TE0950, Vitis 2023.2.1, AI 3.5 Applications for AMD DPUCZDX8G
Description
This application note evaluates performance of the AMD Vitis AI 3.0 classification examples accelerated by the AMD DPUCZDX8G HW IP implemented in the PL logic of the device without acceleration by the AI Engine-ML tiles.
The DPUCZDX8G is the Deep learning Processor Unit (DPU) designed by AMD for the Zynq UltraScale+ MPSoC devices.
The DPUCZDX8G can be also compiled from the RTL source code to PL part of the AMD Versal XCVE2302 device.
The DPUCZDX8G is a configurable computation engine optimized for convolutional neural networks.
It is a microcoded compute engine which has an efficient, optimized instruction set, and which can support inference of most convolutional neural networks.
The AMD DPUs support a set of highly optimized instructions, and supports most convolutional neural networks, such as VGG, ResNet, GoogLeNet, YOLO, SSD, MobileNet, FPN, and others.
Package Summary
| Title | TE0950, Vitis 2023.2.1, AI 3.5 Applications for AMD DPUCZDX8G |
| Filenames | TE0950_2023_2_1_dpuczdx8g_AI_3_5_v1_0.pdf (806 kB) |
| License | See application note for licensing conditions. |
| Size | PDF file: 826 133 Bytes |
| Required tools & platform |
AMD Vitis 2023.2.1 |
Result Category
| Project number | Project acronym | Year | RIV category | Comment |
| 9A25004 | NeAIxt | 2026 | Gfunk | Functional sample (demo) |

