This youtuber give a good layman overview about
1) AI,
2) ML,
3) DL,
4) Discriminative DL --> your Apple Silicion Bionic Chip and M1/2/3 have the 16-core Neurone Engine can do until here --> without using other data set, but only your own secure data set..

without sacrifice your personal data for the AI/ML learning, it is the best option for now.
Please do train your machine (iPhone/Macbook) on your photo when you free, it can make the discriminative DL smarter.

5) LLM, and regenarative AI







---

The Apple 16-core Neural Engine is a dedicated hardware found within the M1 chip, which is used in Macs¹. It greatly accelerates machine learning tasks, enabling up to 15x faster machine learning performance¹. The Neural Engine is also found in the A-Series Bionic microprocessors designed by the company, which equip devices such as the iPhone and iPad with native or localized machine learning capabilities¹.

Apple Neural Engine is a marketing name for a cluster of highly specialized compute cores optimized for the energy-efficient execution of deep neural networks on Apple devices⁷. It accelerates machine learning (ML) and artificial intelligence (AI) algorithms, offering tremendous speed, memory, and power advantages over the main CPU or GPU⁷.

The 16-core Neural Engine on the A15 Bionic chip on iPhone 13 Pro has a peak throughput of 15.8 teraflops⁴. When training ML models, developers benefit from accelerated training on GPUs with PyTorch and TensorFlow by leveraging the Metal Performance Shaders (MPS) back end⁴.

The latest M2 Pro and M2 Max SoCs take that capability much, much farther. The Neural Engine in these SoCs have 16 neural cores³. They can grind through up to 15.8 trillion operations every second³.

For example, the 16-core Neural Engine in the M2 Pro and M2 Max chips can handle 15.8 trillion operations per second⁸. As semiconductor processes become more advanced over time, manufacturers are able to add more transistors to chips during fabrication⁸.

Source: Conversation with Bing, 11/24/2023
(1) Apple unleashes M1 - Apple. https://www.apple.com/newsroom/2020/11/apple-unleashes-m1/.
(2) undefined. https://www.makeuseof.com/what-is-a-neural-...w-does-it-work/.
(3) Deploying Transformers on the Apple Neural Engine. https://machinelearning.apple.com/research/...ne-transformers.
(4) What Is the Apple Neural Engine and What Does It Do? - The Mac Observer. https://www.macobserver.com/tips/deep-dive/...-neural-engine/.
(5) undefined. https://appleinsider.com/articles/23/02/09/...n-new-interview.
(6) . https://bing.com/search?q=Apple+16-core+neu...+engine+working.
(7) Neural Engine | Apple Wiki | Fandom. https://apple.fandom.com/wiki/Neural_Engine.
(8) Neural Engine Hardware of Apple Explained - Profolus. https://www.profolus.com/topics/neural-engi...pple-explained/.

----




---

Create Machine Learning from Apple Framework: https://developer.apple.com/documentation/CreateML

Core ML framework: https://developer.apple.com/documentation/coreml

This post has been edited by xiaojohn: Nov 24 2023, 07:38 PM

Yes, worth to take a read and digest.

i also open another tered in /k for your reply/chit-chat...different topic and focus...

This post has been edited by xiaojohn: Nov 24 2023, 07:18 PM