• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Subscribers 99 subscribers
  • Views 1415 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

TDA4VM: [multi camera] when graph has 2 tivxVpacMscPyramidNode, vxVerifyGraph is failed

TDA4VM: Model infers on PC emulation but not on target

Gildardo
Prodigy 210 points

Part Number: TDA4VM

Hi,

I have updated to PSDK RTOS 08_01_00_13 / tidl_j7_08_01_00_05. I am testing yolov3 (public version) from

https://github.com/onnx/models/blob/master/vision/object_detection_segmentation/yolov3/model/yolov3-10.onnx

as pointed in the guide:

https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_01_00_13/exports/docs/tidl_j7_08_01_00_05/ti_dl/docs/user_guide_html/md_tidl_models_info.html

I successfully imported the model and tested on PC using PC_dsp_test_dl_algo.out. However, when infering on target with TI_DEVICE_a72_test_dl_algo_host_rt.out there is the following error:


Processing config file #0 : ./testvecs/config/infer/public/onnx/tidl_infer_yolo3.txt
APP: Init ... !!!
MEM: Init ... !!!
MEM: Initialized DMA HEAP (fd=4) !!!
MEM: Init ... Done !!!
IPC: Init ... !!!
IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
 14353.852135 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
 14353.852221 s:  VX_ZONE_INIT:Enabled
 14353.852230 s:  VX_ZONE_ERROR:Enabled
 14353.852241 s:  VX_ZONE_WARNING:Enabled
 14353.852732 s:  VX_ZONE_INIT:[tivxInitLocal:130] Initialization Done !!!
 14353.852934 s:  VX_ZONE_INIT:[tivxHostInitLocal:86] Initialization Done for HOST !!!
 14353.894107 s:  VX_ZONE_ERROR:[ownContextSendCmd:815] Command ack message returned failure cmd_status: -1
 14353.894132 s:  VX_ZONE_ERROR:[ownContextSendCmd:851] tivxEventWait() failed.
 14353.894156 s:  VX_ZONE_ERROR:[ownNodeKernelInit:538] Target kernel, TIVX_CMD_NODE_CREATE failed for node TIDLNode
 14353.894175 s:  VX_ZONE_ERROR:[ownNodeKernelInit:539] Please be sure the target callbacks have been registered for this core
 14353.894192 s:  VX_ZONE_ERROR:[ownNodeKernelInit:540] If the target callbacks have been registered, please ensure no errors are occurring within the create callback of this kernel
 14353.894209 s:  VX_ZONE_ERROR:[ownGraphNodeKernelInit:583] kernel init for node 0, kernel com.ti.tidl ... failed !!!
 14353.894229 s:  VX_ZONE_ERROR:[vxVerifyGraph:2055] Node kernel init failed
 14353.894245 s:  VX_ZONE_ERROR:[vxVerifyGraph:2109] Graph verify failed
TIDL_RT_OVX: ERROR: Verifying TIDL graph ... Failed !!!
TIDL_RT_OVX: ERROR: Verify OpenVX graph failed
Error at line:   452 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt/test/a72/../../../test/src/tidl_tb.c, of function : tidlMultiInstanceTest
Invalid Error Type!
root@j7-evm:/opt/tidl_test# [C7x_1 ]  14353.893927 s:  VX_ZONE_ERROR:[tivxMemBufferUnmap:335] tivxMemBufferUnmap failed (either pointer is NULL or size is 0)
[C7x_1 ]  14353.893956 s:  VX_ZONE_ERROR:[tivxMemBufferUnmap:335] tivxMemBufferUnmap failed (either pointer is NULL or size is 0)

The import configuration:

modelType          = 2
numParamBits       = 8
numFeatureBits     = 8
quantizationStyle  = 3

inputNetFile      =  "/mnt/d/Research/Features/OD/yolo3_public/model/yolov3-10.onnx"
outputNetFile      = "/mnt/d/Research/Features/OD/yolo3_public/ti/model_out/tild_net_yolov3-10.bin"
outputParamsFile   = "/mnt/d/Research/Features/OD/yolo3_public/ti/model_out/tidl_io_yolov3-10_"
inDataNorm  = 1
inMean = 0 0 0
inScale = 0.003921568627 0.003921568627 0.003921568627
inDataFormat = 1
inWidth  = 416
inHeight = 416 
inNumChannels = 3

numFrames = 1
inFileFormat  = 1
inData = "/mnt/d/Research/Features/OD/yolo3_public/ti/test_images/ti_lindau_000020.bmp"

perfSimConfig = "/mnt/d/Research/Features/OD/yolo3_public/ti/scripts/device_config.cfg"
inElementType = 0
metaArchType = 4
metaLayersNamesList = "/mnt/d/Research/Features/OD/yolo3_public/ti/scripts/A/tidl_import_yolo3_metaarch.prototxt"
postProcType = 2

