It has two inputs and one output.
Input: [Temperature, Humidity]
Output: [wattage]
I learned as follows
Even after 5 million rotations, it does not work properly.
Did I choose the wrong option?
var input_data = [
[-2.4,2.7,9,14.2,17.1,22.8,281,25.9,22.6,15.6,8.2,0.6],
[58,56,63,54,68,73,71,74,71,70,68,62]
];
var power_data = [239,224,189,189,179,192,243,317,224,190,189,202];
var reason_data = tf.tensor2d(input_data);
var result_data = tf.tensor(power_data);
var X = tf.input({ shape: [2] });
var Y = tf.layers.dense({ units: 1 }).apply(X);
var model = tf.model({ inputs: X, outputs: Y });
var compileParam = { optimizer: tf.train.adam(), loss: tf.losses.meanSquaredError }
model.compile(compileParam);
var fitParam = {
epochs: 500000,
callbacks: {
onEpochEnd: function (epoch, logs) {
console.log('epoch', epoch, logs, "RMSE --> ", Math.sqrt(logs.loss));
}
}
}
model.fit(reason_data, result_data, fitParam).then(function (result) {
var final_result = model.predict(reason_data);
final_result.print();
model.save('file:///path/');
});
The following is the result for 5 million times.
It should be the same as power_data , but it failed.
What should I fix?
While there is rarely one simple reason to point to when a model doesn't perform the way you would expect, here are some options to consider:
You don't have enough data points. Twelve is not nearly sufficient to get an accurate result.
You need to normalize the data of the input tensors. Given that your two features [temperature and humidity] have different ranges, they need to be normalized to give them equal opportunity to influence the output. The following is a normalization function you could start with:
function normalize(tensor, min, max) {
const result = tf.tidy(function() {
// Find the minimum value contained in the Tensor.
const MIN_VALUES = min || tf.min(tensor, 0);
// Find the maximum value contained in the Tensor.
const MAX_VALUES = max || tf.max(tensor, 0);
// Now calculate subtract the MIN_VALUE from every value in the Tensor
// And store the results in a new Tensor.
const TENSOR_SUBTRACT_MIN_VALUE = tf.sub(tensor, MIN_VALUES);
// Calculate the range size of possible values.
const RANGE_SIZE = tf.sub(MAX_VALUES, MIN_VALUES);
// Calculate the adjusted values divided by the range size as a new Tensor.
const NORMALIZED_VALUES = tf.div(TENSOR_SUBTRACT_MIN_VALUE, RANGE_SIZE);
// Return the important tensors.
return {NORMALIZED_VALUES, MIN_VALUES, MAX_VALUES};
});
return result;
}
You should try a different optimizer. Adam might be the best choice, but for a linear regression problem such as this, you should also consider Stochastic Gradient Descent (SGD).
Check out this sample code for an example that uses normalization and sgd. I ran your data points through the code (after making the changes to the tensors so they fit your data), and I was able to reduce the loss to less than 40. There is room for improvement, but that's where adding more data points comes in.
I want to detect played notes and chords using the Web Audio API (using the microphone as an input device). Before I can analyse the data, I need it the individual frequencies mapped to their loudness. I started with the following snippet:
const stream = await navigator.mediaDevices.getUserMedia({
audio: true,
video: false
});
const context = new AudioContext();
const source = context.createMediaStreamSource(stream);
const analyser = context.createAnalyser();
const data = new Uint8Array(analyser.frequencyBinCount);
analyser.getByteFrequencyData(data);
data now is an array of values between 0 to 255. The question I have now is how can I map the frequencies to the loudness values of the data array?
Ideally, I'd like an object like this:
{
...
438: 128,
439: 200,
440: 255,
441: 200,
...
}
Thanks for your help.
The value in data[k] corresponds to the frequency k * Nyquist/frequencyBinCount where Nyquist is one half of the sampling frequency, AudioContext.sampleRate.
I think that's what you're asking for. If not please clarify.
I want to make a image classifier, but I don't know python.
Tensorflow.js works with javascript, which I am familiar with. Can models be trained with it and what would be the steps to do so?
Frankly I have no clue where to start.
