公开(公告)号：WO2020242686A1

公开(公告)日：2020-12-03

申请号：PCT/US2020/029937

申请日：2020-04-24

Applicant: APPLE INC.

Inventor： PAEK, Timothy S. ,  ROSSI, Francesco ,  DHANANI, Jamil ,  AVERY, Keith P. ,  JEONG, Minwoo ,  SHI, Xiaojin ,  KAUR, Harveen ,  WESTING, Brandt M.

IPC: G06N3/063 ,  G06N3/04 ,  G06N3/08

Abstract: The subject technology receives a neural network model in a model format, the model format including information for a set of layers of the neural network model, each layer of the set of layers including a set of respective operations. The subject technology generates neural network (NN) code from the neural network model, the NN code being in a programming language distinct from the model format, and the NN code comprising a respective memory allocation for each respective layer of the set of layers of the neural network model, where the generating comprises determining the respective memory allocation for each respective layer based at least in part on a resource constraint of a target device. The subject technology compiles the NN code into a binary format. The subject technology generates a package for deploying the compiled NN code on the target device.

公开(公告)号：WO2020236904A1

公开(公告)日：2020-11-26

申请号：PCT/US2020/033750

申请日：2020-05-20

Applicant: APPLE INC.

Inventor： ROSSI, Francesco ,  JAGADEESH, Vignesh ,  SHARMA, Vinay ,  ZULIANI, Marco ,  SHI, Xiaojin ,  POULAIN, Benjamin

IPC: G06N3/10 ,  G06N3/04

Abstract: A method includes receiving input data at a trained machine learning model that includes a common part and task-specific parts, receiving an execution instruction that identifies one or more processing tasks to be performed, processing the input data using the common part of the trained machine learning model to generate intermediate data, and processing the intermediate data using one or more of the task-specific parts of the trained machine learning model based on the execution instruction to generate one or more outputs.

公开(公告)号：WO2018222299A1

公开(公告)日：2018-12-06

申请号：PCT/US2018/029201

申请日：2018-04-24

Applicant: APPLE INC.

Inventor： ROSSI, Francesco ,  KAPOOR, Gaurav ,  SIRACUSA, Michael R. ,  MARCH, William B.

IPC: G06F9/50 ,  G06N3/063

Abstract: The subject technology provides for dynamic task allocation for neural network models. The subject technology determines an operation performed at a node of a neural network model. The subject technology assigns an annotation to indicate whether the operation is better performed on a CPU or a GPU based at least in part on hardware capabilities of a target platform. The subject technology determines whether the neural network model includes a second layer. The subject technology, in response to determining that the neural network model includes a second layer, for each node of the second layer of the neural network model, determines a second operation performed at the node. Further the subject technology assigns a second annotation to indicate whether the second operation is better performed on the CPU or the GPU based at least in part on the hardware capabilities of the target platform.

公开(公告)号：EP4283526A2

公开(公告)日：2023-11-29

申请号：EP23202256.6

申请日：2018-04-24

Applicant: Apple Inc.

Inventor： ROSSI, Francesco ,  KAPOOR, Gaurav ,  SIRACUSA, Michael R. ,  MARCH, William B.

IPC: G06N3/063

Abstract: The subject technology provides for dynamic task allocation for neural network models. The subject technology determines an operation performed at a node of a neural network model. The subject technology assigns an annotation to indicate whether the operation is better performed on a CPU or a GPU based at least in part on hardware capabilities of a target platform. The subject technology determines whether the neural network model includes a second layer. The subject technology, in response to determining that the neural network model includes a second layer, for each node of the second layer of the neural network model, determines a second operation performed at the node. Further the subject technology assigns a second annotation to indicate whether the second operation is better performed on the CPU or the GPU based at least in part on the hardware capabilities of the target platform.

公开(公告)号：EP4283526A3

公开(公告)日：2024-01-17

申请号：EP23202256.6

申请日：2018-04-24

Applicant: Apple Inc.

Inventor： ROSSI, Francesco ,  KAPOOR, Gaurav ,  SIRACUSA, Michael R. ,  MARCH, William B.

IPC: G06F9/50 ,  G06N3/063 ,  G06F8/41

Abstract: The subject technology provides for dynamic task allocation for neural network models. The subject technology determines an operation performed at a node of a neural network model. The subject technology assigns an annotation to indicate whether the operation is better performed on a CPU or a GPU based at least in part on hardware capabilities of a target platform. The subject technology determines whether the neural network model includes a second layer. The subject technology, in response to determining that the neural network model includes a second layer, for each node of the second layer of the neural network model, determines a second operation performed at the node. Further the subject technology assigns a second annotation to indicate whether the second operation is better performed on the CPU or the GPU based at least in part on the hardware capabilities of the target platform.

公开(公告)号：EP3977362A1

公开(公告)日：2022-04-06

申请号：EP20728265.8

申请日：2020-04-24

Applicant: Apple Inc.

Inventor： PAEK, Timothy S. ,  ROSSI, Francesco ,  DHANANI, Jamil ,  AVERY, Keith P. ,  JEONG, Minwoo ,  SHI, Xiaojin ,  KAUR, Harveen ,  WESTING, Brandt M.

IPC: G06N3/063 ,  G06N3/04 ,  G06N3/08

公开(公告)号：EP3631633A1

公开(公告)日：2020-04-08

申请号：EP18723333.3

申请日：2018-04-24

Applicant: Apple Inc.

Inventor： ROSSI, Francesco ,  KAPOOR, Gaurav ,  SIRACUSA, Michael R. ,  MARCH, William B.

IPC: G06F9/50 ,  G06N3/063