The infer configuration on target:

inFileFormat    = 2
numFrames   = 1
netBinFile      = "testvecs/config/tidl_models/onnx/tild_net_yolov3-10.bin"
ioConfigFile   = "testvecs/config/tidl_models/onnx/tidl_io_yolov3-10_1.bin"
inData  =   testvecs/config/detection_list.txt
outData =   testvecs/output/tidl_yolo3_od.bin
inResizeMode = 0
debugTraceLevel = 0
writeTraceLevel = 0
postProcType = 2

Thank you for your support

  • 0
    Prodigy 210 points

    Update:

    After rebooting the target, inference on yolov3  and the provided example jacintonet11v2 ran correctly.

    However, I am trying to infer yolov5 on target and shows the following error:

    Error at line:   741 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt/test/a72/../../../test/src/tidl_tb_utils.c, of function : tidl_tb_dataConvert
    Error at line:   741 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt/test/a72/../../../test/src/tidl_tb_utils.c, of function : tidl_tb_dataConvert
    Error at line:   741 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt/test/a72/../../../test/src/tidl_tb_utils.c, of function : tidl_tb_dataConvert
    Error at line:   741 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt^C
    Clean up and exit while handling signal 2
    Application did not close some rpmsg_char devices
    Error at line:   741 : in file /home/gtbldadm/psdk_installer_build_top_workarea/scratch_workarea/ti-processor-sdk-rtos-j721e-evm-08_01_00_13/tidl_j7_08_01_00_05/ti_dl/rt/test/a72/../../../test/src/tidl_tb_utils.c, of function : tidl_tb_dataConvert

    The top lines are repeated until I kill the execution with Ctrl + C. That is why we can not see the first log messages. After getting this error, any other model can be infered on the target, the original error regarding graph verification (see first post above) is shown on already verified models (yolov3 and jacintonet11v2) until rebooting again.

    The same story for yolov5. It is imported correctly as well as PC emulation, just target inference fails.
    I got the yolov5 from EdgeAi Model Zoo repository link :

    github.com/.../weights

    Import configuration

    modelType          = 2
    numParamBits       = 8
    numFeatureBits     = 8
    quantizationStyle  = 3
    inputNetFile      =  "/mnt/d/Research/Features/OD/yolov5_edgeai/model/yolov5s6_640_ti_lite_37p4_56p0.onnx"
    outputNetFile      = "/mnt/d/Research/Features/OD/yolov5_edgeai/ti/model_out/tild_net_yolov5s6_640_ti_lite_37p4_56p0.bin"
    outputParamsFile   = "/mnt/d/Research/Features/OD/yolov5_edgeai/ti/model_out/tidl_io_yolov5s6_640_ti_lite_37p4_56p0_"
    inDataNorm  = 1
    inMean = 0 0 0
    inScale = 0.003921568627 0.003921568627 0.003921568627
    inDataFormat = 1
    inWidth  = 640
    inHeight = 640
    inNumChannels = 3

    numFrames = 5
    inFileFormat  = 0
    inData = "/mnt/d/Research/Features/OD/yolo3_public/ti/test_images/ti_lindau_000020.bmp"

    perfSimConfig = "/mnt/d/Research/Features/OD/yolov5_edgeai/ti/scripts/device_config.cfg"
    inElementType = 0
    metaArchType = 6
    metaLayersNamesList = "/mnt/d/Research/Features/OD/yolov5_edgeai/model/yolov5s6_640_ti_lite_metaarch.prototxt"
    postProcType = 2



    Inference configuration on TDA

    inFileFormat=0
    numFrames   = 1
    netBinFile      = "testvecs/config/tidl_models/onnx/tild_net_yolov5s6_640_ti_lite_37p4_56p0.bin"
    ioConfigFile    = "testvecs/config/tidl_models/onnx/tidl_io_yolov5s6_640_ti_lite_37p4_56p0_1.bin"
    inData  =   "testvecs/input/ti_lindau_000020.bmp"
    outData =   "./out/tidl_yolo5_od.bin"
    inResizeMode = 0
    debugTraceLevel = 0
    writeTraceLevel = 0
    postProcType = 2

  • 0 in reply to Gildardo
    Intellectual 750 points

    Hi Gildardo:

    I met the same problem, have you solved it ? Thanks.

    Kind Regards,

    Damon

  • 0 in reply to Damon@INVO
    Prodigy 210 points

    Hi Damon,

    I guess the main problem is vxVerifyGraph(). It randomly fails. I just skipped the verification of the models and inference was executed successfully. I also re-imported the  models with the same configuration as a precaution. Sometimes I get an error message that he inference failed with no reason. From there I cannot longer infer anything until I reboot or power cycle.

