why i use tensorflow SSD mobilenet v2 training my dataset , get a wrong model can't convert to IR

[fa2233]

it's auto save model    show  ?x?x?x3   

i set training num_steps 5000 

why it output this kind.

[Iffa_Intel]

Greetings,

Based on the errors that you get as per attached photo, it seems like you missed out a couple of instruction required to convert the model such as:

python mo_tf.py -m frozen_inference_graph.pb --tensorflow_use_custom_operations_config ssd_v2_support.json --tensorflow_object_detection_api pipeline.config

This is the complete instruction to convert.

You can refer to this video for the complete and step by step instructions: https://www.youtube.com/watch?v=cbdS3BjjbaQ

Sincerely,

Iffa

[Iffa_Intel]

Greetings,

Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question.

Sincerely,

Iffa