aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Automated g4 rollback of changelist 177619402Gravatar Brennan Saeta2017-12-12
| | | | PiperOrigin-RevId: 178800980
* [XLA] Always fold transposes into convs or dots regardless of use countGravatar David Majnemer2017-12-12
| | | | PiperOrigin-RevId: 178790193
* Sliced Wasserstein Distance metric for GANs evaluation.Gravatar A. Unique TensorFlower2017-12-12
| | | | PiperOrigin-RevId: 178788810
* Add CompositeNodeManager for Grappler VirtualScheduler.Gravatar A. Unique TensorFlower2017-12-12
| | | | | | | | | | | | | | | | CompositeNodeManager has per-device LIFO manager, FirstReadyManagers for _Send and _Recv ops, and chooses FirstReady among the ops from per-device LIFOManager and _Send and _Recv FirstReadyManagers. This one can maximizes producer-consumer locality within a device (with LIFO), but does not introduce previously reported scheduling inefficiency w.r.t. multi-device execution with separately managing _Send and _Recv ops and global FirstReady policy across devices. It's implemented, but not enabled; VirtualScheduler still uses FirstReadyManager. PiperOrigin-RevId: 178787352
* disabling flaky testGravatar Olivia Nordquist2017-12-12
| | | | PiperOrigin-RevId: 178787158
* Associative operator optimization:Gravatar A. Unique TensorFlower2017-12-12
| | | | | | | | | | | | | | Push constants down add/mul to canonicalize chains and possibly create constant nodes at the bottom. Example: + + + / \ / \ / \ c1 + --> x + --> x c1+c2 / \ / \ c2 x c2 c1 Small cleanup: Consolidate code for manipulating names of nodes added or modified during constant folding. PiperOrigin-RevId: 178785218
* Raise exception on missing unused input_map keys with C API enabled.Gravatar Skye Wanderman-Milne2017-12-12
| | | | | | | | | | | | | | | | | Without this change, the C++ ImportGraphDef API returns unused input_map keys (which are plumbed through to the C API as well). However, the Python import_graph_def API requires slightly different semantics: it throws an error for unused input_map keys that are missing from the GraphDef. This change modifies the C and C++ APIs to limit the returned keys to those missing from the GraphDef, and plumbs this through to the C API-enabled import_graph_def implementation. Note that this is a change to the existing C API. Luckily the modified method hasn't been released yet, so it's ok to change it. PiperOrigin-RevId: 178783957
* Integrate tensor pool feature to `gan_loss` function.Gravatar A. Unique TensorFlower2017-12-12
| | | | PiperOrigin-RevId: 178769850
* Disable neutral element and reciprocal optimizations again.Gravatar A. Unique TensorFlower2017-12-12
| | | | PiperOrigin-RevId: 178767676
* This CL makes two improvements to the `map_and_batch` transformation:Gravatar Jiri Simsa2017-12-12
| | | | | | | | 1) It fixes a bug that manifested as `OutOfRange` being returned prematurely. 2) It changes the behavior on sequences of elements whose size is not a multiple of `batch_size`. Previously, the implementation would drop the last small batch (similar to `batch_and_drop_remainder). Newly, the implementation returns the last small batch (similar to `batch`). PiperOrigin-RevId: 178764508
* Automated g4 rollback of changelist 178675527Gravatar Derek Murray2017-12-12
| | | | PiperOrigin-RevId: 178759398
* Add get started `Datasets` docGravatar Mark Daoust2017-12-12
| | | | PiperOrigin-RevId: 178751067
* Simplifying tfe function.pyGravatar Alexandre Passos2017-12-12
| | | | PiperOrigin-RevId: 178740804
* Remove real-data shape check in GANEstimator. Fixes github issue #14257.Gravatar A. Unique TensorFlower2017-12-12
| | | | PiperOrigin-RevId: 178737278
* Disable flaky //tensorflow/contrib/learn:dnn_linear_combined_testGravatar Yifei Feng2017-12-12
| | | | PiperOrigin-RevId: 178734940
* [XLA] Properly set layout requirements in Hlo parser.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178724659
* Use BlockHostUntilDoneWithStatus in various places.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178723711
* Disable flaky random ops test.Gravatar Gunhan Gulsoy2017-12-11
| | | | PiperOrigin-RevId: 178723108
* * HloTestBase now prints out the HLO parser error message when there is one.Gravatar Bjarke Hammersholt Roune2017-12-11
