公开(公告)号：US20230035451A1

公开(公告)日：2023-02-02

申请号：US17783247

申请日：2020-12-09

Applicant: Microsoft Technology Licensing, LLC

Inventor： Yanjie GAO ,  Haoxiang Lin ,  Yuci Liu ,  Mao Yang

IPC: G06N3/08

Abstract: According to implementations of the subject matter described herein, there is provided a solution for predicting the resource usage of the deep learning model. In this solution, information about a deep learning model is obtained, the information comprising first information for describing the deep learning model and second information about an operating environment of a job associated with the deep learning model. The static resource usage of the job is determined based on the first information and a strategy of the job during runtime in the operating environment is determined. Afterwards, resource usage of the job during runtime in the operating environment is predicted based on the strategy and the static resource usage. With this solution, the usage of various resources of the deep learning model, such as computation power consumption, memory consumption, execution time, and the like, under a specific runtime strategy can be accurately predicted.

公开(公告)号：US20220222049A1

公开(公告)日：2022-07-14

申请号：US17615080

申请日：2020-05-06

Applicant: Microsoft Technology Licensing, LLC

Inventor： Haoxiang Lin ,  Mao Yang ,  Shuguang Liu ,  Cheng Chen

IPC: G06F8/34 ,  G06N3/08 ,  G06F3/0486

Abstract: Implementations of the present disclosure relate to visual programming for deep learning. A computer-implemented method comprises presenting a visual representation of an artificial neural network, the visual representation comprising graphical elements representing layers of the artificial neural network; in response to receiving a drag-and-drop operation on the graphical elements, modifying an intermediate representation of the artificial neural network, wherein the intermediate representation is independent of a deep learning framework and the drag-and-drop operation is configured to modify connections between the graphical elements; and modifying, based on the intermediate representation of the artificial neural network, code of the artificial neural network for a target deep learning framework.

公开(公告)号：US20240370237A1

公开(公告)日：2024-11-07

申请号：US18774696

申请日：2024-07-16

Applicant: Microsoft Technology Licensing, LLC

Inventor： Haoxiang Lin ,  Mao Yang ,  Shuguang Liu ,  Cheng Chen

IPC: G06F8/34 ,  G06F3/0486 ,  G06N3/082

Abstract: Implementations of the present disclosure relate to visual programming for deep learning. A computer-implemented method comprises presenting a visual representation of an artificial neural network, the visual representation comprising graphical elements representing layers of the artificial neural network; in response to receiving a drag-and-drop operation on the graphical elements, modifying an intermediate representation of the artificial neural network, wherein the intermediate representation is independent of a deep learning framework and the drag-and-drop operation is configured to modify connections between the graphical elements; and modifying, based on the intermediate representation of the artificial neural network, code of the artificial neural network for a target deep learning framework.

公开(公告)号：US12079600B2

公开(公告)日：2024-09-03

申请号：US17615080

申请日：2020-05-06

Applicant: Microsoft Technology Licensing, LLC

Inventor： Haoxiang Lin ,  Mao Yang ,  Shuguang Liu ,  Cheng Chen

IPC: G06F8/34 ,  G06F3/0486 ,  G06N3/082

CPC classification number: G06F8/34 ,  G06F3/0486 ,  G06N3/082

Abstract: Implementations of the present disclosure relate to visual programming for deep learning. A computer-implemented method comprises presenting a visual representation of an artificial neural network, the visual representation comprising graphical elements representing layers of the artificial neural network; in response to receiving a drag-and-drop operation on the graphical elements, modifying an intermediate representation of the artificial neural network, wherein the intermediate representation is independent of a deep learning framework and the drag-and-drop operation is configured to modify connections between the graphical elements; and modifying, based on the intermediate representation of the artificial neural network, code of the artificial neural network for a target deep learning framework.