aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/core/grappler/optimizers/memory_optimizer.cc
Commit message (Expand)AuthorAge
* Refactor CalculateOutputSize() from VirtualScheduler protected member functio...Gravatar Peter Ma2018-10-08
* Rollforward of rollback:Gravatar A. Unique TensorFlower2018-09-02
* Automated rollback of commit 9e2ce8f4c483e68309a60dc89739bb1b79b4a12eGravatar A. Unique TensorFlower2018-09-01
* Directly import tensor.proto.h (the transitive import will be removed from te...Gravatar Eugene Brevdo2018-08-23
* Use integral power function rather than floating point version. The integral ...Gravatar A. Unique TensorFlower2018-05-17
* Relax the stringent memory allocator constraints in AssignOp if a Grappler gr...Gravatar A. Unique TensorFlower2018-05-01
* Misc. small optimizations in Grappler and shape inference code.Gravatar A. Unique TensorFlower2018-04-12
* Automated g4 rollback of changelist 187582263Gravatar A. Unique TensorFlower2018-03-02
* Automated g4 rollback of changelist 187563544Gravatar Gunhan Gulsoy2018-03-01
* Grappler: Change memory optimizer recomputation name prefix into a regexp. Th...Gravatar A. Unique TensorFlower2018-03-01
* Use optimized ops to handle GPU memory swapping: this avoids the need for 2Gravatar Benoit Steiner2018-02-26
* Turn on swapping heuristic by default to better manage memory usage on GPUGravatar Benoit Steiner2018-02-20
* Explicitely place the swap-in node: this ensures that subsequent rounds ofGravatar Benoit Steiner2018-02-13
* Extend the memory optimizations to also support accumulate_n opsGravatar Benoit Steiner2018-02-12
* Enable the use of scheduling heuristics to reduce peak memory usage by defaultGravatar Benoit Steiner2018-02-12
* Explicitly specify the value of all the attributes for the newly created AssignGravatar Benoit Steiner2018-02-06
* Serialize the evaluation of the AssignAdd nodes to make the result moreGravatar Benoit Steiner2018-02-05
* Made the addn optimization aware of the graph topologyGravatar Benoit Steiner2018-02-01
* Improved heuristics for swappingGravatar Benoit Steiner2018-01-26
* Track swapping candidates directly in memory instead of annotating them in th...Gravatar Benoit Steiner2018-01-22
* Improved detection of swappable nodesGravatar Benoit Steiner2018-01-19
* Entice TensorFlow to swap data out of GPU memory soonerGravatar Benoit Steiner2018-01-18
* Don't try to swap out persistent dataGravatar Benoit Steiner2018-01-18
* Use multiple passes to improve memory since a single pass is often not enough.Gravatar Benoit Steiner2018-01-16
* Avoid creating swap nodes multiple times if the memory optimizer is run moreGravatar Benoit Steiner2018-01-11
* Don't optimize AddN nodes with 2 or fewer inputsGravatar Benoit Steiner2018-01-11
* Implemented heuristic to decrease memory utilization of AddN nodesGravatar Benoit Steiner2018-01-11
* Properly set the type of the swap nodes.Gravatar Benoit Steiner2018-01-05
* Don't swap out inputs that expect a reference.Gravatar Benoit Steiner2018-01-05
* Implemented memory swapping heuristics for GPUGravatar Benoit Steiner2018-01-05
* Make TopologicalSort return an error status if the sorting fails.Gravatar A. Unique TensorFlower2017-12-01
* Added an option to assume that the shape of fed nodes in unknown since any sh...Gravatar Benoit Steiner2017-11-28
* Add heuristics to trigger swappingGravatar Benoit Steiner2017-11-02
* Adds sigmoid to the list of operations that can be recomputed.Gravatar A. Unique TensorFlower2017-10-31
* Grappler memory optimization: allow inputs to gradients with non-standard nam...Gravatar Allen Lavoie2017-07-31
* Merge changes from github.Gravatar Vijay Vasudevan2017-07-28
* Updated the memory optimization config to introduce an explicit default value...Gravatar Benoit Steiner2017-07-26
* Adds RELU6 to the list of cheap operations for potential recompute in memory_...Gravatar A. Unique TensorFlower2017-07-20
* Never recompute Tensors which are feeds in Grappler's memory optimizer.Gravatar Allen Lavoie2017-07-06
* Prepare to remove a bunch of proto.h includes from tensorflow/core headersGravatar Geoffrey Irving2017-06-29
* Properly handle RefEnter, RefExit and RefNextIteration nodes.Gravatar Benoit Steiner2017-06-26
* Fix a bug in the memory optimizer when two inputs to a node are both recomputedGravatar A. Unique TensorFlower2017-06-21
* Add a heuristic to Grappler's memory optimizer to recompute elementwise opsGravatar A. Unique TensorFlower2017-06-14
* Add control dependencies to prevent nodes from being swapped back in too soon.Gravatar Benoit Steiner2017-05-15
* Add a subgraph recomputation helper function to Grappler's memory optimzerGravatar A. Unique TensorFlower2017-04-20
* Added a memory optimizer to grappler.Gravatar Benoit Steiner2017-04-04
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-02 13:01:30 +0000