The only thing I figured out is how to load "mobilenet", which apparently is a set of pre-trained models, and classify images with it:
const tf = require('#tensorflow/tfjs'),
mobilenet = require('#tensorflow-models/mobilenet'),
tfnode = require('#tensorflow/tfjs-node'),
fs = require('fs-extra');
const imageBuffer = await fs.readFile(......),
tfimage = tfnode.node.decodeImage(imageBuffer),
mobilenetModel = await mobilenet.load();
const results = await mobilenetModel.classify(tfimage);
which works, but it's no use to me because I want to train my own model using my images with labels that I create.
=======================
Say I have a bunch of images and labels. How do I use them to train a model?
const myData = JSON.parse(await fs.readFile('files.json'));
for(const data of myData){
const image = await fs.readFile(data.imagePath),
labels = data.labels;
// how to train, where to pass image and labels ?
}
First of all, the images needs to be converted to tensors. The first approach would be to create a tensor containing all the features (respectively a tensor containing all the labels). This should the way to go only if the dataset contains few images.
const imageBuffer = await fs.readFile(feature_file);
tensorFeature = tfnode.node.decodeImage(imageBuffer) // create a tensor for the image
// create an array of all the features
// by iterating over all the images
tensorFeatures = tf.stack([tensorFeature, tensorFeature2, tensorFeature3])
The labels would be an array indicating the type of each image
labelArray = [0, 1, 2] // maybe 0 for dog, 1 for cat and 2 for birds
One needs now to create a hot encoding of the labels
tensorLabels = tf.oneHot(tf.tensor1d(labelArray, 'int32'), 3);
Once there is the tensors, one would need to create the model for training. Here is a simple model.
const model = tf.sequential();
model.add(tf.layers.conv2d({
inputShape: [height, width, numberOfChannels], // numberOfChannels = 3 for colorful images and one otherwise
filters: 32,
kernelSize: 3,
activation: 'relu',
}));
model.add(tf.layers.flatten());
model.add(tf.layers.dense({units: 3, activation: 'softmax'}));
Then the model can be trained
model.fit(tensorFeatures, tensorLabels)
If the dataset contains a lot of images, one would need to create a tfDataset instead. This answer discusses why.
const genFeatureTensor = image => {
const imageBuffer = await fs.readFile(feature_file);
return tfnode.node.decodeImage(imageBuffer)
}
const labelArray = indice => Array.from({length: numberOfClasses}, (_, k) => k === indice ? 1 : 0)
function* dataGenerator() {
const numElements = numberOfImages;
let index = 0;
while (index < numFeatures) {
const feature = genFeatureTensor(imagePath);
const label = tf.tensor1d(labelArray(classImageIndex))
index++;
yield {xs: feature, ys: label};
}
}
const ds = tf.data.generator(dataGenerator).batch(1) // specify an appropriate batchsize;
And use model.fitDataset(ds) to train the model
The above is for training in nodejs. To do such a processing in the browser, genFeatureTensor can be written as follow:
function loadImage(url){
return new Promise((resolve, reject) => {
const im = new Image()
im.crossOrigin = 'anonymous'
im.src = 'url'
im.onload = () => {
resolve(im)
}
})
}
genFeatureTensor = image => {
const img = await loadImage(image);
return tf.browser.fromPixels(image);
}
One word of caution is that doing heavy processing might block the main thread in the browser. This is where web workers come into play.
Consider the exemple https://codelabs.developers.google.com/codelabs/tfjs-training-classfication/#0
What they do is:
take a BIG png image (a vertical concatenation of images)
take some labels
build the dataset (data.js)
then train
The building of the dataset is as follows:
images
The big image is divided into n vertical chunks.
(n being chunkSize)
Consider a chunkSize of size 2.
Given the pixel matrix of image 1:
1 2 3
4 5 6
Given the pixel matrix of image 2 is
7 8 9
1 2 3
The resulting array would be
1 2 3 4 5 6 7 8 9 1 2 3 (the 1D concatenation somehow)
So basically at the end of the processing, you have a big buffer representing
[...Buffer(image1), ...Buffer(image2), ...Buffer(image3)]
labels
That kind of formatting is done a lot for classification problems. Instead of classifying with a number, they take a boolean array.
To predict 7 out of 10 classes we would consider
[0,0,0,0,0,0,0,1,0,0] // 1 in 7e position, array 0-indexed
What you can do to get started
Take your image (and its associated label)
Load your image to the canvas
Extract its associated buffer
Concatenate all your image's buffer as a big buffer. That's it for xs.
