How to use OpenVino to recognize Intel Agilex® 7 Device

[YawenLi]

How to let OpenVino to recoginze the device Intel Agilex® 7 Device ( Terasic* DE10-Agilex Development Board)?

I have installed OpenVino 2022.3.1 through archived files in /opt/intel/openvino_2022.3.1

And intel FPGA AI Suite in /opt/intel/fpga_ai_suite_2023.2.1

The device is recognized through "aocl diagnose"

Device Name: acl0

However, when I run a test python code:

from openvino.runtime import Core

# Initialize OpenVINO runtime

ie = Core()

available_devices = ie.available_devices

print(available_devices)

The result is

(openvino_env) sadmin@desktop-pc:~/dev/project-ASR$ python3 test.py
[E:] [BSL] found 0 ioexpander device
['CPU', 'GNA']

[JohnT_Intel]

Hi,

May I know how do you implement your Intel FPGA AI Suite? Have you program the bitstream generated from the AI Suite?

[YawenLi]

The document I follow about FPGA AI Suite is "Intel® FPGA AI Suite Quick Start Tutorial".  I have done steps 6.1-6.5

https://www.intel.com/content/www/us/en/docs/programmable/768970/2023-3/quick-start-tutorial.html

For the bitstream, the related command I executed are:

jtagdir=$COREDLA_WORK/runtime/build_Release/fpga_jtag_reprogram
bitsdir=$COREDLA_WORK/demo/bitstreams
$jtagdir/fpga_jtag_reprogram $bitsdir/AGX7_Performance.sof
curarch=$COREDLA_ROOT/example_architectures/AGX7_Performance.arch

And there is no error.

[JohnT_Intel]

Hi,

Are you able to run base on https://www.intel.com/content/www/us/en/docs/programmable/768970/2023-3/performing-inference-on-the-pcie-based.html?

[YawenLi]

Yes, I am. It seems run OK.

The screenshots are in the attachment.

[YawenLi]

Hi, can someone help me?

(openvino_env) yawen@yawen-desktop-pc:~/dev/project-ASR/src$ python3 speech_recognition_wav2vec_demo.py -m /home/yawen/dev/project-ASR/demo/models/public/wav2vec2-base/FP16/wav2vec2-base.xml -i /home/yawen/dev/project-ASR/demo/sample_audios/harvard_mono.wav -d 'HETERO:FPGA'
[ INFO ] OpenVINO Runtime
[ INFO ] build: 2023.2.0-13089-cfd42bd2cb0-HEAD
[ INFO ] Reading model /home/yawen/dev/project-ASR/demo/models/public/wav2vec2-base/FP16/wav2vec2-base.xml
Traceback (most recent call last):
File "speech_recognition_wav2vec_demo.py", line 142, in <module>
sys.exit(main() or 0)
File "speech_recognition_wav2vec_demo.py", line 132, in main
model = Wav2Vec(core, args.model, audio.shape, args.device, args.vocab, args.dynamic_shape)
File "speech_recognition_wav2vec_demo.py", line 76, in __init__
compiled_model = core.compile_model(model, device)
File "/opt/intel/openvino_2023.2.0/python/openvino/runtime/ie_api.py", line 543, in compile_model
super().compile_model(model, device_name, {} if config is None else config),
RuntimeError: Exception from src/inference/src/core.cpp:113:
[ GENERAL_ERROR ] Exception from src/plugins/hetero/src/compiled_model.cpp:34:
Standard exception from compilation library: Exception from src/inference/src/dev/core_impl.cpp:559:
Device with "FPGA" name is not registered in the OpenVINO Runtime

[YawenLi]

Hi, I found a script to query the devices, it seems it cannot find the device:

yawen@yawen-desktop-pc:~$ python3 /opt/intel/openvino_2023.2.0/samples/python/hello_query_device/hello_query_device.py
[ INFO ] Available devices:
[ INFO ] CPU :
[ INFO ] SUPPORTED_PROPERTIES:
[ INFO ] AVAILABLE_DEVICES:
[ INFO ] RANGE_FOR_ASYNC_INFER_REQUESTS: 1, 1, 1
[ INFO ] RANGE_FOR_STREAMS: 1, 16
[ INFO ] FULL_DEVICE_NAME: AMD Ryzen 7 5700G with Radeon Graphics
[ INFO ] OPTIMIZATION_CAPABILITIES: FP32, FP16, INT8, BIN, EXPORT_IMPORT
[ INFO ] NUM_STREAMS: 1
[ INFO ] AFFINITY: Affinity.CORE
[ INFO ] INFERENCE_NUM_THREADS: 0
[ INFO ] PERF_COUNT: False
[ INFO ] INFERENCE_PRECISION_HINT: <Type: 'float32'>
[ INFO ] PERFORMANCE_HINT: PerformanceMode.LATENCY
[ INFO ] EXECUTION_MODE_HINT: ExecutionMode.PERFORMANCE
[ INFO ] PERFORMANCE_HINT_NUM_REQUESTS: 0
[ INFO ] ENABLE_CPU_PINNING: True
[ INFO ] SCHEDULING_CORE_TYPE: SchedulingCoreType.ANY_CORE
[ INFO ] ENABLE_HYPER_THREADING: True
[ INFO ] DEVICE_ID:
[ INFO ] CPU_DENORMALS_OPTIMIZATION: False
[ INFO ] CPU_SPARSE_WEIGHTS_DECOMPRESSION_RATE: 1.0
[ INFO ]
[ INFO ] GNA :
[ INFO ] SUPPORTED_PROPERTIES:
[ INFO ] AVAILABLE_DEVICES: GNA_SW
[ INFO ] OPTIMAL_NUMBER_OF_INFER_REQUESTS: 1
[ INFO ] RANGE_FOR_ASYNC_INFER_REQUESTS: 1, 1, 1
[ INFO ] OPTIMIZATION_CAPABILITIES: INT16, INT8, EXPORT_IMPORT
[ INFO ] FULL_DEVICE_NAME: GNA_SW
[ INFO ] GNA_LIBRARY_FULL_VERSION: 3.5.0.2116
[ INFO ] GNA_DEVICE_MODE: GNA_SW_EXACT
[ INFO ] PERFORMANCE_HINT: PerformanceMode.LATENCY
[ INFO ] LOG_LEVEL: LOG_NONE
[ INFO ] EXECUTION_DEVICES: GNA
[ INFO ] GNA_SCALE_FACTOR_PER_INPUT: UNSUPPORTED TYPE
[ INFO ] GNA_FIRMWARE_MODEL_IMAGE:
[ INFO ] GNA_HW_EXECUTION_TARGET: UNSUPPORTED TYPE
[ INFO ] GNA_HW_COMPILE_TARGET: UNSUPPORTED TYPE
[ INFO ] GNA_PWL_DESIGN_ALGORITHM: UNSUPPORTED TYPE
[ INFO ] GNA_PWL_MAX_ERROR_PERCENT: 1.000000
[ INFO ] INFERENCE_PRECISION_HINT: <Type: 'undefined'>
[ INFO ] EXECUTION_MODE_HINT: ACCURACY
[ INFO ] PERFORMANCE_HINT_NUM_REQUESTS: 1

[JohnT_Intel]

Hi,

May I know how do you create the host code? Are you able share with me how write it?

[JohnT_Intel]

Hi,

May I know how do you create the host code? Are you able share with me how write it?

Anticipate your reply.