Deep Learning for Computer Vision

iPhone

Apple used CoreML framework to integrate machine learning in the applications of iPhone. Apple provides a standard list of models that can be integrated directly into the application. You can train a custom deep learning models using TensorFlow and use that in iPhone. In order to deploy custom models, you have to covert the TensorFlow in CoreML framework model. Google released https://github.com/tf-coreml/tf-coreml for converting TensorFlow models into CoreML models. TFcoreML can be installed using the following code:

pip install -U tfcoreml

The model can be exported using the following code:

import tfcoreml as tf_convertertf_converter.convert(tf_model_path='tf_model_path.pb',                     mlmodel_path='mlmodel_path.mlmodel',                     output_feature_names ...