Model compiled but not contain allowedNode.txt

[ParaImage]

compile a onnx model with "tidl_model_import.out" successfully but out files not contain "allowedNode.txt".

model file:

c.zip

config file :

modelType          = 2
inputNetFile      = "/mnt/Codes/Installtion/TI/compile/c.onnx"
outputNetFile      = "/mnt/Codes/Installtion/TI/compile/out/tidl_model.bin"
outputParamsFile   = "/mnt/Codes/Installtion/TI/compile/out/tidl_params_"
inDataNorm  = 1
inMean = 123.675 116.28 103.53
inScale = 0.017125 0.017507 0.017429
resizeWidth = 256
resizeHeight = 256
inWidth  = 256
inHeight = 256 
inNumChannels = 3
postProcType = 1
inData  =   "/mnt/Codes/Installtion/TI/edgeai-tidl-tools/tidl_tools/sample_list.txt"

out log file:

./tidl_model_import.out com.cfg
ONNX Model (Proto) File  : /mnt/Codes/Installtion/TI/compile/c.onnx  
TIDL Network File      : /mnt/Codes/Installtion/TI/compile/out/tidl_model.bin  
TIDL IO Info File      : /mnt/Codes/Installtion/TI/compile/out/tidl_params_  
Current ONNX OpSet Version   : 11  
Empty prototxt path, running calibration

~~~~~Running TIDL in PC emulation mode to collect Activations range for each layer~~~~~

Processing config file #0 : /mnt/Codes/Installtion/TI/compile/out/com.cfg.qunat_stats_config.txt 
 Freeing memory for user provided Net
 ----------------------- TIDL Process with REF_ONLY FLOW ------------------------

#    0 . .. T     343.15  .... ..... ...
 A :     0, 0.0000, 0.0000,   599 .... .....
#    1 . .. T     335.91  .... ..... ...
 A :     0, 0.0000, 0.0000,   725 .... .....
#    2 . .. T     334.78  .... ..... ...
 A :     0, 0.0000, 0.0000,   725 .... .....
#    3 . .. T     335.81  .... ..... ...
 A :     0, 0.0000, 0.0000,   725 .... .....
#    4 . .. T     335.02  .... ..... ...
 A :     0, 0.0000, 0.0000,   599 .... .....
#    5 . .. T     335.30  .... ..... ...
 A :     0, 0.0000, 0.0000,   599 .... .....
 
 
 *****************   Calibration iteration number 0 started ************************ 
 
 
 
Empty prototxt path, running calibration

~~~~~Running TIDL in PC emulation mode to collect Activations range for each layer~~~~~

Processing config file #0 : /mnt/Codes/Installtion/TI/compile/out/com.cfg.qunat_stats_config.txt 
 Freeing memory for user provided Net
 ----------------------- TIDL Process with REF_ONLY FLOW ------------------------

#    0 . .. T     144.98  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
#    1 . .. T     142.05  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
#    2 . .. T     141.93  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
#    3 . .. T     141.66  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
#    4 . .. T     142.19  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
#    5 . .. T     142.15  .... ..... ...
 A :     0, 1.0000, 1.0000,     0 .... .....
 
 
 *****************   Calibration iteration number 0 completed ************************ 
 
 
 
Empty prototxt path, running calibration

------------------ Network Compiler Traces -----------------------------
successful Memory allocation
successful Workload Creation
****************************************************
**                ALL MODEL CHECK PASSED          **
****************************************************

[vtrip97]

@ParaImage - Can you refer to python examples for model compilation & inference? Those artifacts are only generated if you are using the ONNX-RT based flow to compile your model (Take a look at ort_ep.py)

[VoNhatVinh]

I also meet this issue. Did anyone solve it?