aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* Add default_batch_size to ExportEstimator.Gravatar Phil Stahlfeld2016-07-18
| | | | Change: 127753127
* Include cwise_ops.h in cwice_ops_gradients.h, without this change compiler ↵Gravatar A. Unique TensorFlower2016-07-18
| | | | | | couldn't find base<> template when creating custom filegroup to reduce binary size. Change: 127751354
* Rename the histogram series chart to vz-histogram-chart.Gravatar Dan Mané2016-07-18
| | | | | Update miscellaneous files to allow the vz- prefix. Change: 127750985
* Add C++ shape inference functions for:Gravatar A. Unique TensorFlower2016-07-18
| | | | | | | | | | - LookupTableFind - also validated default_value. - LookupTableInsert - also added validation for keys and values compatible. - LookupTableSize - LookupTableExport - also made the two outputs share the first dim. - InitializeTable - InitializeTableFromTextFile Change: 127750785
* Fix overflow detection in ConsumeLeadingDigits.Gravatar A. Unique TensorFlower2016-07-18
| | | | Change: 127749101
* Minor simplification in sampling_ops.py. Effectively a noop.Gravatar A. Unique TensorFlower2016-07-18
| | | | Change: 127747773
* Static shape inference added to array_ops.batch_matrix_transpose. AfterGravatar Ian Langmore2016-07-18
| | | | | | transposition with a shape tensor, all static information is lost. This change preserves shape if static shape is fully defined. Change: 127746146
* Add C++ shape inference function for Concat.Gravatar A. Unique TensorFlower2016-07-18
| | | | | Make shape_inference::InferenceContext::Add support adding two dimension. Change: 127744929
* Fix bernoulli_test.pyGravatar A. Unique TensorFlower2016-07-18
| | | | Change: 127737446
* Make bernoulli_test and normal_test more robust to randomness.Gravatar A. Unique TensorFlower2016-07-18
| | | | | | | Increased the number of samples taken, as well as tweaked the initial parameters and tolerance, such that the test rarely fails (if I remove the seed, the tests fail > 50% of the time). Change: 127731468
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-07-18
| | | | Change: 127728304
* Add C++ shape inference functions for several ops in io_ops.cc andGravatar A. Unique TensorFlower2016-07-18
| | | | | training_ops.cc. Change: 127727818
* Fix failure to detect name collisions in Saver.Gravatar Zongheng Yang2016-07-18
| | | | | | This triggers when save_path is a relative path pointing to a file named "checkpoint", but does not trigger when it is an absolute path. Change: 127727572
* Simplify shape inference test macros by removing the extra _WITH_DEF andGravatar A. Unique TensorFlower2016-07-18
| | | | | | _WITH_TENSORS macros in shape_inference_testutil.h. Instead, pass def and tensors through a new ShapeInferenceTestOp struct. Change: 127715692
* Fix minor docstring bug in tensorflow_dataframe.pyGravatar David Soergel2016-07-18
| | | | Change: 127715478
* Reduce overhead of running kernels:Gravatar A. Unique TensorFlower2016-07-18
| | | | | | | | | | - Change ExecutorState::Entry to construct the Tensor val late, avoiding default construction and moving the tensor in favor of calling the constructor directly. - Change DeviceContextMap to be vector, and make the check cheaper when no nodes are registered. - Cache in Params whether the device requires registering tensor accesses. Change: 127714854
* Automated rollback of change 127562075Gravatar Shanqing Cai2016-07-18
| | | | Change: 127709092
* Correct bug in non_max_suppression which caused failures to some tests.Gravatar A. Unique TensorFlower2016-07-17
| | | | Change: 127682508
* Automated rollback of change 127435209Gravatar A. Unique TensorFlower2016-07-17
| | | | Change: 127668670
* Exclude strided_slice from Android build.Gravatar A. Unique TensorFlower2016-07-16
| | | | Change: 127638036
* Several small changes to reduce the number of memory allocations alongGravatar A. Unique TensorFlower2016-07-16
| | | | | | | | | | | | | | Send/Recv paths: o Allow SendOp and RecvOp implementations to directly use the string buffer contained in a Rendezvous::ParsedKey object, rather than allocating their own string object. Saves two allocations per Send/Recv pair. o Use std::move in a few places to avoid copying a std::function object. o Eliminated unused ParsedName variable declaration in DeviceNameUtils::ParseLocalName. Change: 127630066
* In contrib/quantization, use eigen threadpool when calling gemmlowp toGravatar A. Unique TensorFlower2016-07-16