    Best

    Gildardo

  • 0 in reply to Gildardo
    TI__Guru 56820 points

    Hi Gildardo,

        I see that you are using inFileFormat=1 ( raw data) in import config file whereas you are using inFileFormat = 0 in infer config file. Can you try to use any one of them in both the files. Please read the following documentation to understand inFileFormat and inData parameters: 

    https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_01_00_11/exports/docs/tidl_j7_08_01_00_05/ti_dl/docs/user_guide_html/md_tidl_sample_test.html#tidl_inference_2

       Let me know if this solves your issue.

    Regards,

    ANshu

  • 0 in reply to Gildardo
    Intellectual 750 points

    Hi Gildardo:

    Thank you for your answer.  For some reason, I didn't seem to change anything, but now it's fine.  Ha ha.

    Kind regards,

    Damon

  • 0 in reply to Damon@INVO
    TI__Guru 56820 points

    Hi Gildardo,

         Can you confirm if based on above suggestion ( mentioned in my last reply), this issue is resolved for you also?


    Regards,

    Anshu

  • 0 in reply to Anshu Jain
    Prodigy 210 points

    Hi Anshu, the suggestion that you did works in combination with TI_DEVICE_a72_test_dl_algo_host_rt.out. However, the problem of vxVerifyGraph() in my own application is still there.

    Best

  • 0 in reply to Gildardo
    TI__Guru 56820 points

    Hi Gildardo,

         Can you set debugTraceLevel = 1 in inference config file and share the logs with it? Make sure to follow all steps as mentioned in below documentation:

    Regards,

    Anshu

  • 0 in reply to Anshu Jain
    TI__Guru 56820 points

    One correction. Please set debugTraceLevel = 2

    Regards,

    Anshu

  • 0 in reply to Anshu Jain
    Prodigy 210 points

    Below the inference log. I re-imported the model with inFileFormat=0 and use the same in infer .I re-enable vxVerifyGraph() in my application and apply the newly imported models. So far I have not been able to reproduce again the error. I will keep it enabled and let you know when it happens again.

