what's new on MLKit
Generated on 8/7/2024
1 search
What's New in MLKit
While the context provided does not specifically mention MLKit, it does cover a range of updates and new features in Apple's machine learning ecosystem, which might be relevant to your interests. Here are some highlights:
-
CreateML Enhancements:
- Object Tracking Template: The CreateML app now includes an object tracking template, which is particularly useful for creating spatial experiences on visionOS. This allows you to train reference objects to anchor spatial experiences.
- Time Series Classification and Forecasting: New components for time series classification and forecasting are available, making it easier to integrate these capabilities into your app.
- Data Annotation: Improved tools for exploring and inspecting data annotations prior to training.
For more details, you can check out the session What’s new in Create ML (03:27).
-
Core ML Updates:
- ML Tensor Type: A new ML tensor type has been introduced to simplify the computational glue code for stitching models together.
- Performance Tools: Updated performance reports and new tools to optimize model execution.
- State Management: Key-value caches for efficient decoding of large language models with states.
For more details, you can check out the session Deploy machine learning and AI models on-device with Core ML (17:22).
-
Vision Framework:
- New Swift API: A new Swift API with Swift 6 support for Vision, including capabilities like text extraction, face detection, and body pose recognition.
- Hand Pose Detection: New hand pose detection and body pose requests.
For more details, you can check out the session Explore machine learning on Apple platforms (03:54).
-
Running Models on Device:
- Wide Array of Models: Support for running various models like Whisper, Stable Diffusion, and Mistral on Apple devices.
- Core ML Integration: Core ML segments models across CPU, GPU, and neural engine to maximize hardware utilization.
For more details, you can check out the session Explore machine learning on Apple platforms (10:26).
Relevant Sessions
- What’s new in Create ML
- Deploy machine learning and AI models on-device with Core ML
- Explore machine learning on Apple platforms
If you have any more specific questions or need further details, feel free to ask!
Create enhanced spatial computing experiences with ARKit
Learn how to create captivating immersive experiences with ARKit’s latest features. Explore ways to use room tracking and object tracking to further engage with your surroundings. We’ll also share how your app can react to changes in your environment’s lighting on this platform. Discover improvements in hand tracking and plane detection which can make your spatial experiences more intuitive.
Platforms State of the Union
Discover the newest advancements on Apple platforms.
What’s new in Create ML
Explore updates to Create ML, including interactive data source previews and a new template for building object tracking models for visionOS apps. We’ll also cover important framework improvements, including new time-series forecasting and classification APIs.
Deploy machine learning and AI models on-device with Core ML
Learn new ways to optimize speed and memory performance when you convert and run machine learning and AI models through Core ML. We’ll cover new options for model representations, performance insights, execution, and model stitching which can be used together to create compelling and private on-device experiences.
Explore machine learning on Apple platforms
Get started with an overview of machine learning frameworks on Apple platforms. Whether you’re implementing your first ML model, or an ML expert, we’ll offer guidance to help you select the right framework for your app’s needs.