Take all your associated labels, map them as a boolean array, and concatenate them.
Below, I subclass MNistData::load (the rest can be let as is (except in script.js where you need to instantiate your own class instead)
I still generate 28x28 images, write a digit on it, and get a perfect accuracy since I don't include noise or voluntarily wrong labelings.
import {MnistData} from './data.js'
const IMAGE_SIZE = 784;// actually 28*28...
const NUM_CLASSES = 10;
const NUM_DATASET_ELEMENTS = 5000;
const NUM_TRAIN_ELEMENTS = 4000;
const NUM_TEST_ELEMENTS = NUM_DATASET_ELEMENTS - NUM_TRAIN_ELEMENTS;
function makeImage (label, ctx) {
ctx.fillStyle = 'black'
ctx.fillRect(0, 0, 28, 28) // hardcoded, brrr
ctx.fillStyle = 'white'
ctx.fillText(label, 10, 20) // print a digit on the canvas
}
export class MyMnistData extends MnistData{
async load() {
const canvas = document.createElement('canvas')
canvas.width = 28
canvas.height = 28
let ctx = canvas.getContext('2d')
ctx.font = ctx.font.replace(/\d+px/, '18px')
let labels = new Uint8Array(NUM_DATASET_ELEMENTS*NUM_CLASSES)
// in data.js, they use a batch of images (aka chunksize)
// let's even remove it for simplification purpose
const datasetBytesBuffer = new ArrayBuffer(NUM_DATASET_ELEMENTS * IMAGE_SIZE * 4);
for (let i = 0; i < NUM_DATASET_ELEMENTS; i++) {
const datasetBytesView = new Float32Array(
datasetBytesBuffer, i * IMAGE_SIZE * 4,
IMAGE_SIZE);
// BEGIN our handmade label + its associated image
// notice that you could loadImage( images[i], datasetBytesView )
// so you do them by bulk and synchronize after your promises after "forloop"
const label = Math.floor(Math.random()*10)
labels[i*NUM_CLASSES + label] = 1
makeImage(label, ctx)
const imageData = ctx.getImageData(0, 0, canvas.width, canvas.height);
// END you should be able to load an image to canvas :)
for (let j = 0; j < imageData.data.length / 4; j++) {
// NOTE: you are storing a FLOAT of 4 bytes, in [0;1] even though you don't need it
// We could make it with a uint8Array (assuming gray scale like we are) without scaling to 1/255
// they probably did it so you can copy paste like me for color image afterwards...
datasetBytesView[j] = imageData.data[j * 4] / 255;
}
}
this.datasetImages = new Float32Array(datasetBytesBuffer);
this.datasetLabels = labels
//below is copy pasted
this.trainIndices = tf.util.createShuffledIndices(NUM_TRAIN_ELEMENTS);
this.testIndices = tf.util.createShuffledIndices(NUM_TEST_ELEMENTS);
this.trainImages = this.datasetImages.slice(0, IMAGE_SIZE * NUM_TRAIN_ELEMENTS);
this.testImages = this.datasetImages.slice(IMAGE_SIZE * NUM_TRAIN_ELEMENTS);
this.trainLabels =
this.datasetLabels.slice(0, NUM_CLASSES * NUM_TRAIN_ELEMENTS);// notice, each element is an array of size NUM_CLASSES
this.testLabels =
this.datasetLabels.slice(NUM_CLASSES * NUM_TRAIN_ELEMENTS);
}
}
I found a tutorial [1] how to use existing model to train new classes. Main code parts here:
index.html head:
<script src="https://unpkg.com/#tensorflow-models/knn-classifier"></script>
index.html body:
<button id="class-a">Add A</button>
<button id="class-b">Add B</button>
<button id="class-c">Add C</button>
index.js:
const classifier = knnClassifier.create();
....
// Reads an image from the webcam and associates it with a specific class
// index.
const addExample = async classId => {
// Capture an image from the web camera.
const img = await webcam.capture();
// Get the intermediate activation of MobileNet 'conv_preds' and pass that
// to the KNN classifier.
const activation = net.infer(img, 'conv_preds');
// Pass the intermediate activation to the classifier.
classifier.addExample(activation, classId);
// Dispose the tensor to release the memory.
img.dispose();
};
// When clicking a button, add an example for that class.
document.getElementById('class-a').addEventListener('click', () => addExample(0));
document.getElementById('class-b').addEventListener('click', () => addExample(1));
document.getElementById('class-c').addEventListener('click', () => addExample(2));
....
