aboutsummaryrefslogtreecommitdiffhomepage
path: root/RELEASE.md
Commit message (Collapse)AuthorAge
...
* Automated rollback of change 142316271Gravatar Andrew Selle2016-12-16
| | | | Change: 142324147
* Remove tf.complex_abs from the TensorFlow Python API. tf.abs should be used ↵Gravatar A. Unique TensorFlower2016-12-16
| | | | | | for both real and complex tensors. Change: 142316271
* Added release note for tf.pack --> tf.stack changes.Gravatar A. Unique TensorFlower2016-12-15
| | | | Change: 142215409
* Merge changes from github.Gravatar Martin Wicke2016-12-14
| | | | Change: 142074581
* Final breaking change, removing .shape from SparseTensor.Gravatar Martin Wicke2016-12-14
| | | | Change: 142071642
* Added a release note about tf.concat deprecation.Gravatar A. Unique TensorFlower2016-12-13
| | | | Change: 141982462
* Release notes updateGravatar Andrew Selle2016-12-13
| | | | Change: 141951932
* Automated rollback of change 141490813Gravatar A. Unique TensorFlower2016-12-08
| | | | Change: 141521124
* Merge changes from github.Gravatar Andrew Harp2016-12-08
| | | | | | | | Additionally: - change single quotes to double quotes to make path rewriting easier - guard windows lib reference with PLATFORM_WINDOWS - fixed failing kmeans test Change: 141515942
* Final breaking change of SparseTensor.shape -> SparseTensor.dense_shape rename.Gravatar Eugene Brevdo2016-12-08
| | | | | Removing shape property from SparseTensor. Change: 141490813
* Deprecate tf.select since it is getting replaced by tf.where.Gravatar A. Unique TensorFlower2016-11-30
| | | | Change: 140632089
* Update release.md to reflect breaking changes to RNN variable names.Gravatar Eugene Brevdo2016-11-29
| | | | Change: 140489119
* Change division in TensorFlow to flooring semantics.Gravatar Andrew Selle2016-11-22
| | | | | | | | | - tf.div changes to new behavior, but it will be deprecated - tf.divide is currently a synonym for tf.div but will remain - tf.mod changes to new behavior, but it will be deprecated, you can use % or tf.floormod in the future. - the op FloorDiv now is extended to work on reals Change: 139922734
* Remove tf.RegisterShape from public API. No longer needed sinceGravatar Vijay Vasudevan2016-11-21
| | | | | all real shape functions are now registered via the C++ API. Change: 139803467
* Automated rollback of change 138583261Gravatar Jonathan Hseu2016-11-18
| | | | Change: 139651838
* Rename TF_Port to TF_Output and add a TF_Input type.Gravatar Jonathan Hseu2016-11-16
| | | | Change: 139377074
* C API: Rename TF_SessionWithGraph to TF_Session.Gravatar Asim Shankar2016-11-09
| | | | | | | | | This makes the preferred API for graph construction and execution have the simpler, more sensible names (TF_Session, TF_NewSession etc.) and the older one which we hope to eventually remove have the more awkward ones (TF_DeprecatedSession, TF_NewDeprecatedSession etc.) Change: 138641615
* Rename tf.Tensor to tf.OutputGravatar Jonathan Hseu2016-11-08
| | | | | | - Swap the alias direction and fix tests broken by the swap - Export Output in tensorflow.__all__ Change: 138583261
* C API: Rename TF_Session to TF_DeprecatedSession.Gravatar Asim Shankar2016-11-08
| | | | | | | | | | | | This is one step towards having a stable C API by the time we get to TensorFlow 1.0. A follow-up step will involve renaming TF_SessionWithGraph to TF_Session. We want to encourage all client languages to use TF_SessionRun, TF_GraphImportGraphDef etc. (instead of TF_Run and TF_ExtendGraph), hence the choice of names. Ideally, all client languages will use these functions, but it is likely that Python will continue to use TF_DeprecatedSession for a while. Change: 138454433
* Change FileExists to return tensorflow::Status.Gravatar Jonathan Hseu2016-11-04
| | | | | | | | | | | Also done separately by @llhe at github.com/tensorflow/tensorflow/pull/5370. We needed to do this change internally to fix all callers. Motivation: The existing FileExists interface doesn't allow callers to distinguish between file not found vs. filesystem errors. Semantics changes: - gfile.Exists in Python now throws an exception for filesystem errors. It continues to return true/false if it can accurately determine whether a file exists. - RecursivelyCreateDir now returns errors for filesystem errors when calling FileExists. Change: 138224013
* Merge changes from github.Gravatar Vijay Vasudevan2016-11-03
| | | | Change: 138143557
* Merge changes from github.Gravatar Patrick Nguyen2016-10-20
| | | | Change: 136750267
* Replace enum BusAdjacency with protobuf DeviceLocality for describing theGravatar A. Unique TensorFlower2016-10-19
| | | | | topological neighborhood of a device. Change: 136663586
* Merge changes from github.Gravatar A. Unique TensorFlower2016-10-10
| | | | Change: 135698415
* Make regexp.h internal, only visible to the new tfprof contrib lib.Gravatar Vijay Vasudevan2016-09-23
