IBM Watson Visual Recognition API
This PHP library provides integration with the IBM Watson Visual Recognition service.
See API documentation.
The structure of this library was inspired by the Omnipay suite of packages.
One of the things that this means is the library doesn't hide away that fact that API requests are being made. For each API request, the request should be created with parameters passed in, sent and then response reviewed.
All requests are available via the Client.
use Bobbyshaw\WatsonVisualRecognition\Client; use Bobbyshaw\WatsonVisualRecognition\Classifier; $client = new Client();
The client should then be initialized with parameters, e.g. username and password (your IBM Watson Service credentials).
$client->initialize(['username' => 'abcdef', 'password' => '12356])
The following can be used find default parameters
Each method matches an API request and returns a request object ready to be sent. Most of the time you'll want to send the request straight away.
$request = $client->getClassifiers(); $response = $request->send();
Each response has a class which helps to manage the response, e.g with the Classifier and Image classes.
/** @var Classifier $classifiers */ $classifiers = $response->getClassifiers();
The library also comes with a set of commands to use on the command line
php app/console classifiers:get [-d|--version-date="..."] username password
php app/console classifiers:classify [-c|--classifiers="..."] [-d|--version-date="..."] username password images
Get Classifier Info
php app/console classifier:get [-d|--version-date="..."] username password classifier_id
php app/console classifier:create [-d|--version-date="..."] username password positive_examples negative_examples name
php app/console classifier:delete [-d|--version-date="..."] username password classifier_id
Run phpunit tests with:
This is also using grumphp to automatically check for PSR style formatting as well.
Test images are provided by Pixabay.
PHPDocumentor is being used for creating library documentation. So make sure to add function comments
You may need to install graphViz on your machine if you don't already have it. This can be done on mac with homebrew
brew install graphviz