aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/seq2seq/python/kernel_tests
Commit message (Expand)AuthorAge
* Compute `axes` and `free` statically during graph creation.Gravatar A. Unique TensorFlower2018-09-17
* Support coverage penalty for beam search decoder (according to https://arxiv....Gravatar A. Unique TensorFlower2018-09-12
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-08-21
* Fix masking of beam ids in gather_tree_from_arrayGravatar Guillaume Klein2018-07-10
* Merge commit for internal changesGravatar Yifei Feng2018-04-19
|\
| * Various lint fixes to TensorFlow detected after GitHub merge.Gravatar Michael Case2018-04-18
* | Merge commit for internal changesGravatar Yifei Feng2018-04-17
|\|
| * Use fixed sized tensor arrays and max loop iterations in dynamic_decode if th...Gravatar A. Unique TensorFlower2018-04-16
* | Add test case for Bahdanau attention when normalized=True and dtype = float16/32Gravatar Yong Tang2018-04-16
* | Support passing layer instances to produce attentional hidden states (#14974)Gravatar Guillaume Klein2018-04-15
| * Merge changes from github.Gravatar Michael Case2018-04-10
* | Fix issue with Luong attention when scale=True and dtype of tf.float16/tf.flo...Gravatar Yong Tang2018-03-30
|/
* Makes tf.gather not silently snapshot resource variables.Gravatar Alexandre Passos2018-03-26
* Merge changes from github.Gravatar Jacques Pienaar2018-03-21
* Fixed linter errors.Gravatar Jianwei Xie2018-01-24
* Merge changes from github.Gravatar Jianwei Xie2018-01-24
* Propagate static shape info in AttentionWrapperState.clone() if possible.Gravatar Rui Zhao2018-01-10
* Does not use constants for zeros/ones.Gravatar Alexandre Passos2018-01-05
* Merge changes from github.Gravatar Patrick Nguyen2017-12-28
* [tf.contrib.seq2seq] Modify AttentionMechanisms to propagate state.Gravatar Eugene Brevdo2017-12-15
* Fix tf.contrib.seq2seq._monotonic_probability_fn to use a hard sigmoid when m...Gravatar Colin Raffel2017-11-10
* [tf.contrib.seq2seq] Reserve -1s in GatherTree for error states.Gravatar Eugene Brevdo2017-10-19
* [tf.contrib.seq2seq] Bugfixes to BeamSearchDecoder and GatherTree.Gravatar Eugene Brevdo2017-10-17
* [tf.contrib.seq2seq] Some light cleanup in beam search decoder code.Gravatar Eugene Brevdo2017-10-16
* Register int types for one_hot GPU ops.Gravatar Rui Zhao2017-09-22
* Merge changes from github.Gravatar A. Unique TensorFlower2017-08-15
* Allow the next input in ScheduledOutputTrainingHelper to be computed by any c...Gravatar Adam Roberts2017-08-11
* Allow specification of sample_id_shape and sample_id_dtype in seq2seq.BasicDe...Gravatar Adam Roberts2017-08-11
* Add multi-head attention capabilities to AttentionWrapper via the specificati...Gravatar Adam Roberts2017-07-19
* Merge changes from github.Gravatar A. Unique TensorFlower2017-06-27
* Automated g4 rollback of changelist 160182040Gravatar A. Unique TensorFlower2017-06-26
* Merge changes from github.Gravatar A. Unique TensorFlower2017-06-26
* [tf contrib seq2seq] Add monotonic attention mechanismsGravatar A. Unique TensorFlower2017-06-21
* Make ScheduledOutputTrainingHelper.sample() returns dtypes.int32 sample_ids.Gravatar A. Unique TensorFlower2017-06-20
* [tf contrib seq2seq] Expand tile_batch to handle nested structures.Gravatar Eugene Brevdo2017-06-02
* [tf contrib seq2seq] Improve BeamSearchDecoder's ability to handle unknown sh...Gravatar Eugene Brevdo2017-05-30
* Add SampleEmbeddingHelper to do sampling at inference timeGravatar A. Unique TensorFlower2017-05-30
* [tf contrib seq2seq] Bugfixes to BeamSearchDecoderGravatar Eugene Brevdo2017-05-29
* Fix attention score padding. Use -inf as default instead of 0 for softmax.Gravatar A. Unique TensorFlower2017-05-26
* Move many of the "core" RNNCells and rnn functions back to TF core.Gravatar Eugene Brevdo2017-05-22
* Update RNN helpers to be able to handle dynamic state sizes.Gravatar Eugene Brevdo2017-05-15
* [tf contrib seq2seq] Updates to AttentionMechanism APIGravatar Eugene Brevdo2017-05-12
* [tf contrib seq2seq] Update BeamSearchDecoder + AttentionWrapper API:Gravatar Eugene Brevdo2017-04-26
* Add an option to allow use context vector as attention in Attention Wrapper.Gravatar A. Unique TensorFlower2017-04-24
* [tf contrib seq2seq] Changes to dynamic decodingGravatar Eugene Brevdo2017-04-20
* BeamSearchDecoder:Gravatar A. Unique TensorFlower2017-04-19
* [TF contrib seq2seq] Beam search tree decoder op "GatherTree".Gravatar Eugene Brevdo2017-04-18
* [tf contrib seq2seq] Change 'attention_history' to 'alignment_history'Gravatar Eugene Brevdo2017-04-10
* Add a public `clone()` method to AttentionCellWrapper.Gravatar Eugene Brevdo2017-03-31
* Avoid sampling overhead when no sampled inputs are to be used.Gravatar Adam Roberts2017-03-30
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-27 13:32:41 +0000