experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Adds clarity on supported formats for TFLiteConverter 2.0.

Browse Source 
PiperOrigin-RevId: 238324267
This commit is contained in:
Nupur Garg 2019-03-13 15:36:43 -07:00 committed by TensorFlower Gardener
parent dd7daa8cc7
commit 278bddcee2
1 changed files with 18 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
tensorflow/lite/g3doc/r2/convert/python_api.md
View File

@ -150,7 +150,20 @@ output_data = interpreter.get_tensor(output_details[0]['index'])
print(output_data)
```
## Summary of changes in TFLiteConverter in 1.X and 2.0 <a name="differences"></a>
## Summary of changes in `TFLiteConverter` between 1.X and 2.0 <a name="differences"></a>
The following section summarizes the changes in `TFLiteConverter` from 1.X to
2.0. If any of the changes raise concerns, please file a
[GitHub issue](https://github.com/tensorflow/tensorflow/issues).
### Supported formats
`TFLiteConverter` in 2.0 supports SavedModels and Keras model files generated in
both 1.X and 2.0. However, the conversion process no longer supports frozen
`GraphDefs` generated in 1.X. Users who want to convert frozen `GraphDefs` to
TensorFlow Lite should use `tensorflow.compat.v1`.
### Quantization-aware training
The following attributes and methods associated with
[quantization-aware training](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/quantize)
@ -173,6 +186,8 @@ removed in the 2.0 API until the new quantization API is launched. Users who
want to convert models generated by the rewriter function can use
`tensorflow.compat.v1`.
### Changes to attributes
The `target_ops` attribute has become an attribute of `TargetSpec` and renamed
to `supported_ops` in line with future additions to the optimization framework.
@ -190,15 +205,14 @@ Additionally, the following attributes have been removed:
* `dump_graphviz_dir`
* `dump_graphviz_video`
### Deprecated APIs
The following methods that were previously deprecated in 1.X will no longer be
exported in 2.0:
* `lite.toco_convert`
* `lite.TocoConverter`
If any of the changes raise concerns, please file a
[GitHub issue](https://github.com/tensorflow/tensorflow/issues).
## Installing TensorFlow <a name="versioning"></a>
### Installing the TensorFlow 2.0 nightly <a name="2.0-nightly"></a>