https://github.com/ytakzk/CoreML-samples

Sample code for Core ML using ResNet50 provided by Apple and a custom model generated by coremltools.
https://github.com/ytakzk/CoreML-samples

coreml coreml-framework deep-learning ios ios11 ml resnet swift

Last synced: about 1 month ago
JSON representation

Sample code for Core ML using ResNet50 provided by Apple and a custom model generated by coremltools.

• Host: GitHub
• URL: https://github.com/ytakzk/CoreML-samples
• Owner: ytakzk
• License: mit
• Created: 2017-06-14T23:53:43.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-07-20T14:23:35.000Z (over 7 years ago)
• Last Synced: 2024-11-26T02:03:43.583Z (about 2 months ago)
• Topics: coreml, coreml-framework, deep-learning, ios, ios11, ml, resnet, swift
• Language: Jupyter Notebook
• Homepage:
• Size: 89.9 MB
• Stars: 42
• Watchers: 4
• Forks: 16
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-ios - CoreML-samples - Sample code for Core ML using ResNet50 provided by Apple and a custom model generated by coremltools. (Machine Learning / Other Hardware)
• awesome-ios-star - CoreML-samples - Sample code for Core ML using ResNet50 provided by Apple and a custom model generated by coremltools. (Machine Learning / Other Hardware)

README

        
# CoreML-samples

This is the sample code for Core ML using ResNet50 provided by [Apple](https://developer.apple.com/machine-learning/).  

[ResNet50](https://arxiv.org/abs/1512.03385) can categorize the input image to 1000 pre-trained categories.  

What's more, this includes a sample code for coremltools converting keras model to mlmodel.



## Source Code for the prediction 

```

guard let image = imageView.image, let ref = image.buffer() else {

        return

}

do {

    // predict

    let output = try model.prediction(image: ref)

    print(output.classLabel)

    print(output.classLabelProbs)

} catch {

    print(error)

}

```