| | | | | | | | | | | | | RE2 is not as portable as we would like, so we want to try to discourage the use of it in core libraries. After having removed it from all core uses, the only remaining use is in tfprof, so we move its visibility to only be to tfprof, which is an external tool. tfprof appears to fundamentally rely on regexp as a user interface, so it cannot be easily removed. However, perhaps one day tfprof will move into its own repo and then we can drop RE2 from our builds entirely. Change: 134120532
* Remove dot.cc utility, not really needed since we haveGravatar Vijay Vasudevan2016-09-21
| | | | | TensorBoard. Change: 133917299
* Merge changes from github.Gravatar Dan Mané2016-09-09
| | | | Change: 132733397
* Rename NoGradient -> NotDifferentiable, to make it clearGravatar Vijay Vasudevan2016-09-09
| | | | | | | about when it should be used. Keep the old name around for temporary backwards compatibility. Change: 132700646
* graph.proto no longer needs to export NodeDef, clients should useGravatar A. Unique TensorFlower2016-08-25
| | | | | node_def.proto instead. Change: 131290995
* Adds _PartitionInfo to variable initializer signatureGravatar Wei Ho2016-08-24
| | | | Change: 131229727
* Remove special handling of host-memory/device-memory for int32 arguments in ↵Gravatar A. Unique TensorFlower2016-08-12
| | | | | | | type lists. The current special case behavior is preserved only for functions and their gradients. Change: 130100547
* Document breaking change to several RNNCells.Gravatar Eugene Brevdo2016-08-11
| | | | Change: 130070042
* Change DeviceFactory functions that create devices to propagateGravatar Vijay Vasudevan2016-08-08
| | | | | | | | | | | | | Statuses, so that failures to initialize devices don't crash the program. Changes swig for device_lib to be a lot simpler, thanks to mrry@ and keveman@'s help. Change allocation of eigen scratch memory to go through the allocator. Re-enable test for local devices now that python3 issue is fixed. Change: 129678132
* Merge changes from github.Gravatar A. Unique TensorFlower2016-07-31
| | | | Change: 128958134
* Improvements to the C++ graph building API.Gravatar Manjunath Kudlur2016-07-15
| | | | | | TESTED: - passed opensource_build: http://ci.tensorflow.org/job/tensorflow-cl-presubmit-multijob/2780/ Change: 127585603
* Include recent C graph-construction API in release notes.Gravatar A. Unique TensorFlower2016-07-13
| | | | Change: 127385333
* Merge changes from github.Gravatar Andrew Harp2016-07-11
| | | | Change: 127101926
* Merge changes from github.Gravatar A. Unique TensorFlower2016-06-30
| | | | Change: 126335170
* TensorFlow: Remove old Env:: APIs that don't pass unique_ptr.Gravatar Vijay Vasudevan2016-06-21
| | | | | | Also update RELEASE.md to forward port from 0.9 branch, and add this new 'breaking change to the API.' Change: 125465285
* Move CTC out of contrib and document.Gravatar Eugene Brevdo2016-06-15
| | | | Change: 125022295
* Make RNN api public.Gravatar Eugene Brevdo2016-06-07
| | | | Change: 124305799
* Update the tensorflow RELEASE.md to include information about runtime ↵Gravatar Nikhil Thorat2016-05-27
| | | | | | | statistics in the tensorboard graph visualizer. Change: 123427735
* Allow callers to provide a shift value in tf.nn.moments().Gravatar A. Unique TensorFlower2016-05-24
| | | | Change: 123095477
* Documented TensorBoard improvements that will make it into the 0.9.0 release.Gravatar Dan Mané2016-05-16
| | | | Change: 122460440
* Merge changes from github.Gravatar Illia Polosukhin2016-04-18
| | | | Change: 120185825
* Fix dependencies bugsGravatar Eugene Brevdo2016-03-11
| | | | Change: 116925769
* Merge changes from github.Gravatar Vijay Vasudevan2016-02-17
| | | | Change: 114882676
* Fix tf.test for PEP-8 and documentGravatar Geoffrey Irving2016-02-10
| | | | | | | | tf.test now has appropriate snake case function names (get_temp_dir and is_built_with_cuda) and has normal toplevel module documentation. Also fix a bug in make_all. Change: 114351269
* Generalize tf.image.random_crop to dimension-independent tf.random_cropGravatar Geoffrey Irving2016-02-08
| | | | | | | | | | | The C++ 3-D-only RandomCrop op is now deprecated at GraphDef version 8, replaced with a python tf.random_crop that works for any dimension. This will allow random_crop to be used for other purposes. Unfortunately, tf.image.random_crop took 2 sizes rather than 3 for 3-D tensors. The new tf.random_crop always takes n sizes for rank n tensors; pass 3 as the last element if you want to not crop a last dimension of size 3. Change: 114135451
* Breaking change in TF RNN python api: Return the final state instead of theGravatar Eugene Brevdo2016-01-27
| | | | | | | | | list of states when calling tf.nn.rnn() and tf.nn.state_saving_rnn() This is necessary for further cleanup of RNN state propagation code (currently dynamic RNN calculations when passing sequence_length do not return the proper final state, this is a necessary fix to make that fix efficient). Change: 113203893
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-10 20:00:31 +0000