Convert TensorFlow Lite models to IR

[Reagan]

Hi,

  Is there a plan to support conversion of TensorFlow Lite models to IR using Model Optimizer?

Regards

 Reagan

[Munesh_Intel]

Hi Reagan,

Thanks for reaching out to us.

Model Optimizer supports several frozen quantized topologies hosted on the TensorFlow Lite site. The frozen model file (.pb file) is required for conversion to IR format by Model Optimizer.

More information is available in the section “Supported Frozen Quantized Topologies” at the following link:

https://docs.openvinotoolkit.org/2021.3/openvino_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_TensorFlow.html#supported_topologies

As of now, Model Optimizer doesn’t support native conversion from .tflite file to IR format. However, we are unable to comment on future support or any planned enhancements, as they are subject to changes.

Regards,

Munesh

[Munesh_Intel]

Hi Reagan,

This thread will no longer be monitored since we have provided explanations. If you need any additional information from Intel, please submit a new question.

Regards,

Munesh