Name | Date | Size | |
---|---|---|---|
.. | 10-Aug-2018 | 4 KiB | |
allocate_transient_arrays.cc | 10-Aug-2018 | 11.7 KiB | |
allocate_transient_arrays.h | 10-Aug-2018 | 1.9 KiB | |
args.h | 10-Aug-2018 | 8.1 KiB | |
BUILD | 10-Aug-2018 | 11.5 KiB | |
dump_graphviz.cc | 10-Aug-2018 | 13.8 KiB | |
dump_graphviz.h | 10-Aug-2018 | 1,013 | |
export_tensorflow.cc | 10-Aug-2018 | 78.5 KiB | |
export_tensorflow.h | 10-Aug-2018 | 1.1 KiB | |
format_port.h | 10-Aug-2018 | 2.9 KiB | |
g3doc/ | 10-Aug-2018 | 4 KiB | |
graph_transformations/ | 10-Aug-2018 | 4 KiB | |
import_tensorflow.cc | 10-Aug-2018 | 78.5 KiB | |
import_tensorflow.h | 10-Aug-2018 | 1.6 KiB | |
model.h | 10-Aug-2018 | 53.8 KiB | |
model_cmdline_flags.cc | 10-Aug-2018 | 18.2 KiB | |
model_cmdline_flags.h | 10-Aug-2018 | 1.7 KiB | |
model_flags.proto | 10-Aug-2018 | 7.3 KiB | |
python/ | 10-Aug-2018 | 4 KiB | |
README.md | 10-Aug-2018 | 1 KiB | |
runtime/ | 10-Aug-2018 | 4 KiB | |
tensorflow_graph_matching/ | 10-Aug-2018 | 4 KiB | |
tensorflow_util.cc | 10-Aug-2018 | 7.6 KiB | |
tensorflow_util.h | 10-Aug-2018 | 1.2 KiB | |
tflite/ | 10-Aug-2018 | 4 KiB | |
toco.cc | 10-Aug-2018 | 4.7 KiB | |
toco_cmdline_flags.cc | 10-Aug-2018 | 10.8 KiB | |
toco_cmdline_flags.h | 10-Aug-2018 | 1.6 KiB | |
toco_flags.proto | 10-Aug-2018 | 6.3 KiB | |
toco_graphviz_dump_options.cc | 10-Aug-2018 | 895 | |
toco_graphviz_dump_options.h | 10-Aug-2018 | 1.2 KiB | |
toco_port.cc | 10-Aug-2018 | 5.9 KiB | |
toco_port.h | 10-Aug-2018 | 3.1 KiB | |
toco_port_test.cc | 10-Aug-2018 | 1.8 KiB | |
toco_tooling.cc | 10-Aug-2018 | 12.6 KiB | |
toco_tooling.h | 10-Aug-2018 | 1.9 KiB | |
toco_types.h | 10-Aug-2018 | 1.4 KiB | |
tooling_util.cc | 10-Aug-2018 | 64.5 KiB | |
tooling_util.h | 10-Aug-2018 | 13 KiB | |
tooling_util_test.cc | 10-Aug-2018 | 3.7 KiB | |
types.proto | 10-Aug-2018 | 1,003 |
README.md
1# The TensorFlow Lite Optimizing Converter 2 3The TensorFlow Lite Optimizing Converter's most typical use is converting from the TensorFlow GraphDef to the TensorFlow Lite 4format, but it supports much more than that. 5 6## Usage documentation 7 8Usage information is given in these documents: 9 10* [Command-line examples](g3doc/cmdline_examples.md) 11* [Command-line reference](g3doc/cmdline_reference.md) 12* [Python API](g3doc/python_api.md) 13 14## Design documentation 15 16Coming soon! 17 18## Where the converter fits in the TensorFlow landscape 19 20In the typical case, an application developer is using TensorFlow to design and 21train models, then uses TensorFlow's freeze_graph.py to generate a frozen 22inference graph, then uses the converter to convert that into a TensorFlow Lite flatbuffer file, 23then ships that file to client devices where the TensorFlow Lite interpreter handles them 24on-device. This is represented in the following diagram: 25 26![drawing](https://storage.googleapis.com/download.tensorflow.org/example_images/tensorflow_landscape.svg) 27