diff options
| author |  Shanqing Cai <cais@google.com> | 2018-08-24 18:49:19 -0700 |
|---|---|---|
| committer |  TensorFlower Gardener <gardener@tensorflow.org> | 2018-08-24 18:53:20 -0700 |
| commit | 5afcc0fa269f9b41f1aecd029f4e3ebea35a420b (patch) | |
| tree | c1b0bd98d57e90649fa69e1db0b2de5e5c98c028 /tensorflow/docs_src | |
| parent | d936338f32b826cc6b7ab835efed427d78741f81 (diff) | |
Update TensorFlow.js roadmap
* remove old (completed) items
* add new items
PiperOrigin-RevId: 210190738
Diffstat (limited to 'tensorflow/docs_src')
| -rw-r--r-- | tensorflow/docs_src/community/roadmap.md | 8 |
1 files changed, 5 insertions, 3 deletions
diff --git a/tensorflow/docs_src/community/roadmap.md b/tensorflow/docs_src/community/roadmap.md index 0463ca05fe..d11b6ed467 100644 --- a/tensorflow/docs_src/community/roadmap.md +++ b/tensorflow/docs_src/community/roadmap.md @@ -58,10 +58,12 @@ across image recognition, speech, object detection, and * Increase support for devices beyond Android and iOS (eg. RPi, Cortex-M) #### TensorFlow.js: -* Release package for Node.js bindings to the TensorFlow C API through the TensorFlow.js backend interface -* Expand support for importing TensorFlow SavedModels and Keras models into browser with unified APIs supporting retraining in browser -* Improve Layers API and allow model exporting/saving +* Continue to expand support for importing TensorFlow SavedModels and Keras models into browser with unified APIs supporting retraining in browser +* Improve inference and training performance in both browser and Node.js environments +* Widen the collection of pre-built models in [tfjs-models](https://github.com/tensorflow/tfjs-models), + including but not limited to audio- and speech-oriented models * Release tfjs-data API for efficient data input pipelines +* Integration with [TF-Hub](https://www.tensorflow.org/hub/) #### TensorFlow with Swift: * Establish open source project including documentation, open design, and code availability. |