from Hacker News

Core ML: Integrate machine learning models into your app

by yonilevy on 6/6/17, 2:51 AM with 40 comments

  • by eggie5 on 6/6/17, 6:57 AM

    Apple's new iOS CoreML inference engine supports Keras models! Developers will be able to design and train model using Keras and then convert the architecture to run on the CoreML engine. I suppose you can run TensorFlow models too if you designed them w/ Keras.
  • by blt on 6/6/17, 1:07 PM

    Cool that the file format is officially specified as a Protocol Buffer: https://developer.apple.com/documentation/coreml/converting_... (bottom of page)
  • by seanmcdirmid on 6/6/17, 7:09 AM

    When "Core ML" is no longer a mini language to reason about ML the PL semantics...
  • by singularity2001 on 6/6/17, 7:02 AM

    "BNNS does not do training, however. Its purpose is to provide very high performance inference on already trained neural networks." :(

    https://developer.apple.com/documentation/accelerate/bnns

  • by kensai on 6/6/17, 6:58 AM

    "Core ML is optimized for on-device performance, which minimizes memory footprint and power consumption."

    This is major, if they have managed to achieve it reasonably. But before opening a Sekt, I want to see some benchmarks. :)

  • by jamesswift3 on 6/6/17, 1:49 PM

    Federighi says Core ML on iPhone is 6x faster than Google Pixel and Samsung Galaxy S8.

    How they actually compare?

  • by hackerbot on 6/6/17, 8:36 AM

    using it means need to handle Android with another framework separately
  • by db3d on 6/6/17, 2:11 PM

    Lots of options to explore, but no reinforcement learning yet.

    Also, some converted Core ML Models ready to use here: developer.apple.com/machine-learning

  • by ajay-d on 6/6/17, 3:39 PM

    The python package, coremltools, to convert the trained model is only for python 2.7??
  • by dmix on 6/6/17, 5:48 PM

    Having a directory of trained models to download is an interesting concept. This will certainly accelerate the adoption of ML.
  • by OldeElk on 6/6/17, 5:39 PM

    Does it support caffe/TensorFlow/MXnet model inferencing?
  • by mtw on 6/6/17, 10:15 AM

    Why not caffe2? And why mention libsvm, are we in the 90s?
  • by killin_dan on 6/6/17, 7:20 AM

    Machine Learning needs a new acronym. ML is already taken! Get your own!