Well... as the title suggests Creates a MobileNetV2 model instance and saves it as a Keras .h5 file
Freezes the previously created model, optimises it with TensorRT and saves the optimised model engine
Loads optimised TensorRT model engine and tests it
Text file which contains model input and output layer info