coremltools : Working around with models in iOS devices
Akanksha Sharma (~akanksharma) |
Core ML is an Apple Framework for integrating machine learning on apps running across all its platforms. coremltools is a python package for creating, examining and testing models. Apple has provided a public file format .mlmodel for a broad set of ML methods including deep neural networks (both convolutional and recurrent), tree ensembles with boosting, and generalized linear models. Models in this format can be directly integrated into apps through Xcode. Following three operations can be done by coremltools : - Converting a trained model to Core ML format. - Using a Core ML model to make predictions from Python. - Changing properties of an existing Core ML model.
Topics Covered in the talk
The talk will a guide for converting a trained (TFlite) model to .mlmodel and use it within an app for making predictions. Following topics would be covered
- Understanding the prerequisites for creating .mlmodel
- Converting a trained model
- Adding metadata
- Creating a pipeline
- Getting the insights of the model
- Finally adding it in the app.
- Questions and Answers
- Basic idea of machine learning
- Basic idea of Core ML
Working on content. The code will be pushed here : Github repo
Slides will be updated here : coremltools
I am working as a Senior Software Engineer in a MNC located in Mumbai. With total 6+ years of experience in iOS app development, I have a special interest in Machine Learning and CoreML.
In one of the app Flo, we used CoreML create an Intelligent Video Editor for iPhones which was featured by Apple in 2017. Along with this, I have gained decent experience in creating apps for varied domains. I am an active member of developer groups and meet-ups. When free, I read about politics and binge watch.