aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/core/grappler/costs/graph_properties_test.cc
Commit message (Expand)AuthorAge
* Fix bug in Grappler constant folding: The logic detecting full reductions was...Gravatar A. Unique TensorFlower2018-10-04
* Improve static shape inference in grappler by propagating tensors_as_shapes b...Gravatar Doe Hyun Yoon2018-09-12
* Replace Placeholder with Const to GrapplerFunctionItem for function shape inf...Gravatar Doe Hyun Yoon2018-09-06
* Directly import tensor.proto.h (the transitive import will be removed from te...Gravatar Eugene Brevdo2018-08-23
* Add graphdef version number to GrapplerFunctionItem.Gravatar Doe Hyun Yoon2018-08-23
* Update Grappler to use existing functions for retrieving a node'sGravatar A. Unique TensorFlower2018-07-20
* Function shape inference for Grappler.Gravatar A. Unique TensorFlower2018-07-10
* Set shapes and types to queue ops, if not set by enqueue ops.Gravatar A. Unique TensorFlower2018-06-15
* Added support for strided slicing of symbolic shapesGravatar Benoit Steiner2018-05-14
* Simplified shape inference.Gravatar Benoit Steiner2018-05-01
* Run shape inference directly on the graphdef instead of building an intermedi...Gravatar Benoit Steiner2018-04-24
* Don't rely on graph contruction for an initial shape inference.Gravatar Benoit Steiner2018-04-16
* Don't fail when running shape inference on a graph that contains functions.Gravatar Benoit Steiner2018-04-09
* Disable shape inference for functions since it doesn't seem to work whenGravatar Benoit Steiner2018-03-29
* Implement partial constant folding for Concat.Gravatar A. Unique TensorFlower2018-03-14
* Automated g4 rollback of changelist 188501394Gravatar A. Unique TensorFlower2018-03-09
* Implement partial constant folding for Concat.Gravatar A. Unique TensorFlower2018-03-09
* Fixed a typo that resulted in the graph being processed in reverse topologicalGravatar Benoit Steiner2018-01-02
* Make sure we serialize the creation and deletion of clusters from python to a...Gravatar Benoit Steiner2017-12-11
* Dont optimize away control triggersGravatar Benoit Steiner2017-12-07
* Made sure the unknown shapes of placeholders always propagate to their fanoutsGravatar Benoit Steiner2017-11-29
* Added an option to assume that the shape of fed nodes in unknown since any sh...Gravatar Benoit Steiner2017-11-28
* Track symbolic shapes through shapeN operationsGravatar Benoit Steiner2017-11-27
* Improved shape inferenceGravatar Benoit Steiner2017-11-17
* Implemented Processor<ShapeHandle>, which allows us to merge shapes of unknow...Gravatar Benoit Steiner2017-11-10
* Simplify graph construction with an option to not validate colocation constra...Gravatar Yao Zhang2017-11-10
* Track merged shapes and dimensions more accurately.Gravatar Benoit Steiner2017-11-07
* Improved encoding on shapes in grappler.Gravatar Benoit Steiner2017-11-06
* Enable shape inference on functions in grappler.Gravatar A. Unique TensorFlower2017-10-30
* Do shape inference through Enqueue ops only for Queue ops and Enter ops with ...Gravatar A. Unique TensorFlower2017-09-28
* Annotate the graph properties with input values if they're known statically.Gravatar Benoit Steiner2017-08-30
* Infer output shape for restore op.Gravatar Yuefeng Zhou2017-08-22
* Deleted the code that infers shapes of restores since it crashesGravatar Benoit Steiner2017-08-09
* Set the fanin shapes of assign op;Gravatar Yuefeng Zhou2017-08-04
* Infer output shape for restore op.Gravatar Yuefeng Zhou2017-07-31
* Move Grappler test GraphDefs to separate filesGravatar A. Unique TensorFlower2017-07-31
* Prepare to remove a bunch of proto.h includes from tensorflow/core headersGravatar Geoffrey Irving2017-06-29
* Only wait for one non-control input for Merge nodes if there is a loop. This isGravatar A. Unique TensorFlower2017-06-28
* Infer shapes for loops during Grappler static inferenceGravatar A. Unique TensorFlower2017-06-20
* Initialize tensor in graph_properties_test, to avoid msan complaint.Gravatar A. Unique TensorFlower2017-06-06
* Added a test to make sure that graph properties for variables are properlyGravatar Benoit Steiner2017-06-05
* Change shape inference so that a single resource tensor can carryGravatar A. Unique TensorFlower2017-05-25
* Calling InferenceContext::UnknownShapes twice produces 2 shape handles for wh...Gravatar Benoit Steiner2017-05-02
* Reset the inputs to ShapeRefiner::RunShapeFn so that it behaves the same ever...Gravatar Benoit Steiner2017-05-02
* Enable grappler to propagate shapes through queues.Gravatar Benoit Steiner2017-05-01
* Feed the test input with random numbers instead of constants to preventGravatar Benoit Steiner2017-03-27
* Started to open source Grappler. First application is the GPU layout optimizer.Gravatar Benoit Steiner2017-03-08