Xilinx DPU IP Product Manual page 41

Dpu for convolutional neural network v1.2
Table of Contents

Advertisement

Demo Execution
To run the demo:
1. After generating the BOOT.BIN file, copy BOOT.BIN and image.ub (which is in
to the SD card.
2. Copy the common directory in $TRD_HOME/images to the SD card.
3. Copy the pre-built resnet50 in $TRD_HOME/images/resnet50 or the newly generated resnet50 in
$TRD_HOME/apu/apps/resnet50/build to the resnet50 directory on the SD card (for example,
/home/resnet50). If the directory does not exist, create a new directory.
4. Insert the SD card into the ZCU102 and boot up the board. After the Linux boot, run as follows:
% cd /home/resnet50/
% ./resnet50
The screenshot is shown in Figure 32.
The input images name is displayed in each line beginning with "Load image", followed by the expected
result of the input image. The predicted results of DPU is below, and the top-5 prediction probability of
image classification are printed. If the Top-0 predict results are the same as the expected result, then
the DPU is working properly.
DPU IP Product Guide
PG338 (v1.2) March 26, 2019
Chapter 6: Example Design
www.xilinx.com
folder)
image/linux
41
Send Feedback

Hide quick links:

Advertisement

Table of Contents
loading
Need help?

Need help?

Do you have a question about the DPU IP and is the answer not in the manual?

Questions and answers

This manual is also suitable for:

B512B1024B1152B1600B800B3136 ... Show all

Table of Contents