edgeai-tidl-tools
edgeai-tidl-tools copied to clipboard
Failed to compile onnx model if use PSDK-RTOS's tidl_tools
SDK_version : ti-processor-sdk-rtos-j721s2-evm-08_05_00_11 TIDL_version : tidl_j721s2_08_05_00_16
I run commands below
export DEVICE=j7
export TIDL_TOOLS_PATH=${PSDKR_PATH}/tidl_j721s2_08_05_00_16/tidl_tools
source ./setup.sh
and all passed. but error occured when I compile the onnx model: resnet18_opset9.onnx
MEM: Init ... !!!
MEM: Init ... Done !!!
0.0s: VX_ZONE_INIT:Enabled
0.2s: VX_ZONE_ERROR:Enabled
0.3s: VX_ZONE_WARNING:Enabled
0.929s: VX_ZONE_INIT:[tivxInit:184] Initialization Done !!!
0.21883s: VX_ZONE_ERROR:[tivxTIDLValidate:193] 'outArgs' should be a user_data_object of name:
TIDL_outArgs
0.21892s: VX_ZONE_ERROR:[ownGraphNodeKernelValidate:531] node kernel validate failed for kernel com.ti.tidl:1:1 at index 0
0.21894s: VX_ZONE_ERROR:[vxVerifyGraph:1941] Node kernel Validate failed
0.21895s: VX_ZONE_ERROR:[vxVerifyGraph:2109] Graph verify failed
TIDL_RT_OVX: ERROR: Verifying TIDL graph ... Failed !!!
TIDL_RT_OVX: ERROR: Verify OpenVX graph failed
0.22304s: VX_ZONE_ERROR:[tivxTIDLValidate:193] 'outArgs' should be a user_data_object of name:
TIDL_outArgs
0.22308s: VX_ZONE_ERROR:[ownGraphNodeKernelValidate:531] node kernel validate failed for kernel com.ti.tidl:1:1 at index 0
0.22309s: VX_ZONE_ERROR:[vxVerifyGraph:1941] Node kernel Validate failed
0.22310s: VX_ZONE_ERROR:[vxVerifyGraph:2109] Graph verify failed
0.22330s: VX_ZONE_ERROR:[ownGraphScheduleGraphWrapper:799] graph is not in a state required to be scheduled
0.22332s: VX_ZONE_ERROR:[vxProcessGraph:734] schedule graph failed
0.22333s: VX_ZONE_ERROR:[vxProcessGraph:739] wait graph failed
ERROR: Running TIDL graph ... Failed !!!
********** Frame Index 1 : Running float inference **********
0.48069s: VX_ZONE_ERROR:[tivxTIDLValidate:193] 'outArgs' should be a user_data_object of name:
TIDL_outArgs
0.48079s: VX_ZONE_ERROR:[ownGraphNodeKernelValidate:531] node kernel validate failed for kernel com.ti.tidl:1:1 at index 0
0.48081s: VX_ZONE_ERROR:[vxVerifyGraph:1941] Node kernel Validate failed
0.48082s: VX_ZONE_ERROR:[vxVerifyGraph:2109] Graph verify failed
0.48103s: VX_ZONE_ERROR:[ownGraphScheduleGraphWrapper:799] graph is not in a state required to be scheduled
0.48104s: VX_ZONE_ERROR:[vxProcessGraph:734] schedule graph failed
0.48106s: VX_ZONE_ERROR:[vxProcessGraph:739] wait graph failed
ERROR: Running TIDL graph ... Failed !!!
********** Frame Index 2 : Running fixed point mode for calibration **********
~~~~~Running TIDL in PC emulation mode to collect Activations range for each layer~~~~~
...
This error is disappeared if I export TIDL_TOOLS_PATH=${this_repo}/tidl_tools
, and run the same python script
MEM: Init ... !!!
MEM: Init ... Done !!!
0.0s: VX_ZONE_INIT:Enabled
0.13s: VX_ZONE_ERROR:Enabled
0.14s: VX_ZONE_WARNING:Enabled
0.8960s: VX_ZONE_INIT:[tivxInit:184] Initialization Done !!!
********** Frame Index 1 : Running float inference **********
********** Frame Index 2 : Running fixed point mode for calibration **********
What should I do to use ${PSDKR_PATH}/***/tidl_tools to compile onnx model?
Is TDA4VL could not use ${this_repo}/tidl_tools/osrt_deps/*
which build by TDA4VM sdk?
And how to build tidl_tools/osrt_deps/*
use TDA4VL sdk?