aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Merge changes from github.Gravatar Andrew Harp2017-03-01
| | | | Change: 148954491
* assert_none_equal added to check_opsGravatar Ian Langmore2017-03-01
| | | | Change: 148952814
* [XLA] Add support for profiling multiple computationsGravatar David Majnemer2017-03-01
| | | | | While we are here, add support for getting the cost analysis for call HLOs. Change: 148952748
* BUGFIX: MultivariateNormalLinearOperator was not broadcasting "loc" whenGravatar Ian Langmore2017-03-01
| | | | | computing mean(). Change: 148951756
* Migrate the tf_image_dashboard component to use webfiles.Gravatar A. Unique TensorFlower2017-03-01
| | | | | Updated the demo to use only local data files relevant to the demo. Also made the demo an iron snippet so folks can see the code next to the demo. Change: 148950452
* Website Generation: Split page building from doc parsingGravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148947675
* Move estimator.predictGravatar Mustafa Ispir2017-03-01
| | | | Change: 148945600
* Adding support for higher rank SparseFeatures.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148942856
* VariableScope custom_getters now nest: a child VS with a new custom_getter willGravatar Eugene Brevdo2017-03-01
| | | | | | | call the parent scope's custom_getter instead of overriding it. The child-most scope passes the "true" VariableScope getter all the way through to the parent-most getter for the very innermost variable access. Change: 148940536
* Fix DeviceNameUtils::ParsedNameToString to render non-CPU/GPU devices correctly.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148939552
* Fix layers documentation to not lie about the data format used.Gravatar Martin Wicke2017-03-01
| | | | Change: 148936943
* Add documentation in nce_loss and sampled_softmax_loss to prevent a common bug.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148934142
* C++ API: create Scope::Impl class to hide private members/methods.Gravatar Skye Wanderman-Milne2017-03-01
| | | | | | | Note that this very slightly changes the existing API: * removes unused refiner() getter * adds copy assignment operator (before there was only an implicit copy constructor) Change: 148933743
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148930597
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148930269
* Allows missing indices in segment reduction operations.Gravatar A. Unique TensorFlower2017-03-01
| | | | | | | This makes all such operations match the behavior of unsorted_segment_sum, for which completely missing segment IDs are output as zero initialized. In this case output is either zero or one initialized depending on the aggregation function. This makes such functions more easily usable for embeddings when some features are optional. Zero/one initialization is only performed on the missing segment ids to avoid performance regressions. Change: 148929039
* Upgraded to the latest version of Eigen that should compile withGravatar Benoit Steiner2017-03-01
| | | | | cuda-clang Change: 148928329
* tfdbg: Add ability to inspect Python source against TF graphsGravatar Shanqing Cai2017-03-01
| | | | Change: 148926382
* Corrected the URL of the wheel files in "Installing TensorFlow on Windows."Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148925596
* Fixed broken links in Installation guides.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148922978
* The first application of a template made with ↵Gravatar A. Unique TensorFlower2017-03-01
| | | | | | tf.make_template(create_scope_now_=True) will now create its operations in the same name scope as its variables. This means that all variables and ops created by a template that is only applied once end up in the same scope. Change: 148922089
* Check for whether to show health pill stats based on presence of -, 0, and + ↵Gravatar A. Unique TensorFlower2017-03-01
| | | | | | | | | | | | | values. We used to rely on the mean being infinite, but oddly, Tensorboard gives us those errant values as strings: [1, 256, 256, 0, 0, 0, 0, 0, "Infinity", "-Infinity", "NaN", "NaN"] This breaks logic that checks for !== Infinity because a string is not Infinity. Something in our stack might have difficulty reading Infinity and NaN from JSON. For now, I think we might want to resort to checking for valid values. This resolves a console error. Change: 148920617
* Add interfaces in OpKernelContext and ResourceBase for tracking persistent ↵Gravatar Yuefeng Zhou2017-03-01
| | | | | | | storage. Record persistent tensor memory and persistent memory (originally auxiliary memory) in the cost model. Change: 148920117
* Fix cmake windows build.Gravatar Jingyue Wu2017-03-01
| | | | Change: 148919678
* Internal-only changesGravatar Pete Warden2017-03-01
| | | | Change: 148915145
* Add an options to quantize_nodes, allowing it to not quantize some types of Ops.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148913966
* Handles None train_op_fn properly. Makes it required for TRAIN mode. ↵Gravatar Zakaria Haque2017-03-01
| | | | | | | Provides a util function to create noon train op for cases when users want to handle optimization in model func. Also, Allows a dict of logits for multihead. See attached bug for more details. Change: 148913307
* Make a version of generate.py that works with the 1.0 release branchGravatar A. Unique TensorFlower2017-03-01
