问题
What is the difference between the .lite and the .tflite formats (TensorFlow formats)? And if there is no difference, why there are two of them?
In addition to the question, it seems that I can't upload my model with the .lite extension file to Firebase ML kit. What might be the reason for that?
回答1:
ML Developers first train a TensorFlow model, and then use TOCO to convert it to TensorFlow Lite model. When running the TOCO command, you can specify whatever output name for the converted Lite model. All TensorFlow Lite TOCO samples use .tflite extension; but .lite seems another popular extension people would like to choose.
So as long as it's a TensorFlow Lite FlatBuffer formatted model, TensorFlow Lite would be able to load / run the model regardless of the extension.
But unfortunately, ML Kit Console at this moment only takes files with .tflite extension. We can consider remove that enforcement. In the meantime, if you are sure it's a TensorFlow Lite model, simply rename the extension and upload it.
来源：https://stackoverflow.com/questions/52121533/what-is-the-difference-between-the-lite-and-the-tflite-formats