Get unlimited public & private packages + package-based permissions with npm Pro.Get started »

@cloud-annotations/models

0.0.9 • Public • Published

Cloud Annotations JavaScript SDK

NPM Version NPM Downloads

The Cloud Annotations SDK makes it easy to use your custom trained object detection or classification models in the browser or on the backend with TensorFlow.js. Simply load() your model_web folder and pass the loaded model a <canvas>, <img> or <video> reference.

Installation

npm install @cloud-annotations/models

Usage

Load a model

import models from '@cloud-annotations/models'
 
const model = await models.load('/model_web')
 
const img = document.getElementById('img')
 
// for object detection
const predictions = await model.detect(img)
 
// predictions =>
[{
  label: 'dog',
  bbox: [x, y, width, height],
  score: 0.92
},
{
  label: 'cat',
  bbox: [x, y, width, height],
  score: 0.72
}]
 
// for classification
const predictions = await model.classify(img)
 
// predictions =>
[
  { label: 'dog', score: 0.92 },
  { label: 'cat', score: 0.72 }
]

Object detection

const img = document.getElementById('img')
const predictions = await model.detect(img)
 
// predictions =>
[{
  label: 'dog',
  bbox: [x, y, width, height],
  score: 0.92
},
{
  label: 'cat',
  bbox: [x, y, width, height],
  score: 0.72
}]

Classification

const img = document.getElementById('img')
const predictions = await model.classify(img)
 
// predictions =>
[
  { label: 'dog', score: 0.92 },
  { label: 'cat', score: 0.72 }
]

Usage via Script Tag

No npm install required. Just import via the script tag.

<script src="https://cdn.jsdelivr.net/npm/@cloud-annotations/models"></script>
<script>
  const img = document.getElementById('img')
  models.load('/model_web')
    .then(model => model.detect(img))
    .then(predictions => {
      console.log(predictions)
    })
</script> 

Example usage: Real-Time Object Detection With React.

Install

npm i @cloud-annotations/models

DownloadsWeekly Downloads

14

Version

0.0.9

License

MIT

Unpacked Size

1.77 MB

Total Files

9

Last publish

Collaborators

  • avatar