changes to this project will be documented in this file.
 
+## [1.1.4] - 2020-06-05
+### Fixed
+ - Better usage of `HTMLVideoElement` class imported from `dom.window`
+ - Better accuracy representation in the node status without decimals
+
+### Changed
+ - Some variable types from `const` to `var`
+
 ## [1.1.3] - 2020-06-05
 ### Fixed
  - Prediction does not work when save_image's box is checked - [#14](https://github.com/bonastreyair/node-red-contrib-teachable-machine/issues/14)
diff --git a/package.json b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "node-red-contrib-teachable-machine",
-  "version": "1.1.3",
+  "version": "1.1.4",
   "description": "Simplifies integration with Teachable Machine models from Google",
   "dependencies": {
     "@tensorflow/tfjs": "1.3.1",
diff --git a/teachable_machine.js b/teachable_machine.js
@@ -2,14 +2,13 @@ module.exports = function (RED) {
   /* Initial Setup */
   // Simulate real HTML
   const { JSDOM } = require('jsdom')
-  var dom = new JSDOM('<!doctype html><html></html>')
+  var dom = new JSDOM('')
   global.document = dom.window.document
-  // Require basic libraries
-  const tmImage = require('@teachablemachine/image')
+  global.HTMLVideoElement = dom.window.HTMLVideoElement
   const canvas = require('canvas')
+  // Require basic libraries
   global.fetch = require('node-fetch')
-  // Teachable Machine needs global scope of HTMLVideoElement class to do a check
-  global.HTMLVideoElement = class HTMLVideoElement { }
+  const tmImage = require('@teachablemachine/image')
 
   function setNodeStatus (node, status) {
     switch (status) {
@@ -79,8 +78,8 @@ module.exports = function (RED) {
     }
 
     function getBestPrediction (predictions) {
-      let className = ''
-      let probability = 0
+      var className = ''
+      var probability = 0
       for (let i = 0; i < predictions.length; i++) {
         if (predictions[i].probability > probability) {
           className = predictions[i].className
@@ -97,7 +96,7 @@ module.exports = function (RED) {
     }
 
     function changeKeyResults (results) {
-      const out = []
+      var out = []
       for (let i = 0; i < results.length; i++) {
         out.push({
           class: results[i].className,
@@ -110,35 +109,36 @@ module.exports = function (RED) {
     // Converts the image, makes inference and treats predictions
     async function inference (msg) {
       setNodeStatus(node, 'infering')
-      const image = new canvas.Image()
+      var image = new canvas.Image()
       image.src = msg.image
-      msg.classes = node.model.getClassLabels()
-      const predictions = await node.model.predict(image)
+
+      var predictions = await node.model.predict(image)
 
       predictions.sort(byProbabilty)
-      const percentage = predictions[0].probability.toFixed(2) * 100
-      const bestPredictionText = percentage.toString() + '% - ' + predictions[0].className
+      var percentage = (predictions[0].probability * 100).toFixed(0)
+      var bestPredictionText = percentage.toString() + '% - ' + predictions[0].className
 
       if (node.output === 'best') {
         msg.payload = [{ class: predictions[0].className, score: predictions[0].probability }]
         setNodeStatus(node, bestPredictionText)
       } else if (node.output === 'all') {
-        let filteredPredictions = predictions
+        var filteredPredictions = predictions
         filteredPredictions = node.activeThreshold ? filteredPredictions.filter(prediction => prediction.probability > node.threshold / 100) : filteredPredictions
         filteredPredictions = node.activeMaxResults ? filteredPredictions.slice(0, node.maxResults) : filteredPredictions
         filteredPredictions = changeKeyResults(filteredPredictions)
 
         if (filteredPredictions.length > 0) {
           setNodeStatus(node, bestPredictionText)
         } else {
-          const statusText = 'score < ' + node.threshold + '%'
+          var statusText = 'score < ' + node.threshold + '%'
           setNodeStatus(node, statusText)
           msg.payload = []
           node.send(msg)
           return
         }
         msg.payload = filteredPredictions
       }
+      msg.classes = node.model.getClassLabels()
       node.send(msg)
     }
 

Original file line number	Diff line number	Diff line change
`@@ -1,6 +1,6 @@`
`1`	`1`	`{`
`2`	`2`	`"name": "node-red-contrib-teachable-machine",`
`3`		`- "version": "1.1.3",`
	`3`	`+ "version": "1.1.4",`
`4`	`4`	`"description": "Simplifies integration with Teachable Machine models from Google",`
`5`	`5`	`"dependencies": {`
`6`	`6`	`"@tensorflow/tfjs": "1.3.1",`