| Commit message (Collapse) | Author | Age |
|
|
|
| |
Change: 152837661
|
|
|
|
| |
Change: 152508170
|
|
|
|
| |
Change: 152200430
|
|
|
|
| |
Change: 151046259
|
|
|
|
| |
Change: 149800363
|
|
|
|
| |
Change: 148954491
|
|
|
|
| |
Change: 147897309
|
|
|
|
| |
Change: 146918929
|
|
|
|
|
|
| |
Output, etc. now that all uses have migrated to using the tensorflow
namespace.
Change: 145119259
|
|
|
|
| |
Change: 143412147
|
|
|
|
|
|
| |
from the Python API.
Change: 142598006
|
|
|
|
|
|
|
|
|
| |
less accurate one. This enables, but does not yet provide, significant speed improvements for image models bottlenecked on image-preprocessing.
RELNOTES: tf.image.decode_jpeg by default uses the faster DCT method, sacrificing
a little fidelity for improved speed. One can revert to the old
behavior by specifying the attribute dct_method='INTEGER_ACCURATE'.
Change: 142517485
|
|
|
|
| |
Change: 142324147
|
|
|
|
|
|
| |
for both real and complex tensors.
Change: 142316271
|
|
|
|
| |
Change: 142215409
|
|
|
|
| |
Change: 142074581
|
|
|
|
| |
Change: 142071642
|
|
|
|
| |
Change: 141982462
|
|
|
|
| |
Change: 141951932
|
|
|
|
| |
Change: 141521124
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
| |
Removing shape property from SparseTensor.
Change: 141490813
|
|
|
|
| |
Change: 140632089
|
|
|
|
| |
Change: 140489119
|
|
|
|
|
|
|
|
|
| |
- 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
|
|
|
|
|
| |
all real shape functions are now registered via the C++ API.
Change: 139803467
|
|
|
|
| |
Change: 139651838
|
|
|
|
| |
Change: 139377074
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
|
| |
- Swap the alias direction and fix tests broken by the swap
- Export Output in tensorflow.__all__
Change: 138583261
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
| |
Change: 138143557
|
|
|
|
| |
Change: 136750267
|
|
|
|
|
| |
topological neighborhood of a device.
Change: 136663586
|
|
|
|
| |
Change: 135698415
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
| |
TensorBoard.
Change: 133917299
|
|
|
|
| |
Change: 132733397
|
|
|
|
|
|
|
| |
about when it should be used.
Keep the old name around for temporary backwards compatibility.
Change: 132700646
|
|
|
|
|
| |
node_def.proto instead.
Change: 131290995
|
|
|
|
| |
Change: 131229727
|
|
|
|
|
|
|
| |
type lists.
The current special case behavior is preserved only for functions and their gradients.
Change: 130100547
|
|
|
|
| |
Change: 130070042
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
| |
Change: 128958134
|
|
|
|
|
|
| |
TESTED:
- passed opensource_build: http://ci.tensorflow.org/job/tensorflow-cl-presubmit-multijob/2780/
Change: 127585603
|
|
|
|
| |
Change: 127385333
|
|
|
|
| |
Change: 127101926
|
|
|
|
| |
Change: 126335170
|