    Processing config file #0 : tidl_infer_yolov5s6_public_tidl80113_TDA_Test.cfg
    Input : dataId=0, name=images_original, scale=1.000000
    Ouput : dataId=118, name=tidl_yol_od_output_layer, scale=1.000000
         14324368,     13.661 0xffff5a930010
    worstCaseDelay for Pre-emption is 0.5001593
    Network File Read done
    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
      Num|TIDL Layer Name               |Group |#Ins  |#Outs |Inbuf Ids                       |Outbuf Id |In NCHW                             |Out NCHW                            |
    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
        0|TIDL_DataLayer                |     0|    -1|     1|  x   x   x   x   x   x   x   x |  0       |       0        0        0        0 |       1        3      640      640 |
        1|TIDL_ConvolutionLayer         |     1|     1|     1|  0   x   x   x   x   x   x   x |  1       |       1        3      640      640 |       1       12      320      320 |
        2|TIDL_ConvolutionLayer         |     1|     1|     1|  1   x   x   x   x   x   x   x |  2       |       1       12      320      320 |       1       32      320      320 |
        3|TIDL_ConvolutionLayer         |     1|     1|     1|  2   x   x   x   x   x   x   x |  3       |       1       32      320      320 |       1       64      160      160 |
        4|TIDL_ConvolutionLayer         |     1|     1|     1|  3   x   x   x   x   x   x   x |  4       |       1       64      160      160 |       1       32      160      160 |
        5|TIDL_ConvolutionLayer         |     1|     1|     1|  3   x   x   x   x   x   x   x |  5       |       1       64      160      160 |       1       32      160      160 |
        6|TIDL_ConvolutionLayer         |     1|     1|     1|  5   x   x   x   x   x   x   x |  6       |       1       32      160      160 |       1       32      160      160 |
        7|TIDL_ConvolutionLayer         |     1|     1|     1|  6   x   x   x   x   x   x   x |  7       |       1       32      160      160 |       1       32      160      160 |
        8|TIDL_EltWiseLayer             |     1|     2|     1|  5   7   x   x   x   x   x   x |  8       |       1       32      160      160 |       1       32      160      160 |
        9|TIDL_ConcatLayer              |     1|     2|     1|  8   4   x   x   x   x   x   x |  9       |       1       32      160      160 |       1       64      160      160 |
       10|TIDL_ConvolutionLayer         |     1|     1|     1|  9   x   x   x   x   x   x   x | 10       |       1       64      160      160 |       1       64      160      160 |
       11|TIDL_ConvolutionLayer         |     1|     1|     1| 10   x   x   x   x   x   x   x | 11       |       1       64      160      160 |       1      128       80       80 |
       12|TIDL_ConvolutionLayer         |     1|     1|     1| 11   x   x   x   x   x   x   x | 12       |       1      128       80       80 |       1       64       80       80 |
       13|TIDL_ConvolutionLayer         |     1|     1|     1| 12   x   x   x   x   x   x   x | 14       |       1       64       80       80 |       1       64       80       80 |
       14|TIDL_ConvolutionLayer         |     1|     1|     1| 14   x   x   x   x   x   x   x | 15       |       1       64       80       80 |       1       64       80       80 |
       15|TIDL_EltWiseLayer             |     1|     2|     1| 12  15   x   x   x   x   x   x | 16       |       1       64       80       80 |       1       64       80       80 |
       16|TIDL_ConvolutionLayer         |     1|     1|     1| 16   x   x   x   x   x   x   x | 17       |       1       64       80       80 |       1       64       80       80 |
       17|TIDL_ConvolutionLayer         |     1|     1|     1| 17   x   x   x   x   x   x   x | 18       |       1       64       80       80 |       1       64       80       80 |
       18|TIDL_EltWiseLayer             |     1|     2|     1| 16  18   x   x   x   x   x   x | 19       |       1       64       80       80 |       1       64       80       80 |
       19|TIDL_ConvolutionLayer         |     1|     1|     1| 19   x   x   x   x   x   x   x | 20       |       1       64       80       80 |       1       64       80       80 |
       20|TIDL_ConvolutionLayer         |     1|     1|     1| 20   x   x   x   x   x   x   x | 21       |       1       64       80       80 |       1       64       80       80 |
       21|TIDL_EltWiseLayer             |     1|     2|     1| 19  21   x   x   x   x   x   x | 22       |       1       64       80       80 |       1       64       80       80 |
       22|TIDL_ConvolutionLayer         |     1|     1|     1| 11   x   x   x   x   x   x   x | 13       |       1      128       80       80 |       1       64       80       80 |
       23|TIDL_ConcatLayer              |     1|     2|     1| 22  13   x   x   x   x   x   x | 23       |       1       64       80       80 |       1      128       80       80 |
       24|TIDL_ConvolutionLayer         |     1|     1|     1| 23   x   x   x   x   x   x   x | 24       |       1      128       80       80 |       1      128       80       80 |
       25|TIDL_ConvolutionLayer         |     1|     1|     1| 24   x   x   x   x   x   x   x | 25       |       1      128       80       80 |       1      256       40       40 |
       26|TIDL_ConvolutionLayer         |     1|     1|     1| 25   x   x   x   x   x   x   x | 26       |       1      256       40       40 |       1      128       40       40 |
       27|TIDL_ConvolutionLayer         |     1|     1|     1| 26   x   x   x   x   x   x   x | 28       |       1      128       40       40 |       1      128       40       40 |