| | | | | | | * TestUtils now supports generating random literals with more than one constraint. There is still an error if the constraints conflict. PiperOrigin-RevId: 178720092
* [XLA] Add stringification to BatchNormTestParam.Gravatar Justin Lebar2017-12-11
| | | | | | | | | This way when a test fails, it prints out useful information about the failure, instead of "<48-byte object with these bytes: de ad be ef ...>" PiperOrigin-RevId: 178719733
* [XLA] Don't call timer->Nanoseconds() on a not-ok stream.Gravatar Justin Lebar2017-12-11
| | | | | | | | If the stream is not OK, the timer might not have been initialized and finalized, in which case calling timer->Nanoseconds() is illegal and will crash. PiperOrigin-RevId: 178717089
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178715353
* [XLA:CPU] Teach the CPU layout assignment about dot dimension numbersGravatar Sanjoy Das2017-12-11
| | | | | | | | There is no great need for this yet, but I noticed that the test cases were broken (they were constructing dots with unset dimension numbers), and one thing led to another. PiperOrigin-RevId: 178713597
* prefer_static_* functions added to CORE/distributions/util.pyGravatar Ian Langmore2017-12-11
| | | | PiperOrigin-RevId: 178710439
* Fix the handling of unknown rank. Previous code would wrongly treat a tensor ↵Gravatar Yao Zhang2017-12-11
| | | | | | | | of unknown rank as a scalar. PiperOrigin-RevId: 178710185
* Includes <cstdio> in the TF Lite kernels/op_macros.h to fix a compile errorsGravatar A. Unique TensorFlower2017-12-11
| | | | | | | | when building externally using either the Makefile or Bazel. The macros use stderr and fprintf which may not be defined depending on the order of headers included by the .cc files. PiperOrigin-RevId: 178708839
* Don't materialize BroadcastGradientArgs by default.Gravatar Benoit Steiner2017-12-11
| | | | PiperOrigin-RevId: 178703180
* closes #15281Gravatar Gunhan Gulsoy2017-12-11
| | | | PiperOrigin-RevId: 178701096
* Automated g4 rollback of changelist 178634559Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178695724
* Initialize local_resources during session initialization.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178694869
* [XLA] Optimize dot(concat(..), constant)Gravatar Sanjoy Das2017-12-11
| | | | | | | | dot(concat(..), constant) and dot(constant, concat(..)) can be rewritten to avoid the concatenate. This can itself be a win, but can also help unlock other optimization opportunities. PiperOrigin-RevId: 178691585
* Strength reduce division by a constant to multiplication by the reciprocal ↵Gravatar A. Unique TensorFlower2017-12-11
| | | | | | constant. PiperOrigin-RevId: 178689056
* Fix the remote call graph construction so that the created _Send ops for ↵Gravatar A. Unique TensorFlower2017-12-11
| | | | | | | | returning the results point to the correct return value. List available workers when the remote call target is not available. PiperOrigin-RevId: 178687525
* [XLA:CPU] Minor refactor to the CPU layout assignment codeGravatar Sanjoy Das2017-12-11
| | | | | | | Lifts unnecessary lambdas to reduce clutter. This will make a later change more readable. PiperOrigin-RevId: 178686976
* [tf.data] Use a more efficient dispatch mechanism for functions in datasets.Gravatar Derek Murray2017-12-11
| | | | | | | | | | | | | | | | | This change adds an overload of the `FunctionLibraryRuntime::Run()` method that allows users to pass argument and return value containers in a `CallFrameInterface` object, rather than using the current (and expensive for large arities) default `FunctionCallFrame` implementation. It also specializes `CapturedFunction` to use this interface. Note that the new overload currently only supports local function execution, and more restructuring will be required to take advantage of it in the remote function execution case. This change should especially benefit datasets where each element has a large number of components (typically when training data have many features). PiperOrigin-RevId: 178684431
* Fix for variable naming when executing eagerlyGravatar Allen Lavoie2017-12-11
