aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Add tests for wrong defaults.Gravatar Jianmin Chen2016-03-08
| | | | Change: 116595261
* Assume errors are rare in OP_REQUIRES*() macros.Gravatar Josh Levenberg2016-03-08
| | | | Change: 116594819
* Prototype of an in-process gRPC server for TensorFlow/Python.Gravatar Derek Murray2016-03-08
| | | | | | | | | Adds support for binding a TensorFlow server to any port, to support single-process testing. This interface is a work in progress. In particular, it supports launching a server, but the support for clean shutdown is incomplete. Change: 116593644
* Changed the cuda_crosstool_condition to check for a define of using_cuda.Gravatar Manjunath Kudlur2016-03-07
| | | | | | Checking for a specific crosstool_top directory doesn't work when TensorFlow is a sub-module for a different project. Change: 116592676
* TensorFlow: select op: add support for empty tensors: selectGravatar Vijay Vasudevan2016-03-07
| | | | | can propagate empty tensors. Change: 116588842
* Make RNN calls require a known input depth (via shape inference).Gravatar Eugene Brevdo2016-03-07
| | | | | | This is a necessary change to ensure that we can remove input_size from the RNNCell objects. Change: 116586469
* fix docstring for zero_fractionGravatar A. Unique TensorFlower2016-03-07
| | | | Change: 116586466
* Add comment about SparseMatMul safetyGravatar Geoffrey Irving2016-03-07
| | | | Change: 116584285
* Add a TODO to Tensor::SummarizeValueGravatar Geoffrey Irving2016-03-07
| | | | Change: 116583469
* Add support for hinge loss (SVMs) to the Stohastic Dual Coordinate Ascent ↵Gravatar A. Unique TensorFlower2016-03-07
| | | | | | (SDCA) optimizer. Change: 116582004
* TensorFlow: Return an error on racy inputs to listdiff_op that would otherwiseGravatar Vijay Vasudevan2016-03-07
| | | | | cause a crash. Change: 116574701
* TensorFlow: Make extract_glimpses support empty tensors as input.Gravatar Vijay Vasudevan2016-03-07
| | | | Change: 116574344
* First attempt at an optimized implementation of DepthwiseConv2D for CPU.Gravatar A. Unique TensorFlower2016-03-07
| | | | | | | | | | | | | | | | | | | | | // OLD Benchmark Time(ns) CPU(ns) Iterations ------------------------------------------------------------------- BM_ConvFloatDepthwiseFwdCPU1_conv0 247698841 247715520 100 667.6M items/s 32_112_112_3_8_128_3_3_1_2_cpu1 BM_ConvFloatDepthwiseFwdCPU1_conv1 662664406 662723089 100 665.5M items/s 32_112_112_64_1_128_3_3_1_2_cpu1 // NEW Benchmark Time(ns) CPU(ns) Iterations ------------------------------------------------------------------- BM_ConvFloatDepthwiseFwdCPU1_conv0 60316894 60215905 100 2.7G items/s 32_112_112_3_8_24_3_3_1_2_cpu1 BM_ConvFloatDepthwiseFwdCPU1_conv1 158600898 158571194 100 2.7G items/s 32_112_112_64_1_64_3_3_1_2_cpu1 // NEW 4-THREADS Benchmark Time(ns) CPU(ns) Iterations ------------------------------------------------------------------- BM_ConvFloatDepthwiseFwdCPU4_conv0 16703436 64535709 100 9.7G items/s 32_112_112_3_8_24_3_3_1_2_cpu4 BM_ConvFloatDepthwiseFwdCPU4_conv1 51874080 182896805 100 8.3G items/s 32_112_112_64_1_64_3_3_1_2_cpu4 Change: 116555067
* Enables all `image_ops` functions to accept NumPy arrays.Gravatar Derek Murray2016-03-07
| | | | | Fixes #1409. Change: 116554875
* Adding visitor support to DataByExample.Gravatar A. Unique TensorFlower2016-03-07
| | | | Change: 116478086
* TensorFlow: use GOOGLE_CUDA to prevent gpu library code from needingGravatar Vijay Vasudevan2016-03-05
| | | | | to be linked into CPU-only binaries. Change: 116473970
* TensorFlow: enable cuda host memory allocation for GPU compatibleGravatar Vijay Vasudevan2016-03-05
| | | | | | | | buffers when copying to the CPU device. Re-arranges some of the internal gpu libraries to be library vs. runtime specific. Change: 116472314
* Refactor logistic loss into component parts.Gravatar A. Unique TensorFlower2016-03-05
| | | | Change: 116471151
* Improve BiasGrad for NCHW. Add a number of shared replicas to avoid bankGravatar Xiaoqiang Zheng2016-03-05
| | | | | conflict. Change: 116468477
* Minor documentation improvement.Gravatar A. Unique TensorFlower2016-03-05
| | | | Change: 116456020
* Change L2/L1 regularization strength to in be the units of num_examples.Gravatar A. Unique TensorFlower2016-03-05
| | | | Change: 116453666
* Adding a few more validations and tests.Gravatar A. Unique TensorFlower2016-03-05
| | | | Change: 116434899
* Augmenting DeltasByExample as DataByExample to provision for extraGravatar A. Unique TensorFlower2016-03-05
| | | | | informaton needed for streamed-data duality-gam computation. Change: 116434080
* TensorFlow: Change apply_dense and apply_sparse to use a colocationGravatar Vijay Vasudevan2016-03-05
| | | | | constraint rather than ops.device, since colocation is more portable. Change: 116431514
* TensorFlow python graphdef importer: fill in op_def of created ops.Gravatar Vijay Vasudevan2016-03-05
| | | | | Adds test that failed before and passes after. Change: 116431333
* Add backwards_compatibility_test and its associated machinery.Gravatar Josh Levenberg2016-03-05
| | | | Change: 116429368