       28|TIDL_ConvolutionLayer         |     1|     1|     1| 28   x   x   x   x   x   x   x | 29       |       1      128       40       40 |       1      128       40       40 |
       29|TIDL_EltWiseLayer             |     1|     2|     1| 26  29   x   x   x   x   x   x | 30       |       1      128       40       40 |       1      128       40       40 |
       30|TIDL_ConvolutionLayer         |     1|     1|     1| 30   x   x   x   x   x   x   x | 31       |       1      128       40       40 |       1      128       40       40 |
       31|TIDL_ConvolutionLayer         |     1|     1|     1| 31   x   x   x   x   x   x   x | 32       |       1      128       40       40 |       1      128       40       40 |
       32|TIDL_EltWiseLayer             |     1|     2|     1| 30  32   x   x   x   x   x   x | 33       |       1      128       40       40 |       1      128       40       40 |
       33|TIDL_ConvolutionLayer         |     1|     1|     1| 33   x   x   x   x   x   x   x | 34       |       1      128       40       40 |       1      128       40       40 |
       34|TIDL_ConvolutionLayer         |     1|     1|     1| 34   x   x   x   x   x   x   x | 35       |       1      128       40       40 |       1      128       40       40 |
       35|TIDL_EltWiseLayer             |     1|     2|     1| 33  35   x   x   x   x   x   x | 36       |       1      128       40       40 |       1      128       40       40 |
       36|TIDL_ConvolutionLayer         |     1|     1|     1| 25   x   x   x   x   x   x   x | 27       |       1      256       40       40 |       1      128       40       40 |
       37|TIDL_ConcatLayer              |     1|     2|     1| 36  27   x   x   x   x   x   x | 37       |       1      128       40       40 |       1      256       40       40 |
       38|TIDL_ConvolutionLayer         |     1|     1|     1| 37   x   x   x   x   x   x   x | 38       |       1      256       40       40 |       1      256       40       40 |
       39|TIDL_ConvolutionLayer         |     1|     1|     1| 38   x   x   x   x   x   x   x | 39       |       1      256       40       40 |       1      384       20       20 |
       40|TIDL_ConvolutionLayer         |     1|     1|     1| 39   x   x   x   x   x   x   x | 40       |       1      384       20       20 |       1      192       20       20 |
       41|TIDL_ConvolutionLayer         |     1|     1|     1| 39   x   x   x   x   x   x   x | 41       |       1      384       20       20 |       1      192       20       20 |
       42|TIDL_ConvolutionLayer         |     1|     1|     1| 41   x   x   x   x   x   x   x | 42       |       1      192       20       20 |       1      192       20       20 |
       43|TIDL_ConvolutionLayer         |     1|     1|     1| 42   x   x   x   x   x   x   x | 43       |       1      192       20       20 |       1      192       20       20 |
       44|TIDL_EltWiseLayer             |     1|     2|     1| 41  43   x   x   x   x   x   x | 44       |       1      192       20       20 |       1      192       20       20 |
       45|TIDL_ConcatLayer              |     1|     2|     1| 44  40   x   x   x   x   x   x | 45       |       1      192       20       20 |       1      384       20       20 |
       46|TIDL_ConvolutionLayer         |     1|     1|     1| 45   x   x   x   x   x   x   x | 46       |       1      384       20       20 |       1      384       20       20 |
       47|TIDL_ConvolutionLayer         |     1|     1|     1| 46   x   x   x   x   x   x   x | 47       |       1      384       20       20 |       1      512       10       10 |
       48|TIDL_ConvolutionLayer         |     1|     1|     1| 47   x   x   x   x   x   x   x | 48       |       1      512       10       10 |       1      256       10       10 |
       49|TIDL_PoolingLayer             |     1|     1|     1| 48   x   x   x   x   x   x   x | 49       |       1      256       10       10 |       1      256       10       10 |
       50|TIDL_PoolingLayer             |     1|     1|     1| 48   x   x   x   x   x   x   x | 50       |       1      256       10       10 |       1      256       10       10 |
       51|TIDL_PoolingLayer             |     1|     1|     1| 50   x   x   x   x   x   x   x | 52       |       1      256       10       10 |       1      256       10       10 |
       52|TIDL_PoolingLayer             |     1|     1|     1| 48   x   x   x   x   x   x   x | 51       |       1      256       10       10 |       1      256       10       10 |
       53|TIDL_PoolingLayer             |     1|     1|     1| 51   x   x   x   x   x   x   x | 53       |       1      256       10       10 |       1      256       10       10 |
       54|TIDL_PoolingLayer             |     1|     1|     1| 53   x   x   x   x   x   x   x | 54       |       1      256       10       10 |       1      256       10       10 |
       55|TIDL_ConcatLayer              |     1|     4|     1| 48  49  52  54   x   x   x   x | 55       |       1      256       10       10 |       1     1024       10       10 |
       56|TIDL_ConvolutionLayer         |     1|     1|     1| 55   x   x   x   x   x   x   x | 56       |       1     1024       10       10 |       1      512       10       10 |
       57|TIDL_ConvolutionLayer         |     1|     1|     1| 56   x   x   x   x   x   x   x | 57       |       1      512       10       10 |       1      256       10       10 |
       58|TIDL_ConvolutionLayer         |     1|     1|     1| 57   x   x   x   x   x   x   x | 59       |       1      256       10       10 |       1      256       10       10 |
       59|TIDL_ConvolutionLayer         |     1|     1|     1| 59   x   x   x   x   x   x   x | 60       |       1      256       10       10 |       1      256       10       10 |