| | | | | avoid creating a new one each time. Change: 127624630
* Automated rollback of change 127590512Gravatar Vijay Vasudevan2016-07-15
| | | | Change: 127608996
* Adding Rename to the FileIO python APIGravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127606165
* Extend Monitor semantics to be optionally runnable on non-chief workers.Gravatar A. Unique TensorFlower2016-07-15
| | | | | This is especially useful for signaling scenarios where the chief worker wants to signal the other workers to early stop or execute some action. Change: 127603432
* Infer the number of outputs properly in shape_inference::InferenceContext,Gravatar A. Unique TensorFlower2016-07-15
| | | | | based on the nodedef and opdef. Change: 127602218
* Automated rollback of change 127564438Gravatar Suharsh Sivakumar2016-07-15
| | | | Change: 127601718
* Temporarily disable raw_rnn vs. dynamic_rnn input gradients comparison.Gravatar Eugene Brevdo2016-07-15
| | | | | | | There seems to be flakiness in this comparison. I believe it's a bug in the while_loop gradients or in a missing control flow dependency in TensorArray gradients. Change: 127598540
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127597496
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127597133
* Move OptimizationRegistry framework to open source.Gravatar A. Unique TensorFlower2016-07-15
| | | | | | Enables optimization passes in DirectSession. Refactors SimpleClientGraph and SimpleGraphExecutionState to maintain a separate mutable function library per client graph. Change: 127595477
* Adds a non_max_suppression op. This operation performs non max suppression ofGravatar A. Unique TensorFlower2016-07-15
| | | | | | | bounding boxes, greedily selecting a subset of bounding boxes in descending order of score, and pruning away boxes that have high intersection-over-union (IOU) overlap with previously selected boxes. Change: 127595411
* Adds a method in session_bundle.h to sanity check a TF export directory.Gravatar Vinu Rajashekhar2016-07-15
| | | | | - The method could have false positives but no false negatives. Change: 127595141
* Implement more common shape functions in C++ with tests:Gravatar Vijay Vasudevan2016-07-15
| | | | | | | | - MatMulShape - BiasAddShape - NoOutputs - ScalarShape Change: 127592611
* Add (linear) Support Vector Machines in tf.learn. Currently, only ↵Gravatar A. Unique TensorFlower2016-07-15
| | | | | | SDCAOptimizer is supported as the underlying optimizer for hinge_loss. Change: 127590598
* Port python session_bundle from tensorflow_serving/session_bundle to ↵Gravatar Kiril Gorovoy2016-07-15
| | | | | | tensorflow/contrib/session_bundle. Change: 127590512
* Fix partially unknown MultivariateNormal input shapes, test fully unknown caseGravatar A. Unique TensorFlower2016-07-15
| | | | | Avoids dynamic batch/non-batch dispatching when it's not necessary. Also possibly works around a bug in cond(), but that's an issue for another CL. Change: 127588880
* 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
* Rename Tensorflow Android example java classes to TensorFlow to fix nativeGravatar Andrew Harp2016-07-15
| | | | | method binding error. Change: 127585524
* Reduce the compilation time of strided_sliceGravatar Andrew Selle2016-07-15
| | | | | | | | | | | | - Break strided_slice op into multiple translation units by dimension - Refactor code to allow instantiation from multiple translation units by using a free function instead of a class member - strided_slice_op.cc used to take 163-180s. Now the aggregate of all translation units is 196s but each strided_slice_op_inst_* takes between 24-34s each - cast to a canonical type for each POD size to avoid redundant instantiations for say int32 and float32. Change: 127578995
* Fixed documentation for slice and concat.Gravatar A. Unique TensorFlower2016-07-15
| | | | | | | Documentation does not seem to be rendered properly. All other uses of ```python have a blank line before them. Change: 127574769
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127570597
* Moving slim layers to third_party.Gravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127568733
* Design an API for serving data to charts used by TensorBoard and write a ↵Gravatar Renato Utsch2016-07-15
| | | | | | | component, tf-chart-scaffold, responsible for this. Update the tf-line-chart component to receive data from tf-chart-scaffold instead of talking to the backend directly. Change: 127568346
