- 专利标题: Accessing data in multi-dimensional tensors using adders
-
申请号: US15903991申请日: 2018-02-23
-
公开(公告)号: US10534607B2公开(公告)日: 2020-01-14
- 发明人: Olivier Temam , Harshit Khaitan , Ravi Narayanaswami , Dong Hyuk Woo
- 申请人: Google LLC
- 申请人地址: US CA Mountain View
- 专利权人: Google LLC
- 当前专利权人: Google LLC
- 当前专利权人地址: US CA Mountain View
- 代理机构: Fish & Richardson P.C.
- 主分类号: G06F9/302
- IPC分类号: G06F9/302 ; G06F9/355 ; G06F17/16 ; G06N3/00 ; G06F9/30 ; G06F9/34 ; G06F9/32 ; G06N3/04 ; G06N3/063
摘要:
Methods, systems, and apparatus, including an apparatus for accessing a N-dimensional tensor, the apparatus including, for each dimension of the N-dimensional tensor, a partial address offset value element that stores a partial address offset value for the dimension based at least on an initial value for the dimension, a step value for the dimension, and a number of iterations of a loop for the dimension. The apparatus includes a hardware adder and a processor. The processor obtains an instruction to access a particular element of the N-dimensional tensor. The N-dimensional tensor has multiple elements arranged across each of the N dimensions, where N is an integer that is equal to or greater than one. The processor determines, using the partial address offset value elements and the hardware adder, an address of the particular element and outputs data indicating the determined address for accessing the particular element of the N-dimensional tensor.
公开/授权文献
- US20180341479A1 ACCESSING DATA IN MULTI-DIMENSIONAL TENSORS USING ADDERS 公开/授权日:2018-11-29
信息查询