       60|TIDL_ConvolutionLayer         |     1|     1|     1| 56   x   x   x   x   x   x   x | 58       |       1      512       10       10 |       1      256       10       10 |
       61|TIDL_ConcatLayer              |     1|     2|     1| 60  58   x   x   x   x   x   x | 61       |       1      256       10       10 |       1      512       10       10 |
       62|TIDL_ConvolutionLayer         |     1|     1|     1| 61   x   x   x   x   x   x   x | 62       |       1      512       10       10 |       1      512       10       10 |
       63|TIDL_ConvolutionLayer         |     1|     1|     1| 62   x   x   x   x   x   x   x | 63       |       1      512       10       10 |       1      384       10       10 |
       64|TIDL_ResizeLayer              |     1|     1|     1| 63   x   x   x   x   x   x   x | 64       |       1      384       10       10 |       1      384       20       20 |
       65|TIDL_ConcatLayer              |     1|     2|     1| 64  46   x   x   x   x   x   x | 65       |       1      384       20       20 |       1      768       20       20 |
       66|TIDL_ConvolutionLayer         |     1|     1|     1| 65   x   x   x   x   x   x   x | 66       |       1      768       20       20 |       1      192       20       20 |
       67|TIDL_ConvolutionLayer         |     1|     1|     1| 66   x   x   x   x   x   x   x | 68       |       1      192       20       20 |       1      192       20       20 |
       68|TIDL_ConvolutionLayer         |     1|     1|     1| 68   x   x   x   x   x   x   x | 69       |       1      192       20       20 |       1      192       20       20 |
       69|TIDL_ConvolutionLayer         |     1|     1|     1| 65   x   x   x   x   x   x   x | 67       |       1      768       20       20 |       1      192       20       20 |
       70|TIDL_ConcatLayer              |     1|     2|     1| 69  67   x   x   x   x   x   x | 70       |       1      192       20       20 |       1      384       20       20 |
       71|TIDL_ConvolutionLayer         |     1|     1|     1| 70   x   x   x   x   x   x   x | 71       |       1      384       20       20 |       1      384       20       20 |
       72|TIDL_ConvolutionLayer         |     1|     1|     1| 71   x   x   x   x   x   x   x | 72       |       1      384       20       20 |       1      256       20       20 |
       73|TIDL_ResizeLayer              |     1|     1|     1| 72   x   x   x   x   x   x   x | 73       |       1      256       20       20 |       1      256       40       40 |
       74|TIDL_ConcatLayer              |     1|     2|     1| 73  38   x   x   x   x   x   x | 74       |       1      256       40       40 |       1      512       40       40 |
       75|TIDL_ConvolutionLayer         |     1|     1|     1| 74   x   x   x   x   x   x   x | 75       |       1      512       40       40 |       1      128       40       40 |
       76|TIDL_ConvolutionLayer         |     1|     1|     1| 75   x   x   x   x   x   x   x | 77       |       1      128       40       40 |       1      128       40       40 |
       77|TIDL_ConvolutionLayer         |     1|     1|     1| 77   x   x   x   x   x   x   x | 78       |       1      128       40       40 |       1      128       40       40 |
       78|TIDL_ConvolutionLayer         |     1|     1|     1| 74   x   x   x   x   x   x   x | 76       |       1      512       40       40 |       1      128       40       40 |
       79|TIDL_ConcatLayer              |     1|     2|     1| 78  76   x   x   x   x   x   x | 79       |       1      128       40       40 |       1      256       40       40 |
       80|TIDL_ConvolutionLayer         |     1|     1|     1| 79   x   x   x   x   x   x   x | 80       |       1      256       40       40 |       1      256       40       40 |
       81|TIDL_ConvolutionLayer         |     1|     1|     1| 80   x   x   x   x   x   x   x | 81       |       1      256       40       40 |       1      128       40       40 |
       82|TIDL_ResizeLayer              |     1|     1|     1| 81   x   x   x   x   x   x   x | 82       |       1      128       40       40 |       1      128       80       80 |
       83|TIDL_ConcatLayer              |     1|     2|     1| 82  24   x   x   x   x   x   x | 83       |       1      128       80       80 |       1      256       80       80 |
       84|TIDL_ConvolutionLayer         |     1|     1|     1| 83   x   x   x   x   x   x   x | 84       |       1      256       80       80 |       1       64       80       80 |
       85|TIDL_ConvolutionLayer         |     1|     1|     1| 84   x   x   x   x   x   x   x | 86       |       1       64       80       80 |       1       64       80       80 |
       86|TIDL_ConvolutionLayer         |     1|     1|     1| 86   x   x   x   x   x   x   x | 87       |       1       64       80       80 |       1       64       80       80 |
       87|TIDL_ConvolutionLayer         |     1|     1|     1| 83   x   x   x   x   x   x   x | 85       |       1      256       80       80 |       1       64       80       80 |
       88|TIDL_ConcatLayer              |     1|     2|     1| 87  85   x   x   x   x   x   x | 88       |       1       64       80       80 |       1      128       80       80 |
       89|TIDL_ConvolutionLayer         |     1|     1|     1| 88   x   x   x   x   x   x   x | 89       |       1      128       80       80 |       1      128       80       80 |
       90|TIDL_ConvolutionLayer         |     1|     1|     1| 89   x   x   x   x   x   x   x | 90       |       1      128       80       80 |       1      128       40       40 |
       91|TIDL_ConcatLayer              |     1|     2|     1| 90  81   x   x   x   x   x   x | 92       |       1      128       40       40 |       1      256       40       40 |
