PyTorch to CoreML: Writing custom layers with Metal

pytorch-coreml-custom-layer-example

PyTorch to CoreML: Writing custom layers with Metal

Convert PyTorch model

cd Convert
python3 -m pip install -r requirements.txt
python3 convert.py -o /path/to/output/resulting/mlmodel/file

Run the demo app

Open TorchCoreMLDemo.xcodeproj in Xcode, build and run the app.

Shader

The code of the grid sample Metal-shader is here.

More information about grid sample (warper) algorithm you can find here.

GitHub

View Github