TECHNIQUES FOR MACHINE LANGUAGE MODEL CREATION

    公开(公告)号:US20200380301A1

    公开(公告)日:2020-12-03

    申请号:US16670914

    申请日:2019-10-31

    Applicant: Apple Inc.

    Abstract: Embodiments of the present disclosure present devices, methods, and computer readable medium for techniques for creating machine learning models. Application developers can select a machine learning template from a plurality of templates appropriate for the type of data used in their application. Templates can include multiple templates for classification of images, text, sound, motion, and tabular data. A graphical user interface allows for intuitive selection of training data, validation data, and integration of the trained model into the application. The techniques further display a numerical score for both the training accuracy and validation accuracy using the test data. The application provides a live mode that allows for execution of the machine learning model on a mobile device to allow for testing the model from data from one or more of the sensors (i.e., camera or microphone) on the mobile device.

    CAPABILITY ATTRIBUTES BASED APPLICATION PACKAGING
    4.
    发明申请
    CAPABILITY ATTRIBUTES BASED APPLICATION PACKAGING 有权
    基于能力属性的应用程序包装

    公开(公告)号:US20160357536A1

    公开(公告)日:2016-12-08

    申请号:US14732610

    申请日:2015-06-05

    Applicant: Apple Inc.

    CPC classification number: G06F8/61 G06F9/54 H04L67/34

    Abstract: A method and an apparatus for an application thinning mechanism are described. The thinning mechanism can select a subset of components from a universal application to assemble an application variant to be distributed and installed to a specific type of devices. The universal application may include every component, such as asset, resource or executable, built/developed for targeted device attributes to install one common application to multiple devices. For example, the thinning mechanism can use a trait vector associated with a type of devices to iterate through the components and identify assets to be included or packaged into in each target device specific application or application variant.

    Abstract translation: 描述了一种应用稀疏机制的方法和装置。 稀疏机制可以从通用应用程序中选择一个组件的子集,以组装要分发和安装到特定类型设备的应用程序变体。 通用应用可以包括针对目标设备属性构建/开发的每个组件,例如资产,资源或可执行文件,以将一个通用应用安装到多个设备。 例如,稀疏机制可以使用与一类设备相关联的特征向量来遍历组件并且识别要包含或打包在每个目标设备特定应用或应用变体中的资产。

    TOUCH-BASED INTERACTIVE LEARNING ENVIRONMENT
    5.
    发明申请
    TOUCH-BASED INTERACTIVE LEARNING ENVIRONMENT 审中-公开
    基于触摸的交互式学习环境

    公开(公告)号:US20160357431A1

    公开(公告)日:2016-12-08

    申请号:US15173515

    申请日:2016-06-03

    Applicant: Apple Inc.

    Abstract: In an example method, a user interface is presented by a device on a touch sensitive surface of the device. The device receives a first touch input selecting a numeral presented in the user interface. Responsive to receiving the first touch input, a numeric keypad interface and a display element are presented. The device receives a second touch input selecting the display element. Responsive to receiving the second touch input, a rotary interface including the display element is presented. While the user is maintaining physical contact with the touch sensitive surface, the device receives a rotational touch gesture dragging the display element about the rotary interface. Responsive to receiving the rotational touch gesture, the numeral is modified.

    Abstract translation: 在示例性方法中,用户界面由设备的触敏表面上的设备呈现。 该设备接收选择在用户界面中呈现的数字的第一触摸输入。 响应于接收第一触摸输入,数字小键盘接口和显示元件。 设备接收选择显示元件的第二触摸输入。 响应于接收第二触摸输入,呈现包括显示元件的旋转界面。 当用户正在与触敏表面保持物理接触时,该设备接收围绕旋转界面拖动显示元件的旋转触摸手势。 响应于接收旋转触摸手势,数字被修改。

Patent Agency Ranking