| | | | | (additional modules have been sealed since then). Change: 148910557
* Fix metagemm offsets in quantized matmul.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148905332
* Delete .*weights_ and .*bias_ methods from canned estimators.Gravatar A. Unique TensorFlower2017-03-01
| | | | Change: 148899498
* Minor speedups to xla util libraries.Gravatar Jeffrey A. Dean2017-03-01
| | | | Change: 148897757
* LinearOperator name change: LinearOperatorMatrix --> LinearOperatorFullMatrixGravatar Ian Langmore2017-03-01
| | | | Change: 148895969
* Remove disribution.is_continuous property.Gravatar Eugene Brevdo2017-03-01
| | | | | | | Most of its usefulness is determined by the reparameterization type. In the future we'll be adding a distribution Domain object / property to better describe the domain of the values a distribution may take on. Change: 148894335
* Make import_graph_def work with functionsGravatar Skye Wanderman-Milne2017-03-01
| | | | | This patch adds logic for recreating functions originally defined using @Defun and serialized in a GraphDef. Change: 148893654
* histogram_ops BUGFIX: Float64 was not previously handled.Gravatar Ian Langmore2017-03-01
| | | | | Also removing unnecessary tests. Some tests checked proper use of state, and were added at a time when histogram_fixed_width created an internal Variable. Since histogram is now (and has long been) TITO, this is not necessary. Change: 148891639
* Remove workarounds that deal with spaces in macro expansions. SelectiveGravatar A. Unique TensorFlower2017-03-01
| | | | | registrations has now been changed to handle these space properly. Change: 148886703
* LinearOperatorUDVHUpdate argument name change: diag --> update_diagGravatar Ian Langmore2017-03-01
| | | | Change: 148886147
* [TF:XLA] Add implementation of tf.one_hot to the XLA bridge.Gravatar Peter Hawkins2017-03-01
| | | | | Fix crash in normal OneHot kernel for depth < 0. Change: 148881102
* core/platform/profile_utils: Make Windows friendly.Gravatar Asim Shankar2017-03-01
| | | | | | | | | | | Some changes to assist in getting the bazel build for Windows working. In particular: - With https://github.com/tensorflow/tensorflow/commit/8898e88d5b3014a14d269560fb2f928f68562f53 core/platform/profile_utils/* needs to be included in Windows builds (Changes to tensorflow/core/BUILD in this change) - Avoid building the AndroidArmV7ACpuUtilsHelper when not on Android. Change: 148880242
* Add --printoptions flag to inspect_checkpoint, to control how numpy formats ↵Gravatar A. Unique TensorFlower2017-03-01
| | | | | | | | tensor values. For example, this allows --printoptions threshold=1000000 to print all the values in a large tensor instead of ellipsis in place of most of them. Change: 148876650
* Move dynamic_rnn_estimator._construct_rnn_cell() into rnn_common and make itGravatar A. Unique TensorFlower2017-03-01
| | | | | public. Change: 148875698
* Extends fold_batch_norms transform to also fold the mul introduced by batch ↵Gravatar A. Unique TensorFlower2017-03-01
| | | | | | normalization after fully connected layers (MatMul). Change: 148868461
* Improve error message reporting for check_numerics gradient.Gravatar Jack Rae2017-03-01
| | | | | | | At present the op message is only printed if the numeric check fails during the op's 'forward' computation. If the check fails during the gradient, there is no identifier on *which* op's gradient failed. Change: 148866334
* Remove spurious (unreachable) return statement.Gravatar A. Unique TensorFlower2017-03-01
| | | | | Reported by #7917 Change: 148861244
* Add new ops for memory statistics.Gravatar Jingyue Wu2017-02-28
| | | | | | | | | | | This CL adds only the MaxBytesInUse op, which collects the peak memory usage of a device allocator. Other ops can be added similarly when demanded. For now, we only enable MaxBytesInUse for GPU because memory statistics are unreliable for CPU allocators. This CL essentially merges part of Yaroslav Bulatov's work on https://github.com/yaroslavvb/memory_probe_ops to TensorFlow. Change: 148854571
* Windows: Script to generate binary release for the Java and C libraries.Gravatar Asim Shankar2017-02-28
| | | | | A step towards #7877 Change: 148850174
* Add mean and variance atop health pills.Gravatar A. Unique TensorFlower2017-02-28
| | | | Change: 148849242
* Fix TF-learn eval_evalset/eval_evaltrain job failure due to race condition.Gravatar James Qin2017-02-28
| | | | | | | | | | | | | | | | | TF-learn creates two jobs for evaluation: one uses evalset data and the other uses train data. Both job call Experiment._continuous_eval() -> Experiment._maybe_export() -> export_strategy.export() For TF-learn, the export function above is created by saved_model_export_utils.make_export_strategy() which calls garbage_collect_exports() after calling Estimator.export_savedmodel() The two jobs might trigger a race condition where they both try to delete the same dir. Then one fails with `NotFoundError` which isn't caught. Change: 148843701
* Fix the dlopen contrib test hack by making a pywrap_tensorflow module that ↵Gravatar Jonathan Hseu2017-02-28
| | | | | | | | | imports pywrap_tensorflow_internal with RTLD_GLOBAL. Fixes #6568 Change: 148843302
* Fix typo spooted by #7906Gravatar A. Unique TensorFlower2017-02-28
| | | | Change: 148842430
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-26 18:37:04 +0000