spb/tools-spboucher-ai

1{2  "slug": "create-ml",3  "name": "Create ML",4  "aliases": ["CreateML"],5  "vendor": { "name": "Apple Inc.", "slug": "apple", "website": "https://developer.apple.com/" },6  "homepage_url": "https://developer.apple.com/documentation/createml",7  "docs_url": "https://developer.apple.com/documentation/createml",8  "product_type": "application",9  "product_status": "active",10  "summary": "Apple's app and Swift framework for training image, text, sound, tabular and activity models on a Mac without writing training code, exporting directly to Core ML",11  "description": "Create ML is Apple's model-training tool for developers who want a Core ML model without building a training pipeline. It exists as a Mac application bundled with Xcode and as a Swift framework usable in playgrounds and on-device. Users pick a template, such as image, object detection, text classification, sound, tabular regression, activity or recommendation, point it at labeled data, and train with transfer learning on top of the models Apple ships in the operating system, which keeps training short and the resulting models small.\n\nThe app shows accuracy metrics, lets you test with new data and exports a `.mlmodel` ready to drop into an Xcode project. Because training runs on the Mac's own hardware, data never leaves the machine. The framework variant supports training on iOS and iPadOS for personalization scenarios.\n\nIt is tied to Apple's template set and platforms: there is no arbitrary architecture definition, no distributed training and no non-Apple export, so anything beyond the supported tasks is done in PyTorch or MLX and converted with coremltools.",12  "capabilities": ["Template-based training for images, text, sound, tabular data, activity and recommendations", "Transfer learning on Apple's built-in models", "Evaluation and live testing in the app", "Direct export to Core ML format", "Swift framework for programmatic and on-device training"],13  "workflows": ["Train an image classifier from labeled folders and add it to an app", "Build a text classifier for support tickets", "Personalize a model on-device with user data"],14  "advantages": ["No training code required", "Fast, local training with small outputs", "Free with Xcode"],15  "limitations": ["Only Apple's task templates", "Apple platforms only; no other export format", "Not suited to large or custom architectures"],16  "prerequisites": ["Mac with Xcode", "Labeled training data"],17  "audiences": ["App developers adding ML features", "Educators and students"],18  "alternatives": ["core-ml", "pytorch", "scikit-learn", "mlx"],19  "categories": [{ "slug": "apple-silicon-ml", "primary": true }, { "slug": "model-training" }],20  "tags": ["apple-silicon", "macos", "model-training", "no-code", "swift"],21  "use_cases": ["fine-tune-models", "build-native-apps"],22  "operating_systems": ["macos", "ios"],23  "execution_modes": ["local"],24  "interfaces": ["desktop", "sdk"],25  "deployment_models": ["desktop"],26  "integrations": ["Xcode", "Core ML", "Vision", "Natural Language"],27  "data_formats": [".mlmodel", "CSV / JSON tables", "Labeled image folders"],28  "languages": ["Swift"],29  "source_model": "proprietary",30  "pricing_model": ["free"],31  "self_hostable": false,32  "sources": [33    { "url": "https://developer.apple.com/documentation/createml", "type": "official_site", "title": "Create ML | Apple Developer Documentation", "supports": ["name", "summary", "capabilities", "operating_systems"], "checked_at": "2026-11-02" },34    { "url": "https://developer.apple.com/documentation/coreml", "type": "official_docs", "title": "Core ML | Apple Developer Documentation", "supports": ["integrations", "data_formats"], "checked_at": "2026-11-02" },35    { "url": "https://developer.apple.com/documentation/xcode", "type": "official_docs", "title": "Xcode | Apple Developer Documentation", "supports": ["prerequisites"], "checked_at": "2026-11-02" }36  ],37  "confidence": "high",38  "discovery_method": "editorial-research",39  "depth": "standard",40  "editorial_status": "reviewed",41  "visibility": "public",42  "featured": false,43  "author": "batch-C"44}45