Open sandeepmohanadasan opened 3 years ago
@sandeepmohanadasan
what error you met?
1808 ops written to frozen_darknet_yolov4_model.pb.
It runs correctly . No error
It's just the first strp,the tensorflow program .
mo.py is OpenVINO program
So can I ignore the above-mentioned warnings as the .pb file generated?
@sandeepmohanadasan Yes! Just warnings,not errors
@TNTWEN Thanks for the clarification. I have proceeded with the remaining commands .Let me know if you can throw some light on below issue
- call C:\Program Files (x86)\Intel\openvino_2021\bin\setupvars.bat
- python convert_weights_pb.py --weights_file Model\custom-yolov4-detector_last.weights --output_graph model_yolov4.pb --class_names cfg\coco_chip.names --data_format NHWC --size 416
- python C:\Program Files (x86)\Intel\openvino_2021.1.110\deployment_tools\model_optimizer\mo.py --input_model model_yolov4.pb --transformations_config model_yolov4.json --reverse_input_channels --batch 1
below files generated
- model_yolov4.bin
- model_yolov4.mapping
- model_yolov4.xml
- model_yolov4.pb
But when I tried to validate my image, I was getting the below error
(env) C:\Users\<user>\conversion\OpenVINOModelConversion>python object_detection_demo_yolov4_async2021.py -m model_yolov4.xml -i modeltest01.jpg -d CPU
[ INFO ] Creating Inference Engine...
[ INFO ] Loading network
[ INFO ] Preparing inputs
MFX: Unsupported extension: modeltest01.jpg
[ INFO ] Loading model to the plugin
[ INFO ] Starting inference...
To close the application, press 'CTRL+C' here or switch to the output window and press ESC key
To switch between min_latency/user_specified modes, press TAB key in the output window
object_detection_demo_yolov4_async2021.py:233: DeprecationWarning: 'layers' property of IENetwork class is deprecated. For iteration over network please use get_ops()/get_ordered_ops() methods from nGraph Python API
out_blob = out_blob.buffer.reshape(net.layers[net.layers[layer_name].parents[0]].out_data[0].shape)
Traceback (most recent call last):
File "object_detection_demo_yolov4_async2021.py", line 496, in <module>
sys.exit(main() or 0)
File "object_detection_demo_yolov4_async2021.py", line 386, in main
args.keep_aspect_ratio)
File "object_detection_demo_yolov4_async2021.py", line 233, in get_objects
out_blob = out_blob.buffer.reshape(net.layers[net.layers[layer_name].parents[0]].out_data[0].shape)
ValueError: cannot reshape array of size 105456 into shape (1,33,52,52)
It is obvious that you didn't set correct classes in model_yolov4.json 105456 is (1,39,52,52) means classes =8 (1,33,52,52) means classes=5
check your classes file and corresponding parameters
Hi @TNTWEN , I am getting an error while converting YoloV4 weight to OPENVINO format. I have tried the 2021.1 and 2021.3 version. I can replicate the issue on both versions. It will be a great help if you can guide me on this.
Intel Driver details
Packages List