* while_loop now accepts a (possibly nested) list or tuple of TensorsGravatar Eugene Brevdo2016-07-15
| | | | | (numpy arrays or tf Tensors), IndexedSlices, and TensorArrays. Change: 127567260
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2016-07-15
| | | | Change: 127565651
* When transposing a Tensor with a single non-1 dimension, it can be optimized ↵Gravatar A. Unique TensorFlower2016-07-15
| | | | | | | as a reshape. This is a performance optimization that is important for mobile inference graphs. They typically run with single examples (batch_size=1) and so many operations are performed on Tensors with only 1 non-1 dimension and these are just a reshape. In performance measurements on a Nexus5, this reduced times spent in transpose nodes from around 1ms to 0.02ms for a 1x20000 tensor. Change: 127564469
* Initialize random seed in TensorflowTestCase Setup to reduce flakiness in ↵Gravatar Suharsh Sivakumar2016-07-15
| | | | | | | | | | | | | | | unittests. This uncovered: - a few flaky tests, that started failing deterministically :). Their fixes are included in this CL. - random_shuffle_queue with a shared_name always fails if a user provides a graph-level random seed. This is because the op_level seed returned from random_seed.get_seed changes per graph operation. The fix for this is also included in this CL. While I was here, also added a test for random_seed, since the logic in there has become a bit complicated. Change: 127564438
* tfdb: Debug nodes inserterGravatar Shanqing Cai2016-07-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | EXPERIMENTAL: Insert special debug ops (e.g., DebugIdentity) to graph for debugging. Currently, debug ops need to take exactly one input and has the string attribute "tensor_name" to indicate what tensor it watches. For example, before the node insertion, the graph may look like: A:0 -----------1----------> B | ---------2-----------> C wherein the output slot 0 of node A feeds as the input to nodes B through edge 1 and to node C through edge 2. After the node insertion, assuming both B and C have non-Ref input, the graph becomes: A:0 ---3---> Copy -----------4----------> B | ---------5--------> C | ---------6--------> X If a node (e.g., B) has Ref input, the graph becomes: ----------------4---------------> B | A:0 ---3-----> Copy -----------5----------> C | -----------6--------> X In other words, we do not feed Refs to deep-copies to downstream nodes. The Copy node is the inserted deep-copy node that copies the input tensor on-device (e.g., CPU-to-CPU or GPU-to-GPU deep copy) that reduces the likelihood of racy updates during debug tensor-watching. X is the newly created debug node that transforms the input (copy of the watched tensor) into a debug signal. DebugIdentity is the simplest debugging paradigm, in which the debug signal (i.e., X:0) equals the tensor itself. More sophisticated debug ops can be used to transform the tensor into other useful debug signals. An example is the added DebugNanCounter op. If the nodes (A, B and C) are located on GPU and the edges from A to B or C is HOST_MEMORY, the CopyHost op will be used instead of the Copy op. A reserved string attribute "debug_url" is created for the debug ops to make it possible to send debug signals to files or RPC calls in the future. Other points worth noting: * The debug ops have control-edge connections to the original destination node, in order to ensure that the debug signals are deterministically generated before the destination node executes. * More than one debug ops can be added to watch a tensor. * A new field called "DebugTensorWatch" is added to RunOptions to support debug node insertion. * A new method GPUUtil::CopyGPUTensorToSameGPU has been added to make GPU-to-GPU deep-copy of tensors possible. * The two test files (debug_gateway_test.cc and debug_gateway_gpu_test.cc) have been consolidated to the former, by using the GOOGLE_CUDA macro. Change: 127562075
* Delete tensorflow_serving/session_bundle C++ code (copy exists in ↵Gravatar Kiril Gorovoy2016-07-15
| | | | | | | tensorflow/contrib/session_bundle) and introduce forwarding headers instead. This doesn't change any behavior since the copies are identical, but moves forward with the migration, avoids code duplication, and eliminates the possibility of this code diverging. Change: 127556144
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-09 11:08:15 +0000