Main idea is to use existing network to make its prediction and then substitute the found label with your own one.
Complete code is in the tutorial. Another promising, more advanced one in [2]. It needs strict pre processing, so I leave it only here, I mean it is so much more advanced one.
Sources:
[1] https://codelabs.developers.google.com/codelabs/tensorflowjs-teachablemachine-codelab/index.html#6
[2] https://towardsdatascience.com/training-custom-image-classification-model-on-the-browser-with-tensorflow-js-and-angular-f1796ed24934
TL;DR
MNIST is the image recognition Hello World. After learning it by heart, these questions in your mind are easy to solve.
Question setting:
Your main question written is
// how to train, where to pass image and labels ?
inside your code block. For those I found perfect answer from examples of Tensorflow.js examples section: MNIST example. My below links have pure javascript and node.js versions of it and Wikipedia explanation. I will go them through on the level necessary to answer the main question in your mind and I will add also perspectives how your own images and labels have anything to do with MNIST image set and the examples using it.
First things first:
Code snippets.
where to pass images (Node.js sample)
async function loadImages(filename) {
const buffer = await fetchOnceAndSaveToDiskWithBuffer(filename);
const headerBytes = IMAGE_HEADER_BYTES;
const recordBytes = IMAGE_HEIGHT * IMAGE_WIDTH;
const headerValues = loadHeaderValues(buffer, headerBytes);
assert.equal(headerValues[0], IMAGE_HEADER_MAGIC_NUM);
assert.equal(headerValues[2], IMAGE_HEIGHT);
assert.equal(headerValues[3], IMAGE_WIDTH);
const images = [];
let index = headerBytes;
while (index < buffer.byteLength) {
const array = new Float32Array(recordBytes);
for (let i = 0; i < recordBytes; i++) {
// Normalize the pixel values into the 0-1 interval, from
// the original 0-255 interval.
array[i] = buffer.readUInt8(index++) / 255;
}
images.push(array);
}
assert.equal(images.length, headerValues[1]);
return images;
}
Notes:
MNIST dataset is a huge image, where in one file there are several images like tiles in puzzle, each and every with same size, side by side, like boxes in x and y coordination table. Each box has one sample and corresponding x and y in the labels array has the label. From this example, it is not a big deal to turn it to several files format, so that actually only one pic at a time is given to the while loop to handle.
Labels:
async function loadLabels(filename) {
const buffer = await fetchOnceAndSaveToDiskWithBuffer(filename);
const headerBytes = LABEL_HEADER_BYTES;
const recordBytes = LABEL_RECORD_BYTE;
const headerValues = loadHeaderValues(buffer, headerBytes);
assert.equal(headerValues[0], LABEL_HEADER_MAGIC_NUM);
const labels = [];
let index = headerBytes;
while (index < buffer.byteLength) {
const array = new Int32Array(recordBytes);
for (let i = 0; i < recordBytes; i++) {
array[i] = buffer.readUInt8(index++);
}
labels.push(array);
}
assert.equal(labels.length, headerValues[1]);
return labels;
}
Notes:
Here, labels are also byte data in a file. In Javascript world, and with the approach you have in your starting point, labels could also be a json array.
train the model:
await data.loadData();
const {images: trainImages, labels: trainLabels} = data.getTrainData();
model.summary();
let epochBeginTime;
let millisPerStep;
const validationSplit = 0.15;
const numTrainExamplesPerEpoch =
trainImages.shape[0] * (1 - validationSplit);
const numTrainBatchesPerEpoch =
Math.ceil(numTrainExamplesPerEpoch / batchSize);
await model.fit(trainImages, trainLabels, {
epochs,
batchSize,
validationSplit
});
Notes:
Here model.fit is the actual line of code that does the thing: trains the model.
Results of the whole thing:
const {images: testImages, labels: testLabels} = data.getTestData();
const evalOutput = model.evaluate(testImages, testLabels);
console.log(
`\nEvaluation result:\n` +
` Loss = ${evalOutput[0].dataSync()[0].toFixed(3)}; `+
`Accuracy = ${evalOutput[1].dataSync()[0].toFixed(3)}`);
Note:
In Data Science, also this time here, the most faschinating part is to know how well the model survives the test of new data and no labels, can it label them or not? For that is the evaluation part that now prints us some numbers.
