-
公开(公告)号:US20240193721A1
公开(公告)日:2024-06-13
申请号:US18530371
申请日:2023-12-06
发明人: Haoran LI , Fei SUN , Yuan GAO , Ruiguang ZHONG
IPC分类号: G06T1/20
CPC分类号: G06T1/20
摘要: This application describes an accelerator, a computer system, and a method for adaptive graph-to-stream scheduling in processors. An example method may include receiving a computation graph for a GPU, the computation graph comprising (1) a plurality of nodes representing a plurality of kernels for the GPU to execute and (2) a plurality of edges representing execution dependencies among the plurality of kernels; performing one or more wave partitions on the computation graph to determine a plurality of waves of kernels; obtaining a kernel resource table comprising resource usage of each kernel; mapping the plurality of kernels into a plurality of streams based on the plurality of waves and the kernel resource table; and executing the plurality of streams on the GPU, wherein kernels mapped in a same stream are executed sequentially by the GPU, and kernels mapped to different streams are concurrently executable by the GPU.
-
公开(公告)号:US20230402181A1
公开(公告)日:2023-12-14
申请号:US18048595
申请日:2022-10-21
CPC分类号: G16H50/20 , G06T7/0012 , G06T2207/30088
摘要: A method of skin detection for an object includes: acquiring an image of skin covering an outer surface of an object to be detected; evaluating the image of the skin to determine a skin lesion region of the object to be detected, in which the skin lesion region is an image region with a skin lesion feature in the image of the skin; determining a skin lesion attribute of the object to be detected based on the skin lesion feature of the skin lesion region and an object type of the object to be detected, in which the skin lesion attribute is used for describing a skin lesion generated on the object to be detected; and matching lesion data recorded in a lesion database based on the skin lesion attribute to determine a pathological result of the object to be detected.
-
公开(公告)号:US20230306257A1
公开(公告)日:2023-09-28
申请号:US17866194
申请日:2022-07-15
发明人: Fei SUN , Minghai QIN , Haoran LI , Guocai ZHU , Yuan GAO , Guyue HUANG , Yawen ZHANG
摘要: Neural network (NN) model training techniques can include computing activations in a forward pass using a sparse weight matrix that is transpose invariant. The neural network (NN) model training techniques can further include computing activation gradients and weight gradients in a backward pass using the sparse weight matrix.
-
-