PredicSis ML SDK for Javascript.

The official PredicSis ML SDK for Javascript (for pure frontend applications, without node).

Installation

Add PredicSis ML SDK into your bower.json

bower install predicsis_ml_sdk-javascript --save-dev

Configuration

If you want to change PredicSis API host, add these lines to your app.js configuration file.

angular
  .module('YourAngularApplication', ['predicsis.jsSDK'])
  .config(function(predicsisAPIProvider, config) {
    predicsisAPIProvider.setBaseUrl('https://predicsis_api_host');
    predicsisAPIProvider.setOauthToken('4V83j2BWgcPONK8Xw8P7953yPvnTzz784V83j2BWgcPONK8Xw8P7953yPvnTzz78');
  })

  .controller('ExampleCtrl', function(predicsisAPI) {
    var self = this;
    predicsisAPI.Datasets.all().then(function(datasetList) { self.datasets = datasetList; });
  });

You can also setup your personal access token within a run context. This can be useful if your token is stored in a cookie or in localStorage.

angular
  .module('YourAngularApplication', ['predicsis.jsSDK'])
  .config(function(predicsisAPIProvider) {
    predicsisAPIProvider.setBaseUrl('http://localhost:8003');
  })

  .run(function(predicsisAPI, $cookieStore) {
    predicsisAPI.setOauthToken($cookieStore.get('session'));
    predicsisAPI.setErrorHandler(console.log.bind(console));
  })

  .controller('ExampleCtrl', function(predicsisAPI) {
    var self = this;
    predicsisAPI.Datasets.all().then(function(datasetList) { self.datasets = datasetList; });
  });

If you want to send API requests with no Authorization header, you must explicitly use predicsisAPI.setOauthToken(false);

Default values:

• api endpoint: https://api.predicsis.com
• error handler: throw Error(response);
• oauth access token: no-token-defined

Getting Started

You can start using our SDK assuming you already have a user token

Once you're done, you can send API request easily:

• list projects:

  predicsisAPI.Projects
    .all()
    .then(function(projectList) { console.log(projectList); })
  

• generate a dictionary from a dataset:

  predicsisAPI.Dictionaries
    .create({dataset_id: '53c7dea470632d3417020000', name: 'My dictionary'})
    .then(function(dictionary) { console.log(dictionary); });
  

• start a learn process:

  predicsisAPI.Models.learn(project)
  

• ...

See the SDK documentation for more examples.

Upload a file to S3

//Get an HTML5 File instance
fileInput.addEventListener('change', function(evt) {
  var file = evt.target.files[0];
  predicsisAPI.uploadHelper
    .processFile(file, { chunkSize: 50 * 1024 * 1024 });
  $rootScope.$on('jsSDK.upload.starting', function(event, upload) {
    console.log(upload.id, upload.fileName, upload.fileSize, upload.progression);
  });
  $rootScope.$on('jsSDK.upload.progress', function(event, upload) {
    console.log(upload.id, upload.fileName, upload.fileSize, upload.progression, upload.path);
  });
  $rootScope.$on('jsSDK.upload.uploaded', function(event, upload) {
    console.log(upload.id, upload.fileName, upload.fileSize, upload.progression, upload.path);
  });
  $rootScope.$on('jsSDK.upload.cancelled', function(event, upload) {
    console.log(upload.id, upload.fileName, upload.fileSize, upload.progression, upload.path);
  });
  $rootScope.$on('jsSDK.upload.error', function(event, failure, retry) {
    console.log(failure.id, failure.fileName, failure.path, failure.err);
    retryButton.addEventListener('click', retry);
  });
});

Run this project locally

This project comes with a little angular application (located in the app folder). After a complete install, just run:

  bower install
  npm install

  # start the server
  grunt connect:server

And go to http://localhost:8100/app/index.html.

Getting Help

• API documentation
• SDK documentation

License

MIT. See the LICENSE for more details.

Contributing

1. Fork it
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create new Pull Request