| Commit message (Collapse) | Author | Age |
| |
|
|\
| |
| | |
1.7.0rc2 cherry-pick request: Fixes a bug in `tf.contrib.data.map_and_batch()` shape inference
|
|/
|
|
|
|
|
|
| |
Previously, it would incorrectly report that all batches have the same size, not accounting for the possibility of the last batch being partial.
Fixes #17720.
PiperOrigin-RevId: 189121488
|
|\
| |
| | |
Update documentation
|
| |\
| |/
|/| |
|
| |
| |
| |
| |
| |
| |
| | |
path (#17605) (#17751)
* MKL DNN: fix the TF1.6 speed issue by fixing MKL DNN LRN
* fixed typos in the doc for LrnRewrite
|
| |\
| |/
|/| |
|
|\ \
| | |
| | | |
1.7.0rc1 cherry-pick request: Fixes a race condition in function instantiation.
|
| |/
|/| |
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, if the same function was being concurrently instantiated
and released:
1. Thread one could begin to instantiate the function, determine
that it already existed in the runtime, then be preempted.
2. Thread two could release the handle on the function, causing it to
be freed and removed from the `FunctionLibraryRuntime::items_` map.
3. Thread one could then incorrectly assume that the function still
existed, and fail to find it in the `FunctionLibraryRuntime::items_`
map, causing a segfault when it attempted to increment the refcount
on an uninitialized object.
PiperOrigin-RevId: 188661500
|
|\
| |
| | |
Make tensorflow/python:framework_importer_test large
|
|/
|
| |
tensorflow/python:framework_importer_test sometime times out during release builds
|
| |
|
| |
|
| |
|
|\
| |
| | |
Disable tensorflow/contrib/learn:monitors_test for pip gpu
|
|/ |
|
|\
| |
| | |
Including the original path to find bazel.
|
|/ |
|
|\
| |
| | |
Fix pylint error in single_return.py
|
|/ |
|
|
|
| |
Hide os so we don't generate api_docs for it
|
| |
|
|\
| |
| | |
Disable checkpointable_utils_test
|
|/
|
| |
failed http://ci.tensorflow.org/view/Release/job/release-debian-cpu/99/consoleFull
|
|\
| |
| | |
Update version string to 1.7.0rc0 everywhere.
|
|/ |
|
|\
| |
| | |
Update tensorrt import exception.
|
| | |
|
|\ \
| | |
| | | |
Revert "Update external protobuf codebase version for Windows cmake b…
|
| | | |
|
| |/
|/| |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This fix adds the missing `#define OMPI_SKIP_MPICXX` in
`tensorflow/contrib/mpi/mpi_utils.h` so that it is consistent
with other usages of `mpi.h` includes. `OMPI_SKIP_MPICXX` skip
the MPI C++ bindings support.
This fix fixes 17388.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
|
|\ \
| | |
| | | |
Branch 188272354
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | | |
Current implementation modifies TfLiteNode::builtin_data every time when
a loaded graph is executed. The three masks in params will continually
flipping, and cause the op produce incorrect result every two executions.
|
| | |
| | |
| | |
| | | |
This is a workaround to fix build failure caused by packed git refs.
The tf.__git_version__ string will be "unknown" in this case.
|
| | |
| | |
| | | |
__unicode()__ was removed in Python 3 because all str are Unicode so this PR changes four calls to __unicode()__ into calls to [__six.text_type()__](http://six.readthedocs.io/#six.text_type).
|
| | | |
|
| | |
| | |
| | | |
Bring TrainingSpec and EvalSpec pydoc in line with pydoc of estimator.train() and evaluate()
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* Fix build issue with KafkaDataset
This fix tries to address the issue raised in 17210 where
error of `NotFoundError: Op type not registered 'KafkaDataset' in binary.`
returned from kafka ops. The issue was that the inclusion of kafka ops
was removed due to the conflict merge from the other PR. This fix
fixes the issue.
This fix fixes 17210.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Change `import readers.Dataset` to `import dataset_ops.Dataset`,
due to the changes in some other places.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix library dependency issues in bazel
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add dependency to bazel rules
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add license to lib and pip package
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Remove unneeded changes in bazel
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Address review feedback
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix sanity check
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add zlib dependency and include path
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add copts in bazel to address the discrepancy in clang and gcc
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* Fix cmake Dockerfile issue on Linux
When running cmake on Linux with (clean build with no cached docker images):
```
tensorflow/tools/ci_build/ci_build.sh CMAKE tensorflow/tools/ci_build/builds/cmake.sh
```
The following isse was encountered:
```
Step 11/13 : RUN pip install --upgrade termcolor
---> Running in 838167596eb6
Collecting termcolor
Downloading termcolor-1.1.0.tar.gz
......
......
......
error: invalid command 'bdist_wheel'
----------------------------------------
Failed building wheel for termcolor
```
This fix adds the missing `pip install wheel`
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update golang installation in cmake Dockerfile
This fix updates the golang installation in cmake Dockerfile.
Previously, `ppa:ubuntu-lxc/lxd-stable` was used but it has been
deprecated, see:
http://lxc-users.linuxcontainers.narkive.com/IlHLLHqN/lxd-official-ppa-deprecation
That caused the following error:
```
Step 13/14 : RUN add-apt-repository -y ppa:ubuntu-lxc/lxd-stable
---> Running in 09301ba43a33
Cannot add PPA: 'ppa:~ubuntu-lxc/ubuntu/lxd-stable'.
The team named '~ubuntu-lxc' has no PPA named 'ubuntu/lxd-stable'
Please choose from the following available PPAs:
* 'buildd-backports': linuxcontainers.org: buildd backports
* 'daily': linuxcontainers.org: development builds
......
......
```
This fix updates the golang installation and use backported xenial (16.04),
as was suggested in the link:
http://lxc-users.linuxcontainers.narkive.com/IlHLLHqN/lxd-official-ppa-deprecation
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
|
| |/
|/|
| |
| | |
This reverts commit 07bec47ba5db4c2f2e33ecb49f23253a371bfbbe.
|
| | |
|
|\ \
| | |
| | | |
Merge 1.6 back to master.
|
|\ \ \
| | | |
| | | | |
add error message when importing contrib.tensorrt without libnvinfer
|
| | | |
| | | |
| | | |
| | | | |
PiperOrigin-RevId: 188261279
|
| | | |
| | | |
| | | |
| | | | |
PiperOrigin-RevId: 188261273
|