#CoreML

1 month ago

Why can't you train an LLM on your iPhone?
It's not just the speed. It's the Memory Physics.

Ep 6 of Sandboxed is out: "Training vs. Inference."
We break down the wall between "Learning" and "Doing."

logicbridge.dev/sandboxed/6

#OnDeviceAI #CoreML #iOSDev #MachineLearning #AppleSilicon

1 0 0 0

2 months ago

A lot of “Core ML performance problems” aren’t performance problems.
They’re task selection problems.

🎧 Full episode: logicbridge.dev/sandboxed/4

#iOSDevelopment #CoreML #OnDeviceAI #VisionFramework #Sandboxed

0 0 0 0

2 months ago

Your Core ML model isn’t a black box.
It’s adjustable logic.

And if your accuracy is *suspiciously* high… it might be cheating.

Listen now: logicbridge.dev/sandboxed/3

#iOSDevelopment #CoreML #OnDeviceAI #MachineLearning #Sandboxed

0 0 0 0

‎MagicaLCore ‎MagicaLCore offers everything you need to work with machine learning on your iPad. From importing and organizing models to training and live testing, MagicaLCore lets you build and experiment with m...

2 months ago

🎧 99% accuracy? Your model might be cheating.

Episode 03 explains “learning” as: forward pass → loss → nudge weights.
Then why models fail in real apps: overfitting + data leakage.

What’s the sneakiest shortcut you’ve seen in data?

Listen now: logicbridge.dev/sandboxed/3

#iOSDev #CoreML

1 0 0 0

Jörg Klausewitz

@klausewitz.de

3 months ago

Apples Vision framework is insane.

Thank you #Apple.

#swift #swiftui #ai #coreml #ml

0 0 0 0

Sandor

@isandor.bsky.social

7 months ago

I made a #TensorFlow app for #CoreML embedding models, to power an offline, on-device FAQ chatbot (screenshots: iOS Budapest airport app). Added profanity filter to make the bot cheeky.

Next create a SPM with logic and UI. apps.apple.com/gb/app/bud-a...

0 0 0 0

DaMy

@damy.bsky.social

8 months ago

🚨 New app drop!
MagicaLCore for iPad lets you create image classifiers using your own dataset — no coding required.
• Build & test CoreML models on-device
• Export .mlmodel in seconds
• Free to try, with 3-day trial

🔗 apps.apple.com/app/id646644...

#indieDev #CoreML #AI #NoCode

8 0 0 0

ZooScan - Part 4: Using the Swift Vision Framework to Classify Animals In this fourth part of the ZooScan series, we integrate the Swift Vision framework to classify animals based on images captured by the app. We define a protocol for image classification models and imp...

8 months ago

Want to learn how to develop an image classification app with swift and CoreML? In the fourth post in the series, we integrate a Vision classification model and build the first working version of the ZooScan app.

thinkpractice.nl/post/zooscan...

#swift #swiftlang #iOSDev #ML #coreml #Buildinpublic

4 1 1 0

ZooScan – Part 3: Storing Our Scanned Animals and Finalizing the UI In this third part of the ZooScan series, we implement the ViewModel and ScannedAnimal model, and create the Main and Detail views. This sets the stage for the machine learning classification in futur...

9 months ago

Want to learn how to develop an image classification app with swift and CoreML?

In the third post in the series we finish a first version of the ZooScan UI and provide a basis for adding a ML classifier later.

thinkpractice.nl/post/zooscan...

#swift #swiftlang #iOSDev #Buildinpublic #ML #coreml

5 1 0 0

Lalit mukesh

@lalitt.bsky.social

9 months ago

Gonna need that ZooScan app, stat! My cat keeps getting misidentified as a dust bunny 😅 #CoreML #Swift

1 0 0 0

ZooScan - Part 2: Project Setup and the First Steps In this second part of the ZooScan series, we set up the project, create the basic UI, and implement the initial screens of the app. Join us as we start building this exciting animal classification ap...

9 months ago

Want to learn how to develop an image classification app with swift and CoreML?

In the second post of the series we are going to start implementing the ZooScan App's views:

thinkpractice.nl/post/zooscan...

#swift #swiftlang #iOSDev #Buildinpublic #ML #coreml

3 0 1 0