| Commit message (Collapse) | Author | Age |
|
|
|
| |
Change: 117011137
|
|
|
|
| |
Change: 116908297
|
|
|
|
| |
Change: 116906391
|
|
|
|
|
|
|
|
|
|
|
|
| |
This includes a gRPC server (grpc_tensorflow_server) that can serve as both
the master of a distributed TensorFlow computation, and an individual worker
in the computation. The GrpcSession class is included to allow client programs
(including Python clients) to interact with a server.
See tensorflow/core/distributed_runtime/README.md for usage instructions.
This change partially addresses issue #23.
Change: 115634191
|
|
|
|
|
| |
Remove some unnecessary devDependencies, e.g. accessibility_developer_tools and chai.
Change: 115613184
|
|
|
|
| |
Change: 115515678
|
|
|
|
| |
Change: 115280348
|
|
|
|
| |
Change: 115268843
|
|
|
|
|
| |
floats in TensorFlow. The code was tested on Tegra x1.
Change: 115253733
|
|
|
|
|
|
| |
used by tf-regex-group, tf-categorizer and tf-collapsable-pane components. Also make the version number exact instead of ^1.0.0, and make it match the version inside google.
Change: 114987970
|
|
|
|
| |
Change: 114806277
|
|
|
|
| |
Change: 114796968
|
|
|
|
|
|
| |
also fixes the problem where users are asked which version of polymer to install when they run `bower install`.
Change: 114774859
|
|
|
|
| |
Change: 114585944
|
|
|
|
|
|
|
| |
Eigen::CompleteOrthogonalDecomposition (COD), which I recently contributed to Eigen in https://bitbucket.org/eigen/eigen/pull-requests/163/implement-complete-orthogonal/diff
The advantage of COD over column pivoted QR is that it is able to compute the minimum-norm solution when the matrix is rank-deficient, which is usually the desired behavior and makes it consistent with the fast path.
Change: 114483303
|
|
|
|
| |
Change: 114243879
|
|
|
|
|
| |
stability improvements
Change: 113791782
|
|
|
|
| |
Change: 113773458
|
|
|
|
| |
Change: 113371678
|
|
|
|
| |
Change: 113114631
|
|
|
|
| |
Change: 112920860
|
|
|
|
|
|
| |
* Move some checks out of inner loops
* Split the mapper in 2: a base mapper, and a sub-mapper. This reduces the number of variables that are contained in the base mapper and helps reduce register spills
Change: 112809881
|
|
|
|
| |
Change: 112699191
|
|
|
|
|
| |
matrix-vector products and provides various fixes for CUDA.
Change: 112298067
|
|
|
|
|
| |
well the crash reported in https://github.com/tensorflow/tensorflow/issues/713.
Change: 111874622
|
|
|
|
|
|
| |
reductions on GPU"
Change: 111739152
|
|
|
|
| |
Change: 111713364
|
| |
|
|
|
|
|
|
| |
imported.
Change: 110842260
|
|
|
|
| |
Change: 110406666
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change 109418220
Update WORKSPACE to use gmock.BUILD from google/protobuf instead of a duplicate.
Update google/protobuf's commit hash to include damieng@'s commit.
Change 109417314
TensorFlow: add .gitignore to ignore some in-tree modified files.
Change 109400051
Optionally build full TensorFlow for Android.
1. --define ANDROID_TYPES=__ANDROID_TYPES_FULL__ to register ops
for all types, not just float. Today this increases codesize
by ~700K when compiled for ARM, though only for clients who
request full type support.
2. Add more ops to android_extended_ops, sufficient to train on
the linear regression baseball codelab.
Change 109388118
Fix the option changed in templatize. Oops.
Change 109382553
Allows setting a function name in an op's attr in the py frontend.
Change 109380896
Remove assert_same_graph in favor of op_scope. Change the latter to handle tensor-like objects such as SparseTensor, IndexedSlices, and Variable.
Base CL: 109418322
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change 109195845
Fix TensorFlow for build against Bazel 0.1.2rc2
Two things are currently broken with TensorFlow and Bazel 0.1.2:
- Bazel now use sandboxing by default on Linux and have fixed it for cc_* rules.
Undeclared headers are not mounted in the sandbox which make several cc_* rules
fails.
- Bazel now enforce strict header checking and some target were missing
headers even though the headers were mounted in the sandbox. This change
adds a "strict_headers" target that globs every headers of the core
library and add it to the `tf_cc_tests` targets.
Change 109162708
Fix various website issues
- Fix headline in os_setup.md
- Fix #anchor links
Change 109162129
Fix numbers in mnist tutorial, fixes #362
Change 109158967
Fix typo in word2vec tutorial, fixes #347
Change 109151855
Fix tile and its gradient for scalars on GPUs
Eigen doesn't handle scalars on GPUs in all cases. Fortunately, both
tile and its gradient are the identity for scalars, so we can just copy
the input to the output.
Fixes https://github.com/tensorflow/tensorflow/issues/391.
Change 109140763
Support int32 and int64 in tf.random_uniform
This requires a new RandomUniformInt op on the C++ side since the op needs
to know minval and maxval.
Fixes https://github.com/tensorflow/tensorflow/issues/364.
Change 109140738
Fix spacing in docs.
Change 109140030
Fix content nav to not hide the bottom 100 or so px.
Change 109139967
Add license files to TensorBoard files, fix mnist_with_summaries test
Change 109138333
Fix typos in docstring
Change 109138098
Fix some missing resources in the website.
Fixes #366.
Change 109123771
Make sparse_to_dense's default_value default to 0
Nearly all uses of sparse_to_dense use 0 as the default. The
same goes for sparse_tensor_to_dense.
Base CL: 109198336
|
|
TensorFlow is an open source software library for numerical computation
using data flow graphs.
Base CL: 107276108
|