Error compiling horizontal-text-detection-0001 model by myriad_compiler

[asidsunrise]

Hello! I have some problems compiling Intel pre-trained model for text detection

(https://docs.openvinotoolkit.org/2021.2/omz_models_intel_horizontal_text_detection_0001_description_horizontal_text_detection_0001.html).

The error is "get_element_type() must be called on a node with exactly one output."

Other text detection models compile successfully. Is it a bug or I do something wrong? Screenshot is below the post. Thank You!

[Iffa_Intel]

Hi,

According to your screenshot, it's just a Warning & shouldn't be a problem.

I recommend for you to use compile_tool instead and specify the device type (-d MYRIAD).

You can refer to my attachment.

[asidsunrise]

Thanks for the quick response! I executed the same code with compiler_tool and got the same output is in your terminal screenshot. But the .blob file for this model was not created (like it was with models text-detection-0003 and -0004). Was it created in your case?

[Azizul_Intel · Accepted solution]

Hi Alexey,

The model (horizontal-text-detection-0001.xml) does not work on MyriadX due to some layers are not implemented. Thus, you are seeing the indicated error with no .blob file generated.

Regards,

Azizul

[asidsunrise]

Now it’s clear, thank You!

[Azizul_Intel]

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

Regards,

Azizul