neural_network_inference

Neural Network Inference Examples

In this set of notebook examples, we show examples for building and editing mlmodels via coremltools.

• Setting per channel scale image pre-processing
  This notebook shows how an mlmodel can be edited after conversion, to add to it a per channel scale pre-processing layer.

• Power iteration algorithm using a neural network Core ML model
  This notebook shows how to build an mlmodel from scratch using the neural network builder API in coremltools. In particular, this examples shows how a non-neural network algorithm involving control flow features can still be expressed as an mlmodel. The feature of using flexible shaped inputs is also touched upon at the end of the notebook.

• Tensorflow 1 conversion examples

• Tensorflow 2 conversion examples

• ONNX conversion examples