[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-9655":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":24,"defaultBranch":25,"hasWiki":24,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":16,"starSnapshotCount":16,"syncStatus":18,"lastSyncTime":33,"discoverSource":34},9655,"turicreate","apple\u002Fturicreate","apple","Turi Create simplifies the development of custom machine learning models.","",null,"C++",11169,1124,11,501,0,1,2,3,4,44.15,"BSD 3-Clause \"New\" or \"Revised\" License",true,false,"main",[27,28,29],"deep-learning","machine-learning","python","2026-06-12 02:02:10","Quick Links: [Installation](#supported-platforms) | [Documentation](#documentation)\n\n[![Build Status](https:\u002F\u002Ftravis-ci.com\u002Fapple\u002Fturicreate.svg?branch=master)](#)\n[![PyPI Release](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fturicreate.svg)](#)\n[![Python Versions](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fpyversions\u002Fturicreate.svg)](#)\n\n[\u003Cimg align=\"right\" src=\"https:\u002F\u002Fdocs-assets.developer.apple.com\u002Fturicreate\u002Fturi-dog.svg\" alt=\"Turi Create\" width=\"100\">](#)\n\n# Turi Create \n\nTuri Create simplifies the development of custom machine learning models. You\ndon't have to be a machine learning expert to add recommendations, object\ndetection, image classification, image similarity or activity classification to\nyour app.\n\n* **Easy-to-use:** Focus on tasks instead of algorithms\n* **Visual:** Built-in, streaming visualizations to explore your data\n* **Flexible:** Supports text, images, audio, video and sensor data\n* **Fast and Scalable:** Work with large datasets on a single machine\n* **Ready To Deploy:** Export models to Core ML for use in iOS, macOS, watchOS, and tvOS apps\n\nWith Turi Create, you can accomplish many common ML tasks:\n\n| ML Task                 | Description                      |\n|:------------------------:|:--------------------------------:|\n| [Recommender](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Frecommender\u002F)             | Personalize choices for users    |\n| [Image Classification](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fimage_classifier\u002F)    | Label images                     |\n| [Drawing Classification](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fdrawing_classifier)  | Recognize Pencil\u002FTouch Drawings and Gestures                     |\n| [Sound Classification](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fsound_classifier)  | Classify sounds                     |\n| [Object Detection](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fobject_detection\u002F)        | Recognize objects within images  |\n| [One Shot Object Detection](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fone_shot_object_detection\u002F)    | Recognize 2D objects within images using a single example  |\n| [Style Transfer](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fstyle_transfer\u002F)        | Stylize images |\n| [Activity Classification](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Factivity_classifier\u002F) | Detect an activity using sensors |\n| [Image Similarity](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fimage_similarity\u002F)        | Find similar images              |\n| [Classifiers](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fsupervised-learning\u002Fclassifier.html)             | Predict a label           |\n| [Regression](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fsupervised-learning\u002Fregression.html)              | Predict numeric values           |\n| [Clustering](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Fclustering\u002F)              | Group similar datapoints together|\n| [Text Classifier](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide\u002Ftext_classifier\u002F)         | Analyze sentiment of messages    |\n\n\nExample: Image classifier with a few lines of code\n--------------------------------------------------\n\nIf you want your app to recognize specific objects in images, you can build your own model with just a few lines of code:\n\n```python\nimport turicreate as tc\n\n# Load data \ndata = tc.SFrame('photoLabel.sframe')\n\n# Create a model\nmodel = tc.image_classifier.create(data, target='photoLabel')\n\n# Make predictions\npredictions = model.predict(data)\n\n# Export to Core ML\nmodel.export_coreml('MyClassifier.mlmodel')\n```\n \nIt's easy to use the resulting model in an [iOS application](https:\u002F\u002Fdeveloper.apple.com\u002Fdocumentation\u002Fvision\u002Fclassifying_images_with_vision_and_core_ml):\n\n\u003Cp align=\"center\">\u003Cimg src=\"https:\u002F\u002Fdocs-assets.developer.apple.com\u002Fpublished\u002Fa2c37bce1f\u002F689f61a6-1087-4112-99d9-bbfb326e3138.png\" alt=\"Turi Create\" width=\"600\">\u003C\u002Fp>\n\nSupported Platforms\n-------------------\n\nTuri Create supports:\n\n* macOS 10.12+\n* Linux (with glibc 2.10+)\n* Windows 10 (via WSL)\n\nSystem Requirements\n-------------------\n\nTuri Create requires:\n\n* Python 2.7, 3.5, 3.6, 3.7, 3.8\n* x86\\_64 architecture\n* At least 4 GB of RAM\n\nInstallation\n------------\n\nFor detailed instructions for different varieties of Linux see [LINUX\\_INSTALL.md](LINUX_INSTALL.md).\nFor common installation issues see [INSTALL\\_ISSUES.md](INSTALL_ISSUES.md).\n\nWe recommend using virtualenv to use, install, or build Turi Create. \n\n```shell\npip install virtualenv\n```\n\nThe method for installing *Turi Create* follows the\n[standard python package installation steps](https:\u002F\u002Fpackaging.python.org\u002Finstalling\u002F).\nTo create and activate a Python virtual environment called `venv` follow these steps:\n\n```shell\n# Create a Python virtual environment\ncd ~\nvirtualenv venv\n\n# Activate your virtual environment\nsource ~\u002Fvenv\u002Fbin\u002Factivate\n```\nAlternatively, if you are using [Anaconda](https:\u002F\u002Fwww.anaconda.com\u002Fwhat-is-anaconda\u002F), you may use its virtual environment:\n```shell\nconda create -n virtual_environment_name anaconda\nconda activate virtual_environment_name\n```\n\nTo install `Turi Create` within your virtual environment:\n```shell\n(venv) pip install -U turicreate\n```\n\nDocumentation\n-------------\n\nThe package [User Guide](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fuserguide) and [API Docs](https:\u002F\u002Fapple.github.io\u002Fturicreate\u002Fdocs\u002Fapi) contain\nmore details on how to use Turi Create.\n\nGPU Support\n-----------\n\nTuri Create **does not require a GPU**, but certain models can be accelerated 9-13x by utilizing a GPU.\n\n| Linux                     | macOS 10.13+         | macOS 10.14+ discrete GPUs, macOS 10.15+ integrated GPUs |\n| :-------------------------|:---------------------|:---------------------------------------------------------|\n| Activity Classification   | Image Classification | Activity Classification                                  |\n| Drawing Classification    | Image Similarity     | Object Detection                                         |\n| Image Classification      | Sound Classification | One Shot Object Detection                                |\n| Image Similarity          |                      | Style Transfer                                           |\n| Object Detection          |                      |                                                          |\n| One Shot Object Detection |                      |                                                          |\n| Sound Classification      |                      |                                                          |\n| Style Transfer            |                      |                                                          |\n\nmacOS GPU support is automatic. For Linux GPU support, see [LinuxGPU.md](LinuxGPU.md).\n\nBuilding From Source\n---------------------\n\nIf you want to build Turi Create from source, see [BUILD.md](BUILD.md).\n\nContributing\n------------\n\nPrior to contributing, please review [CONTRIBUTING.md](CONTRIBUTING.md) and do\nnot provide any contributions unless you agree with the terms and conditions\nset forth in [CONTRIBUTING.md](CONTRIBUTING.md).\n\nWe want the Turi Create community to be as welcoming and inclusive as possible, and have adopted a [Code of Conduct](CODE_OF_CONDUCT.md) that we expect all community members, including contributors, to read and observe.\n","Turi Create 是一个简化自定义机器学习模型开发的工具。它提供了一系列易于使用的功能，如推荐系统、图像分类、对象检测等，使开发者无需深入了解机器学习算法即可将这些功能集成到自己的应用中。该项目支持多种数据类型（文本、图片、音频、视频和传感器数据），内置了可视化工具帮助用户探索数据，并且能够处理大规模数据集。此外，Turi Create 还支持将训练好的模型导出为 Core ML 格式，方便在 iOS、macOS 等苹果平台上部署使用。适用于希望快速实现基础机器学习功能的应用场景。","2026-06-11 03:24:00","top_topic"]