| | | | | | | | | name_scope bypassed the Graph.name_scope slash-stripping logic (strip a trailing slash if it exists, then add one back unconditionally) when executing eagerly, leading to extra slashes at the end of some variable names and a failure to break out of nested name scopes. PiperOrigin-RevId: 178676873
* Fix definition of tflite_smartreplyGravatar Justine Tunney2017-12-11
| | | | PiperOrigin-RevId: 178675580
* Mark a FunctionDef's signature as stateful when it contains a stateful node.Gravatar Derek Murray2017-12-11
| | | | | | This fixes a bug where two calls to the same stateful function will erroneously be eliminated as common subexpressions. It is also a step towards pruning nodes from function bodies, which is necessary for a variety of `Dataset` optimizations. PiperOrigin-RevId: 178675527
* [XLA] Move BatchDot unrolling from TF2XLA bridge to AlgebraicSimplifier so ↵Gravatar A. Unique TensorFlower2017-12-11
| | | | | | that unrolling can be selectively enabled/disabled per backend (should be no performance change). PiperOrigin-RevId: 178666990
* Tune SQLiteGravatar Justine Tunney2017-12-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change makes sure the b-tree page size isn't 1024 bytes. It also enables WAL mode. This means TensorBoard can perform reads at the same time as TensorFlow is performing writes. We now also fsync() less often. This shouldn't carry any risk of database corruption in WAL mode. Since WAL mode uses shared memory, writes become immediately available to other processes, but they won't become durable until after the OS decides to flush the FS cache. This makes the DB writer faster than the file writer, at least in cases where the DB is tiny. We probably make it go faster still, once we find a way to use transactions. Name Cold ?s Average ?s Flushing ?s Size B ?i.i 1,920 69 0 0 Scalar 1.0 FS 1,623 337 4,258 11,348 Scalar 1.0 TB FS 3,137 527 4,213 17,023 Scalar 2.0 FS 3,319 681 3,917 11,348 Scalar 2.0 DB 2,601 578 217 118,784 Tensor 1.0 FS 4 6,397 558 4,276 14,215 Tensor 2.0 FS 4 1,678 613 3,971 24,455 Tensor 2.0 DB 4 3,605 278 313 118,784 Tensor 1.0 FS 128 1,857 289 4,397 47,111 Tensor 2.0 FS 128 3,558 721 10,894 57,351 Tensor 2.0 DB 128 3,508 585 203 118,784 Tensor 1.0 FS 8192 2,677 525 4,400 2,119,816 Tensor 2.0 FS 8192 2,248 822 4,006 2,130,056 Tensor 2.0 DB 8192 4,346 370 449 126,976 PiperOrigin-RevId: 178666363
* [XLA:CPU] Error on unsupported dot instructionsGravatar Sanjoy Das2017-12-11
| | | | | | This is a stopgap measure to avoid silently miscompiling dot operations. PiperOrigin-RevId: 178666218
* Enable optimizations of operations with neutral/absorbing elements by ↵Gravatar A. Unique TensorFlower2017-12-11
| | | | | | default. We leave removal of addition and subtraction with zero out for now, since it is used as a "hack" to force a copy of a tensor in a few places. Once we have fixed this code, we can enable it. PiperOrigin-RevId: 178665567
* Support different threading modes in GPU device.Gravatar Xiaoqiang Zheng2017-12-11
| | | | | | | All modes are experimental for now. The goal is to find the best setting, and change the default to pick that. PiperOrigin-RevId: 178662212
* Support all unary ops.Gravatar Yao Zhang2017-12-11
| | | | PiperOrigin-RevId: 178662178
* Always inline functions when creating an item.Gravatar Benoit Steiner2017-12-11
| | | | PiperOrigin-RevId: 178661624
* Add ReverseDFSFrom variant that works with const Node*.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178658907
* Rename ABSL MacrosGravatar A. Unique TensorFlower2017-12-11
| | | | | | | | | | | This LSC will rename ABSL macros. Most macro will be renamed with ABSL_ prefix. Some might have completely new names. Please see the list of the macros affected. For example, MUST_USE_RESULT will be renamed ABSL_MUST_USE_RESULT The purpose of this LSC is to avoid name conflicts for the ABSL release. To see the details go/absl-macros. PiperOrigin-RevId: 178655181
* Be more conservative when optimizing full reductionsGravatar Benoit Steiner2017-12-11
| | | | PiperOrigin-RevId: 178652323
* Optimized specializations for 3-channel depthwise with multiplier 2 and 4.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178647824
* Adds remaining tests to _shared_embedding_column.Gravatar A. Unique TensorFlower2017-12-11
| | | | PiperOrigin-RevId: 178644910
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-08 08:02:41 +0000