知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

14 records (took 0.031 seconds)

    Data visualization machine learning model performance
    5.
    发明授权
    Data visualization machine learning model performance 有权

    公开(公告)号：US11580444B2

    公开(公告)日：2023-02-14

    申请号：US16530855

    申请日：2019-08-02

    Applicant: Apple Inc.

    Inventor： Aaron B. Franklin ,  Kanit Wongsuphasawat ,  Naga Rama Abhishek Pratapa ,  Srikrishna Sridhar ,  Zachary A. Nation

    IPC: G06F16/00 ,  G06N20/00 ,  G06F16/904 ,  G06F17/16

    Abstract: The subject technology receives information associated with a machine learning model. The subject technology determines a set of metrics based at least in part on the information associated with the machine learning model, where the set of metrics corresponds to respective indicators of performance of the machine learning model based on input data from a data set, the set of metrics further including a number of errors produced by the machine learning model when applied to the input data from the data set. Further, the subject technology displays a user interface based at least in part on the set of metrics, where the user interface includes a set of graphical elements, and the set of graphical elements further includes representations of the set of metrics, and representations of the input data from the data set utilized by the machine learning model.

    Providing a compact representation of tree structures
    7.
    发明授权
    Providing a compact representation of tree structures 有权

    公开(公告)号：US11216431B2

    公开(公告)日：2022-01-04

    申请号：US15851668

    申请日：2017-12-21

    Applicant: Apple Inc.

    Inventor： Hoyt A. Koepke ,  Srikrishna Sridhar

    IPC: G06F16/22 ,  G06F16/901 ,  G06N20/00

    Abstract: The subject technology provides for generating a set of nodes representing a tree structure, each node comprising a feature index, a flag field indicating branch directions, an execution index storing locations related to the branch directions, and a feature value for comparing with the value stored in the input feature vector. The subject technology generates evaluation data, the evaluation data comprising a first array containing index values, and a second array containing evaluation values respectively corresponding to the index values, the evaluation data representing values of leaf nodes from the set of nodes. Further, the subject technology stores the set of nodes and the evaluation data as a contiguous block of data, where the set of nodes includes a first node and a second node, the second node corresponding to a likely execution path from the first node being physically stored adjacent to the first node.

    Burst image fusion and denoising using end-to-end deep neural networks
    9.
    发明授权
    Burst image fusion and denoising using end-to-end deep neural networks 有权

    公开(公告)号：US11842460B1

    公开(公告)日：2023-12-12

    申请号：US17351820

    申请日：2021-06-18

    Applicant: Apple Inc.

    Inventor： Chen Chen ,  Gustav M. Larsson ,  Jiangkun Liu ,  Srikrishna Sridhar ,  Abdelrahman Abdelhamed

    IPC: G06T5/00 ,  G06T5/50

    CPC classification number: G06T5/002 ,  G06T5/50 ,  G06T2200/24 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20182 ,  G06T2207/20221

    Abstract: Electronic devices, methods, and non-transitory program storage devices for leveraging machine learning to perform improved image fusion and/or noise reduction are disclosed. When a capture request is received, a neural network may be used to perform fusion and denoising operations on a first set of captured input images. According to some embodiments, the neural network's architecture comprises: a first plurality of network layers configured to compute optical flow information between the first set of input images; a second plurality of network layers configured to perform, at least in part, the fusion and denoising operations on the first set of input images; and a third plurality of skip connections between layers of the second plurality of network layers, wherein parameters for each skip connection of the third plurality of skip connections are warped based on at least part of the optical flow information computed by the first plurality of network layers.

Patent Agency Ranking