This project is a demo on using CoreML framework for sentiment analysis of text. .mlmodel was developed from Scikit-learn Pipeline using coremltools python package.

You can also use it in the form of plug and play. It is very simple and real quick.

Steps:

1. Drag and drop SentimentClassificationService.Swift and VKSentimentAnalysis.mlmodel into your project.

2. sentimentLabel.text = SentimentClassificationService.instance.prediction(from: "I love CoreML")?.emoji.

We are done.

If you would like to have a glimpse of the whole code and VKSentimentAnalysis.mlmodel description, you can download the project and run the sample project.

Bonus:

This project covers how to use 3D Touch to implement peek and pop using Custom Presentation Controllers. It also shows how to add a Today Extension to the project.