README.md
1 # TensorFlow Lite Converter 2 3 The TensorFlow Lite Converter converts TensorFlow graphs into 4 TensorFlow Lite graphs. There are additional usages that are also detailed in 5 the usage documentation. 6 7 ## Usage documentation 8 9 Usage information is given in these documents: 10 11 * [Command-line glossary](../g3doc/convert/cmdline_reference.md) 12 * [Command-line examples](../g3doc/convert/cmdline_examples.md) 13 * [Python API examples](../g3doc/convert/python_api.md) 14 15 ## Where the converter fits in the TensorFlow landscape 16 17 Once an application developer has a trained TensorFlow model, the TensorFlow 18 Lite Converter will accept 19 that model and generate a TensorFlow Lite 20 [FlatBuffer](https://google.github.io/flatbuffers/) file. The converter currently supports 21 [SavedModels](https://www.tensorflow.org/guide/saved_model#using_savedmodel_with_estimators), 22 frozen graphs (models generated via 23 [freeze_graph.py](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/tools/freeze_graph.py)), 24 and `tf.Keras` model files. The TensorFlow Lite FlatBuffer file can be shipped 25 to client devices, generally mobile devices, where the TensorFlow Lite 26 interpreter handles them on-device. This flow is represented in the diagram 27 below. 28 29  30
