Mastering Computer Vision with TensorFlow 2.x

Converting from a TensorFlow model developed using Keras

This section describes how to convert a TensorFlow model into OpenVINO IR format. For more information, please refer to the following link: https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_TensorFlow.html.

The steps can be summarized as follows:

1. Configure the model optimizer. As described at the very beginning of this chapter, any model that's being deployed for edge devices must be optimized, which involves removing any unnecessary components without sacrificing accuracy. The following code performs this task globally:

<INSTALL_DIR>/deployment_tools/model_optimizer/install_prerequisites directory and run: $install_prerequisites.sh ...