tflite2onnx 0.1.0

Converting TensorFlow Lite models to ONNX models

Convert TensorFlow Lite models to ONNX

This tflite2onnx package converts TensorFlow Lite (TFLite) models (*.tflite) to ONNX models (*.onnx).

This project is under early stage of development, contributions are welcome at GitHub.

Installation

Simply install via pip. This package requires tflite, onnx and numpy which should be installed automatically.

pip install tflite2onnx

Usage

It's simple as below. Command line tool will be provided soon.

import tflite2onnx

tflite_path = '/path/to/original/tflite/model'
onnx_path = '/path/to/save/converted/onnx/model'

tflite2onnx.convert(tflite_path, onnx_path)

If you'd like to convert a TensorFlow model (*.pb) to ONNX, you may try tf2onnx. Or, you can firstly convert TensorFlow model (*.pb) to TensorFlow Lite models (*.tflite), and then convert to ONNX using this tool.

Resources

• PyPI page.
• GitHub page.
• TensorFlow Lite converter.

License

Apache License Version 2.0.