公开(公告)号：US20150381797A1

公开(公告)日：2015-12-31

申请号：US14843750

申请日：2015-09-02

Applicant: Huawei Technologies Co., Ltd.

Inventor： Yong GAO ,  Hanbin HU

IPC: H04M1/725 ,  H04M1/02

CPC classification number: H04M1/72575 ,  G06F3/03547 ,  G06F3/0362 ,  G06F3/041 ,  H04M1/0254 ,  H04M1/72522

Abstract: Embodiments of the present invention disclose a touch bar and a mobile terminal apparatus. An interface module of the touch bar is connected to a touch panel; the interface module is connected in a matching manner to an interface on the mobile terminal; the touch panel is configured to sense a touch operation, generate a corresponding sensing signal, and transmit it to a processor in the mobile terminal by using the interface module, so that the processor controls the mobile terminal according to the sensing signal and content currently displayed on a display interface of the mobile terminal. Functions such as flicking a browser page up and down, turning pages of an ebook, and adjusting volume are implemented effectively; the operability is enhanced, and the user experience is prevented from being affected because the screen is blocked due to a tap operation or flick operation, improving the user experience effectively.

Abstract translation: 本发明的实施例公开了触摸条和移动终端装置。 触摸板的接口模块连接到触摸面板; 接口模块以匹配的方式连接到移动终端上的接口; 触摸面板被配置为感测触摸操作，产生相应的感测信号，并且通过使用接口模块将其发送到移动终端中的处理器，使得处理器根据当前显示的感测信号和内容来控制移动终端 在移动终端的显示界面上。 上下翻页浏览器页面，翻页电子书和调整音量等功能得到有效实施; 提高了可操作性，并且防止了用户体验受到影响，因为屏幕被点击操作或轻拂操作阻止，从而有效地改善了用户体验。

公开(公告)号：US20240020541A1

公开(公告)日：2024-01-18

申请号：US18476830

申请日：2023-09-28

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Wei GUO ,  Huifeng GUO ,  Yong GAO ,  Ruiming TANG ,  Wenzhi LIU ,  Xiuqiang HE

IPC: G06N3/08

CPC classification number: G06N3/08

Abstract: This application describes a model training method, applied to the field of artificial intelligence. The method includes a computing core of a first processor obtains an embedding used for model training, and writes an updated embedding to a first memory of the first processor instead of transferring the updated embedding to a second processor after model training is completed. In this application, after updating an embedding, the first processor saves the updated embedding to the first memory of the first processor. Without needing to wait for the second processor to complete a process of transferring a second target embedding to a GPU, the first processor may directly obtain the updated embedding and perform model training of a next round based on the updated embedding, provided that the first processor may obtain a latest updated embedding.