UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
655 Views
Registered: ‎02-10-2014

Running precompiled examples

Hi Team,

I try to run precompiled examples from ML suite but faced with error: ERROR: Failed to create compute program from binary -42

andrey@L-RUMOSZ2NB10010K:~$ conda activate ml-suite
(ml-suite) andrey@L-RUMOSZ2NB10010K:~$ cd XILINX/ml-suite/examples/classification/
(ml-suite) andrey@L-RUMOSZ2NB10010K:~/XILINX/ml-suite/examples/classification$
(ml-suite) andrey@L-RUMOSZ2NB10010K:~/XILINX/ml-suite/examples/classification$ ./run.sh -p alveo-u200 -t test_classify -k v3 -b 8 -m resnet50
XILINX_XRT : /opt/xilinx/xrt
PATH : /opt/xilinx/xrt/bin:/home/andrey/anaconda2/envs/ml-suite/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin
LD_LIBRARY_PATH : /opt/xilinx/xrt/lib:/home/andrey/XILINX/ml-suite/overlaybins/alveo-u200/runtime/lib/x86_64/:/home/andrey/XILINX/ml-suite/xfdnn/rt/xdnn_cpp/build/lib:/home/andrey/XILINX/ml-suite/xfdnn/rt/lib:/home/andrey/XILINX/ml-suite/ext/boost/lib:/home/andrey/XILINX/ml-suite/ext/zmq/libs:/home/andrey/XILINX/ml-suite/examples/classification
20182
make: Entering directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms'
cd ./nms_20180209 && make
make[1]: Entering directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms/nms_20180209'
make[1]: Nothing to be done for 'all'.
make[1]: Leaving directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms/nms_20180209'
make: Leaving directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms'

Running:
Test: test_classify
Model: resnet50
Fpgaoutsz: 2048
Platform: alveo-u200
Xclbin: overlay_2.xclbin
Kernel Config: v3
Precision: 8
Accelerator: 0

[XBLAS] # kernels: 1
CL_PLATFORM_VENDOR Xilinx
CL_PLATFORM_NAME Xilinx
CL_DEVICE_0: 0x22796c0
CL_DEVICES_FOUND 1, using 0
loading /home/andrey/XILINX/ml-suite/overlaybins/alveo-u200/overlay_2.xclbin
ERROR: Failed to create compute program from binary -42

 

System details:

Card ALVEO U200

Server ASUS ESC4000

root@L-RUMOSZ2NB10010K:/home/andrey# xbutil flash scan
Card [0]
Card BDF: 0000:02:00.1
Card type: u200
Flash type: SPI
DSA running on FPGA:
xilinx_u200_xdma_201820_1,[TS=0x000000005b891ee3],[BMC=3.1]
DSA package installed in system:
xilinx_u200_xdma_201820_1,[TS=0x000000005b891ee3],[BMC=1.8]

root@L-RUMOSZ2NB10010K:/home/andrey# ^C

DISTRIB_DESCRIPTION="Ubuntu 16.04.4 LTS"

 

0 Kudos
2 Replies
Highlighted
Moderator
Moderator
615 Views
Registered: ‎04-12-2017

Re: Running precompiled examples

Hello @andrey.kochetov

from your log it looks like  you have installed old version of XRT and DSA shell on board.

I will recommend you to start process from scratch.

Please refer UG1301 for installing latest version of XRT and deployment shell which is 2018.3 and then refer ML overlays from following GITHUB

https://github.com/Xilinx/ml-suite/tree/master/overlaybins

I have personally ran this flow Ubuntu and 16.04 and it works successfully.

Thank you

-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
0 Kudos
582 Views
Registered: ‎02-10-2014

Re: Running precompiled examples

Thank you!

I upgraded to newest version of xrt but anyway I see another error( look below). Please advice how to check that python environmnet was installed right?

(ml-suite) andrey@L-RUMOSZ2NB10010K:~/XILINX/ml-suite/examples/classification$ ./run.sh -p alveo-u200 -t test_classify -k v3 -b 8 -m resnet50
XILINX_XRT : /opt/xilinx/xrt
PATH : /opt/xilinx/xrt/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/envs/ml-suite/bin:/home/andrey/anaconda2/bin:/home/andrey/anaconda2/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin
LD_LIBRARY_PATH : /opt/xilinx/xrt/lib:/home/andrey/XILINX/ml-suite/overlaybins/alveo-u200/runtime/lib/x86_64/:/home/andrey/XILINX/ml-suite/xfdnn/rt/xdnn_cpp/build/lib:/home/andrey/XILINX/ml-suite/xfdnn/rt/lib:/home/andrey/XILINX/ml-suite/ext/boost/lib:/home/andrey/XILINX/ml-suite/ext/zmq/libs:/home/andrey/XILINX/ml-suite/examples/classification
20182
make: Entering directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms'
cd ./nms_20180209 && make
make[1]: Entering directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms/nms_20180209'
make[1]: Nothing to be done for 'all'.
make[1]: Leaving directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms/nms_20180209'
make: Leaving directory '/home/andrey/XILINX/ml-suite/apps/yolo/nms'

Running:
Test: test_classify
Model: resnet50
Fpgaoutsz: 2048
Platform: alveo-u200
Xclbin: overlay_2.xclbin
Kernel Config: v3
Precision: 8
Accelerator: 0

Traceback (most recent call last):
File "test_classify.py", line 12, in <module>
import xdnn, xdnn_io
File "/home/andrey/XILINX/ml-suite/xfdnn/rt/xdnn_io.py", line 14, in <module>
import cv2
ImportError: No module named cv2
(ml-suite) andrey@L-RUMOSZ2NB10010K:~/XILINX/ml-suite/examples/classification$

0 Kudos