HubLens › Trending › PaddlePaddle/PaddleCustomDevice
PaddlePaddle

PaddleCustomDevice

AIPaddlePaddleDeep LearningHardware AccelerationNPUGPU
View on GitHub
103
+340

// summary

PaddleCustomDevice is the custom hardware integration solution provided by the PaddlePaddle framework. This project aims to help developers integrate various third-party hardware backends into the PaddlePaddle ecosystem. By referring to the official integration solutions and sample code, users can achieve support for a variety of domestic and mainstream hardware platforms.

// use cases

01
Supports the integration of various hardware backends such as Ascend NPU, Cambricon MLU, and Intel GPU.
02
Provides a comprehensive design and development guide for custom hardware integration.
03
Assists developers in adapting new hardware through CustomCPU sample code.