Could not perform "operation times" after conversion from sdk

Hi,
After a rigorous efforts [here], now I am able to convert the yolov3 model and can run it using produced files on khadas. But still could not perform " each operation times usage" according to the documentation page. It gives the following error. Can you please guide me!
@Frank @numbqq

khadas@Khadas-teco:~/sajjad/eachoptime2/op_test_yolov3$ ./build_vx.sh
  COMPILE /home/khadas/sajjad/eachoptime2/op_test_yolov3/vnn_pre_process.c
  COMPILE /home/khadas/sajjad/eachoptime2/op_test_yolov3/vnn_yolov3.c
vnn_yolov3.c: In function ‘vnn_CreateYolov3’:
vnn_yolov3.c:146:29: warning: unused variable ‘data’ [-Wunused-variable]
  146 |     uint8_t *               data;
      |                             ^~~~
At top level:
vnn_yolov3.c:94:17: warning: ‘load_data’ defined but not used [-Wunused-function]
   94 | static uint8_t* load_data
      |                 ^~~~~~~~~
  COMPILE /home/khadas/sajjad/eachoptime2/op_test_yolov3/main.c
  COMPILE /home/khadas/sajjad/eachoptime2/op_test_yolov3/vnn_post_process.c
make: Nothing to be done for 'all'.
khadas@Khadas-teco:~/sajjad/eachoptime2/op_test_yolov3/bin_r_cv4$ ./yolov3 ../yolov3_88.nb ../1080p.bmp
[     1] HAL user version 6.4.6.2.345497
[     2] HAL kernel version 6.4.6.2.345497
E [vnn_CreateYolov3:196]Create graph fail.
E [vnn_CreateNeuralNetwork:196]CHECK PTR 196
E [main:229]CHECK PTR 229
khadas@Khadas-teco:~/sajjad/eachoptime2/op_test_yolov3/bin_r_cv4$ ./yolov3 ../yolov3.export.data ../1080p.bmp
[     1] HAL user version 6.4.6.2.345497
[     2] HAL kernel version 6.4.6.2.345497
E [vnn_CreateYolov3:196]Create graph fail.
E [vnn_CreateNeuralNetwork:196]CHECK PTR 196
E [main:229]CHECK PTR 229
khadas@Khadas-teco:~/sajjad/eachoptime2/op_test_yolov3/bin_r_cv4$

@enggsajjad GitHub - yan-wyb/Just_for_get_op_time

This repo is still based on 6.4.6.2, you can update it manually, or wait for my update next week

Can you please guide me (some starting points) how to update it manually? Thanks alot!

@enggsajjad All the needed things can be found in aml-npu_sdk, and the corresponding so files can be found directly from /usr/lib on the board

Yes, I already found the first ones, and will see the .so files. Thanks