aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/predictor
Commit message (Collapse)AuthorAge
* Update BUILD files so that Estimator code depends on TF thru tf_no_contrib.Gravatar Michael Case2018-09-25
| | | | PiperOrigin-RevId: 214541221
* Remove tf.contrib.get_signature_def_by_key. This can be replaced by ↵Gravatar A. Unique TensorFlower2018-09-13
| | | | | | meta_graph_def.signature_def[signature_def_key] PiperOrigin-RevId: 212873314
* Fix estimator dependencies in contrib/.Gravatar Michael Case2018-08-09
| | | | PiperOrigin-RevId: 208126204
* Split checkpoint management utility functions out of saver.pyGravatar Allen Lavoie2018-08-02
| | | | | | Pure refactor, in preparation for adding a higher level checkpoint management utility. This utility will also need to work with the Checkpoint proto, and globbing it on to saver.py seems dirty. PiperOrigin-RevId: 207179646
* Add `{input, output}_names` to `predictor.from_saved_model`. This exposes ↵Gravatar A. Unique TensorFlower2018-08-01
| | | | | | existing functionality via factory methods. PiperOrigin-RevId: 206934272
* Support session config in tf.contrib.predictor (#19542)Gravatar Lukas Geiger2018-06-03
| | | | | | | | | | * Support session config in tf.contrib.predictor This PR allows users to supply a custom session config uses by the predictor. This can be essential for some GPU setups in order to play nicely with other processes running on the same GPU. * Test passing session config to tf.contrib.predictor
* Remove all_opensource_files. It's not needed any more.Gravatar Martin Wicke2018-03-28
| | | | PiperOrigin-RevId: 190878279
* Merge changes from github.Gravatar Jacques Pienaar2018-03-21
| | | | PiperOrigin-RevId: 189945839
* Automated g4 rollback of changelist 189231636Gravatar A. Unique TensorFlower2018-03-15
| | | | PiperOrigin-RevId: 189258641
* Merge changes from github.Gravatar Jacques Pienaar2018-03-15
| | | | PiperOrigin-RevId: 189231636
* Add C0330 bad-continuation check to pylint.Gravatar Yifei Feng2018-01-25
| | | | PiperOrigin-RevId: 183270896
* Merge changes from github.Gravatar Raghuraman Krishnamoorthi2018-01-03
| | | | PiperOrigin-RevId: 180746153
* Fix 'tags' parameter in predictor_factories.load_from_model.Gravatar A. Unique TensorFlower2017-12-13
| | | | | | | tags was incorrectly being mapped to inputs. Added basic unit tests. PiperOrigin-RevId: 178916192
* Fix nopip build tag typo in contrib/predictor:test_export_dir.Gravatar Michael Case2017-11-16
| | | | PiperOrigin-RevId: 176011110
* BUILD cleanup in contrib/...Gravatar A. Unique TensorFlower2017-10-30
| | | | PiperOrigin-RevId: 173889798
* Add expected keys to predictor exception if unexpected key detected.Gravatar A. Unique TensorFlower2017-10-18
| | | | PiperOrigin-RevId: 172634275
* Provide a public property to access model_fn of an Estimator. An example ↵Gravatar Mustafa Ispir2017-09-07
| | | | | | | | | | usage of it for extending model_fn is in tf.contrib.estimator.add_metrics function. Design decisions: * `params` is bound to the returned function: model_fn is made public to help framework developers to extend/compose estimators. We assumed that framework developer does not know much information about how one estimator/model_fn is constructed. We think `params` is part of model_fn construction. If a framework combines estimators, each estimator may be created with different params. Combining those params into single one may create conflicts. * model_fn property have a fixed signature instead of optional arguments. We made model_fn arguments as optional to simplify model_fn implementations. In this case users are not model_fn developers but model_fn users. If we kept the arguments optional, then framework developers should handle that optional logic for every model_fn usage. Also model_fn is called by Estimator. Estimator have all those arguments. So there is no need to make them optional. PiperOrigin-RevId: 167914400
* Further BUILD cleanup in tensorflow/contrib/...Gravatar A. Unique TensorFlower2017-07-19
| | | | PiperOrigin-RevId: 162561536
* Automated g4 rollback of changelist 161218103Gravatar A. Unique TensorFlower2017-07-12
| | | | PiperOrigin-RevId: 161671226
* Automated g4 rollback of changelist 161203536Gravatar Suharsh Sivakumar2017-07-07
| | | | PiperOrigin-RevId: 161218103
* Fix dependencies and import statements for predictor module.Gravatar A. Unique TensorFlower2017-07-07
| | | | PiperOrigin-RevId: 161203536
* Introduce Predictor, an interface for efficient, repeated inference.Gravatar A. Unique TensorFlower2017-06-15
PiperOrigin-RevId: 159141010
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-16 18:29:25 +0000