Loss and accuracy: [4]
The lower the loss, the better a model (unless the model has over-fitted to the training data). The loss is calculated on training and validation and its interperation is how well the model is doing for these two sets. Unlike accuracy, loss is not a percentage. It is a summation of the errors made for each example in training or validation sets.
..
The accuracy of a model is usually determined after the model parameters are learned and fixed and no learning is taking place. Then the test samples are fed to the model and the number of mistakes (zero-one loss) the model makes are recorded, after comparison to the true targets.
More information:
In the github pages, in README.md file, there is a link to tutorial, where all in the github example is explained in greater detail.
[1] https://github.com/tensorflow/tfjs-examples/tree/master/mnist
[2] https://github.com/tensorflow/tfjs-examples/tree/master/mnist-node
[3] https://en.wikipedia.org/wiki/MNIST_database
[4] How to interpret "loss" and "accuracy" for a machine learning model
I'm working on a project and I need to add 3d sounds effects, like the sound is continually moving around the listener effects. Is it possible to achieve that with howlerjs i see that with howler i'm able to play a sound from specific coordinates/orientation but how to achieve surrounding/ambisonics sounds ?
Or another library in JavaScript to achieve that?
Thanks for your help.
Half a year late, but yeah that's entirely possible in howler.js, haven't used it myself but judging from the docs you can just update the position. there's some more libraries that do it that I've found, check here how 3dage does exactly what you want:
https://codepen.io/naugtur/pen/QgmvOB?editors=1010
var world = IIIdage.World({
tickInterval: 200
})
var annoyingFly = IIIdage.Thing({
is: ['fly'],
sounds: {
'buzzing constantly': {
sound: 'buzz',
times: Infinity
}
},
reacts: [
{
// to: world.random.veryOften(),
to: world.time.once(),
with: 'buzzing constantly'
}
]
})
// scene should create and expose a default world or accept one
var scene = IIIdage.Scene({
title: 'Annoying fly',
library: {
sounds: {
'buzz': {
src: ['https://webaudiogaming.github.io/3dage/fly.mp3']
}
}
},
world: world,
things: [ // scene iterates all things and spawns them into the world. same can be done manually later on.
annoyingFly({
pos: [-1, -15, 0],
dir: [1, 0, 0],
v: 1
})
]
}).load().run()
setTimeout(function () {
scene.dev.trace(IIIdage.dev.preview.dom())
}, 500)
setInterval(function rotateVector() {
var angleRad = 0.15
var d=scene.things[0].attributes.dir
var x=d[0], y=d[1]
var cosAngle = Math.cos(angleRad), sinAngle = Math.sin(angleRad)
scene.things[0].attributes.dir = [x * cosAngle - y * sinAngle, y * cosAngle + x * sinAngle, 0]
}, 500)
window.scene = scene
There's still some others that do similar stuff:
https://www.npmjs.com/package/songbird-audio
https://www.npmjs.com/package/ambisonics
Hope this pushes you in the right direction if you still want help with it.
I'm using cordova-plugin-audioinput plugin for a JavaScript app that I'm developing. I'm trying to get the different decibel values at different frequencies in realtime using this code:
function startCapture() {
audioinput.start({
audioSourceType: 9,
sampleRate: 44100,
streamToWebAudio: true
});
audioCtx = audioinput.getAudioContext();
analyser = audioCtx.createAnalyser();
analyser.fftSize = 8192;
analyser.maxDecibels = 0;
audioinput.connect(analyser)
bufferLength = analyser.frequencyBinCount;
dataArray = new Uint8Array(bufferLength);
}
The data gets saved into the dataArray using analyser.getByteFrequencyData(dataArray);
Even though I specify maxDecibels to 0, the dataArray gets filled with positive values which doesn't really make any sense to me. I need the end result to be in decibels and even though the values that get pushed into dataArray react accordingly to the volume in realtime, they're not in decibels.
The values returned by getByteFrequencyData() are always in the range from 0 to 255. And these values are mapped linearly from minDecibels to maxDecibels. See the getByteFrequencyData.