* Fix a compiler error due to signed/unsigned comparison.Gravatar Derek Murray2016-03-05
| | | | Change: 116421402
* Fix a broken tutorial linkGravatar A. Unique TensorFlower2016-03-05
| | | | Change: 116413358
* gcs TensorBoard should not print gsutil usage to consoleGravatar Dan Mané2016-03-05
| | | | Change: 116411097
* Added the ability to convert between floats and float16 on Kepler and MaxwellGravatar Benoit Steiner2016-03-05
| | | | | GPUs Change: 116409601
* TensorBoard correctly reads gcs paths from gs://path/on/gcsGravatar Dan Mané2016-03-05
| | | | | This replaces the confusing /gs/path/on/gcs syntax. Change: 116407064
* Support for running TensorBoard against Google Cloud Storage.Gravatar Dan Mané2016-03-04
| | | | | | | If the log path looks like /gs/foo/bar/baz, we assume it's a reference to a GCS object named bar/baz in bucket foo. This requires the gsutil binary to be available somewhere on $PATH. Change: 116401931
* Adding NCHW support for BiasAdd.Gravatar Xiaoqiang Zheng2016-03-04
| | | | | | | | | | | | | | | Switch BiasGrad to use shared atomics. Improve OSS CNN model benchmark performance. OSS CNN Benchmark Performance Comparison. One forward+backward pass in ms. Before After Improvement AlexNet 146 104 40.38% Overfeat 338 303 11.55% OxfordNet 983 678 44.99% GoogleNet V1 943 634 48.74% Change: 116401874
* More work on GPU/CPU memory swaping. Trigger the memory swapping only when ↵Gravatar Yuan Yu2016-03-04
| | | | | | | | | | | | the "in-use" memory reaches a given threshold. Added more bechmarks. batch seqlen units dynamic elapsed_t/seqlen 512 100 512 True 0.006626 512 500 512 True 0.007760 512 800 512 True 0.008175 Line 1 has no memory swapping; line 2 has started memory swapping; and line 3 spills a lot of tensors to CPU. More improvements are possible, but it is in a good enough state for people to use it, in particular for long sequences. Change: 116396980
* Moved config.proto from core/framework to core/protobuf.Gravatar Sherry Moore2016-03-04
| | | | Change: 116396958
* Remove endl which broke OSS.Gravatar Jianmin Chen2016-03-04
| | | | Change: 116396154
* Add native depthwise_convolution op (backward pass) with GPU kernels. Now theGravatar Jianmin Chen2016-03-04
| | | | | | | | | | | | | depthwise convolution op is completed for GPU runs. There are still space for more optimization which will be done in the future CLs. Note that the current CPU kernels with this CL are just reference kernels and not optimized at all. The old depthwise_conv is very inefficient by calling slice() on each input channel on input and filters, followed by a conv() on each input channel, after which is a concat(). Change: 116383911
* Remove script tag referencing non-existing file.Gravatar Dan Smilkov2016-03-04
| | | | Change: 116379470
* Don't put inline functions in the base Session classGravatar Geoffrey Irving2016-03-04
| | | | Change: 116368992
* Allow multiple calls to 'dynamic_rnn'.Gravatar A. Unique TensorFlower2016-03-04
| | | | | | | Usage of a TensorArray with a hardcoded name causes errors when 'dynamic_rnn' is called multiple times. Suggested solution is to use name of the current variable scope as part of TensorArray's name. Change: 116362122
* Fix broken gRPC timeout test.Gravatar Derek Murray2016-03-04
| | | | | | Depending on the test environment, opening a session against example.org might result in an UNAVAILABLE or DEADLINE_EXCEEDED error. Change: 116362101
* Added support for mean reductions on GPUGravatar Benoit Steiner2016-03-04
| | | | Change: 116322420
* TensorFlow: use typedef instead of using typename declaration to fixGravatar Vijay Vasudevan2016-03-03
| | | | | gcc 4.8 compatibility. Fixes #1362. Change: 116317708
* Adding shared_name to various wrappers of queues.Gravatar A. Unique TensorFlower2016-03-03
| | | | | | | | Queues have a parameter `shared_name` required if multiple workers want to access a queue. This is required, for instance, in the recommended use of the filename queue in a multiple worker training run. A shared filename queue should yield a shards to the training data. Different workers should work on different shards. By passing the shared_name paramter through the various wrapper functions (string_input_producer, range_input_producer, etc.) we can use these wrappers also in the case where this resource is shared by various workers. Change: 116312445
* Fix documentation typo in Graph.gradient_override_mapGravatar Geoffrey Irving2016-03-03
| | | | Change: 116310892
* TensorFlow: add exclusive tags to grpc_session_testGravatar Vijay Vasudevan2016-03-03
| | | | Change: 116303083
* Replacing use of legacy graph_util.pin_to_cpu with device('/cpu:0')Gravatar David G. Andersen2016-03-03
| | | | Change: 116302210
* Documentation update so that user ops compilation/linking works in Mac OSXGravatar A. Unique TensorFlower2016-03-03
| | | | Change: 116299181
* Report a better error when we have no input shape for metrics calculations.Gravatar Pete Warden2016-03-03
| | | | Change: 116295883
* Remove extra LICENSE files.Gravatar A. Unique TensorFlower2016-03-03
| | | | Change: 116290685
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-06 12:13:05 +0000