TextClassificationIOS is an example iOS app that uses the Skafos platform for CoreML model integration and delivery. It's a good starting point for diving in, or a good reference for integrating Skafos in to your own app. Skafos is a platform that streamlines CoreML model updates without needing to submit a new version to the app store everytime a new model is ready for use.
This TextClassification example app specifically integrates and deploys a Text Classification machine learning model. Text Classification is a machine learning technique that enables you to assign tags or categories to arbitary text, based on its content. The example model that is provided in this app trains a sentiment classifier model on Yelp reviews. This sentiment classifier model predicts whether or not the text in question is a positive or negative review. For more details about how to use and customize this model, please navigate to the Skafos Turi Text Classifier repo on github.
Before diving in to this example application, make sure you have setup an account at Skafos.
- Clone or fork this repository.
- In the project directory, run
- Open the project workspace (
- In your project's settings under
Generalchange the following:
- Display Name
- Bundle Identifier
- Any other settings specific to your app.
Skafos Framework Setup
AppDelegate.swift make sure to set your Skafos environment keys in:
Skafos.initialize. You can find these in your App Settings on the dashboard.
Now take a moment to click on
TextClassifier.mlmodel and under Model Class section click the arrow next
TextClassifier and have a peek at the class that Xcode generates from the CoreML Model. Now, inside of
MainViewController.swift take a look at the
viewDidAppear function to see an example of
how to load a model using the Skafos framework.
Skafos swift framework uses the Apache2 license, located in the LICENSE file.
Questions? Need Help?
Contact us by email firstname.lastname@example.org