       92|TIDL_ConvolutionLayer         |     1|     1|     1| 92   x   x   x   x   x   x   x | 93       |       1      256       40       40 |       1      128       40       40 |
       93|TIDL_ConvolutionLayer         |     1|     1|     1| 93   x   x   x   x   x   x   x | 95       |       1      128       40       40 |       1      128       40       40 |
       94|TIDL_ConvolutionLayer         |     1|     1|     1| 95   x   x   x   x   x   x   x | 96       |       1      128       40       40 |       1      128       40       40 |
       95|TIDL_ConvolutionLayer         |     1|     1|     1| 92   x   x   x   x   x   x   x | 94       |       1      256       40       40 |       1      128       40       40 |
       96|TIDL_ConcatLayer              |     1|     2|     1| 96  94   x   x   x   x   x   x | 97       |       1      128       40       40 |       1      256       40       40 |
       97|TIDL_ConvolutionLayer         |     1|     1|     1| 97   x   x   x   x   x   x   x | 98       |       1      256       40       40 |       1      256       40       40 |
       98|TIDL_ConvolutionLayer         |     1|     1|     1| 98   x   x   x   x   x   x   x | 99       |       1      256       40       40 |       1      256       20       20 |
       99|TIDL_ConcatLayer              |     1|     2|     1| 99  72   x   x   x   x   x   x |101       |       1      256       20       20 |       1      512       20       20 |
      100|TIDL_ConvolutionLayer         |     1|     1|     1|101   x   x   x   x   x   x   x |102       |       1      512       20       20 |       1      192       20       20 |
      101|TIDL_ConvolutionLayer         |     1|     1|     1|102   x   x   x   x   x   x   x |104       |       1      192       20       20 |       1      192       20       20 |
      102|TIDL_ConvolutionLayer         |     1|     1|     1|104   x   x   x   x   x   x   x |105       |       1      192       20       20 |       1      192       20       20 |
      103|TIDL_ConvolutionLayer         |     1|     1|     1|101   x   x   x   x   x   x   x |103       |       1      512       20       20 |       1      192       20       20 |
      104|TIDL_ConcatLayer              |     1|     2|     1|105 103   x   x   x   x   x   x |106       |       1      192       20       20 |       1      384       20       20 |
      105|TIDL_ConvolutionLayer         |     1|     1|     1|106   x   x   x   x   x   x   x |107       |       1      384       20       20 |       1      384       20       20 |
      106|TIDL_ConvolutionLayer         |     1|     1|     1|107   x   x   x   x   x   x   x |108       |       1      384       20       20 |       1      384       10       10 |
      107|TIDL_ConcatLayer              |     1|     2|     1|108  63   x   x   x   x   x   x |110       |       1      384       10       10 |       1      768       10       10 |
      108|TIDL_ConvolutionLayer         |     1|     1|     1|110   x   x   x   x   x   x   x |111       |       1      768       10       10 |       1      256       10       10 |
      109|TIDL_ConvolutionLayer         |     1|     1|     1|111   x   x   x   x   x   x   x |113       |       1      256       10       10 |       1      256       10       10 |
      110|TIDL_ConvolutionLayer         |     1|     1|     1|113   x   x   x   x   x   x   x |114       |       1      256       10       10 |       1      256       10       10 |
      111|TIDL_ConvolutionLayer         |     1|     1|     1|110   x   x   x   x   x   x   x |112       |       1      768       10       10 |       1      256       10       10 |
      112|TIDL_ConcatLayer              |     1|     2|     1|114 112   x   x   x   x   x   x |115       |       1      256       10       10 |       1      512       10       10 |
      113|TIDL_ConvolutionLayer         |     1|     1|     1|115   x   x   x   x   x   x   x |116       |       1      512       10       10 |       1      512       10       10 |
      114|TIDL_ConvolutionLayer         |     1|     1|     1|116   x   x   x   x   x   x   x |117       |       1      512       10       10 |       1      255       10       10 |
      115|TIDL_ConvolutionLayer         |     1|     1|     1| 89   x   x   x   x   x   x   x | 91       |       1      128       80       80 |       1      255       80       80 |
      116|TIDL_ConvolutionLayer         |     1|     1|     1| 98   x   x   x   x   x   x   x |100       |       1      256       40       40 |       1      255       40       40 |
      117|TIDL_ConvolutionLayer         |     1|     1|     1|107   x   x   x   x   x   x   x |109       |       1      384       20       20 |       1      255       20       20 |
      118|TIDL_DetectionOutputLayer     |     1|     4|     1| 91 100 109 117   x   x   x   x |118       |       1      255       80       80 |       1        1        1     1404 |
      119|TIDL_DataLayer                |     0|     1|    -1|118   x   x   x   x   x   x   x |  0       |       1        1        1     1404 |       0        0        0        0 |
    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
    APP: Init ... !!!
    MEM: Init ... !!!
    MEM: Initialized DMA HEAP (fd=4) !!!
    MEM: Init ... Done !!!
    IPC: Init ... !!!
    IPC: Init ... Done !!!
    REMOTE_SERVICE: Init ... !!!
    REMOTE_SERVICE: Init ... Done !!!
      6167.464148 s: GTC Frequency = 200 MHz
    APP: Init ... Done !!!
      6167.470219 s:  VX_ZONE_INIT:Enabled
      6167.470242 s:  VX_ZONE_ERROR:Enabled
      6167.470250 s:  VX_ZONE_WARNING:Enabled
      6167.472634 s:  VX_ZONE_INIT:[tivxInitLocal:130] Initialization Done !!!
      6167.474577 s:  VX_ZONE_INIT:[tivxHostInitLocal:86] Initialization Done for HOST !!!

     Instance created for  tidl_infer_yolov5s6_public_tidl80113_TDA_Test.cfg

    Processing Cnt :    0, InstCnt :    0 testvecs/config/tidl_models/onnx/tild_net_yolov5s6_640_ti_lite_37p4_56p0Test.bin!
    /home/root/cnn_sdk/test_images/OD_cc_640x/TP16_002340_640x.bmp
     ----------------------- TIDL Process with TARGET DATA FLOW ------------------------

    # NETWORK_EXECUTION_TIME =    38.75 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) =     0.00,     0.00,     0.00 (in Mega Bytes/frame) .../home/root/cnn_sdk/test_images/OD_cc_640x/TP16_002340_640x.bmp
    /home/root/cnn_sdk/test_images/OD_cc_640x/TP16_002340_640x.bmp
     .... .....
    Processing Cnt :    1, InstCnt :    0 testvecs/config/tidl_models/onnx/tild_net_yolov5s6_640_ti_lite_37p4_56p0Test.bin!
    /home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_15-14-29_christian_002903_640x.bmp
     ----------------------- TIDL Process with TARGET DATA FLOW ------------------------

    # NETWORK_EXECUTION_TIME =    38.25 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) =     0.00,     0.00,     0.00 (in Mega Bytes/frame) .../home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_15-14-29_christian_002903_640x.bmp
    /home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_15-14-29_christian_002903_640x.bmp
     .... .....
    Processing Cnt :    2, InstCnt :    0 testvecs/config/tidl_models/onnx/tild_net_yolov5s6_640_ti_lite_37p4_56p0Test.bin!
    /home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_17-09-19_Tarek_000234_640x.bmp
     ----------------------- TIDL Process with TARGET DATA FLOW ------------------------

    # NETWORK_EXECUTION_TIME =    38.17 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) =     0.00,     0.00,     0.00 (in Mega Bytes/frame) .../home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_17-09-19_Tarek_000234_640x.bmp
    /home/root/cnn_sdk/test_images/OD_cc_640x/2021-04-19_17-09-19_Tarek_000234_640x.bmp
     .... .....
    Processing Cnt :    3, InstCnt :    0 testvecs/config/tidl_models/onnx/tild_net_yolov5s6_640_ti_lite_37p4_56p0Test.bin!
    /home/root/cnn_sdk/test_images/OD_cc_640x/s01n_CM_049022_640x.bmp
     ----------------------- TIDL Process with TARGET DATA FLOW ------------------------

    # NETWORK_EXECUTION_TIME =    38.21 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) =     0.00,     0.00,     0.00 (in Mega Bytes/frame) .../home/root/cnn_sdk/test_images/OD_cc_640x/s01n_CM_049022_640x.bmp
    /home/root/cnn_sdk/test_images/OD_cc_640x/s01n_CM_049022_640x.bmp
     .... .....  6168.282993 s:  VX_ZONE_INIT:[tivxHostDeInitLocal:100] De-Initialization Done for HOST !!!
      6168.287350 s:  VX_ZONE_INIT:[tivxDeInitLocal:193] De-Initialization Done !!!
    APP: Deinit ... !!!
    REMOTE_SERVICE: Deinit ... !!!
    REMOTE_SERVICE: Deinit ... Done !!!
    IPC: Deinit ... !!!
    IPC: DeInit ... Done !!!
    MEM: Deinit ... !!!
    MEM: Alloc's: 7 alloc's of 15632485 bytes
    MEM: Free's : 7 free's  of 15632485 bytes
    MEM: Open's : 0 allocs  of 0 bytes
    MEM: Deinit ... Done !!!
    APP: Deinit ... Done !!!

  • +1 in reply to Gildardo
    TI__Guru 56820 points

    Hi Gildardo,

         From the logs it looks like all models are running fine. Let me know if we can close this thread?

    Regards,

    Anshu

  • 0 in reply to Anshu Jain
    Prodigy 210 points

    Hi Anshu, I agree, If the problem comes up, I open a new issue.

    Thanks for checking the log and for your time.

    Best

    Gildardo