- 12 10月, 2018 12 次提交
-
-
由 Dan Moldovan 提交于
Modify converted_call to accept a separate argument when the callable is an attribute, representing the attribute's owner. This gives converted_call a chance to override the attribute lookup, and in turn allows working around the limitation `super` that prevents it from resolving dynamic attributes. PiperOrigin-RevId: 216727687
-
由 Adrian Kuegel 提交于
Also fix the DotLayout test, it would pass even when commenting out the dot specific logic in GpuLayoutAssignment. PiperOrigin-RevId: 216726796
-
由 Tong Shen 提交于
PiperOrigin-RevId: 216725062
-
由 A. Unique TensorFlower 提交于
patch command can exceed the 10s timeout limit on some slow machines. Fixes #22761 PiperOrigin-RevId: 216721485
-
由 Dan Moldovan 提交于
Use the functional operator pattern for logical expressions. The current implementation relied on the TF import, which is now gone. PiperOrigin-RevId: 216719675
-
由 Gunhan Gulsoy 提交于
PiperOrigin-RevId: 216718180
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216714487
-
由 Scott Zhu 提交于
Also fix the initial_state and constants param in the eager context. PiperOrigin-RevId: 216712447
-
由 Michael Case 提交于
PiperOrigin-RevId: 216706771
-
由 Dan Moldovan 提交于
PiperOrigin-RevId: 216706627
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216705389
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216705354
-
- 11 10月, 2018 28 次提交
-
-
由 Ilya Biryukov 提交于
PiperOrigin-RevId: 216697820
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216697425
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216691593
-
由 Adrian Kuegel 提交于
PiperOrigin-RevId: 216688700
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216686849
-
由 Pete Warden 提交于
PiperOrigin-RevId: 216686136
-
由 A. Unique TensorFlower 提交于
We are working on disallowing xla::Add on PRED types because they can be confusing. This is a prerequiste for that to not fail over in the new shape check. PiperOrigin-RevId: 216685015
-
由 A. Unique TensorFlower 提交于
Wonder how that could have ever let this CL to go in in the first place. Automated rollback of commit 905deeaa PiperOrigin-RevId: 216683052
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216663386
-
由 Francois Chollet 提交于
PiperOrigin-RevId: 216639755
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216638239
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216633097
-
由 James Qin 提交于
PiperOrigin-RevId: 216632480
-
由 Justin Lebar 提交于
We have a lot of such warnings and don't make an attempt to clean them up internally. PiperOrigin-RevId: 216631010
-
由 Justin Lebar 提交于
No functional change. PiperOrigin-RevId: 216629980
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 216628574
-
由 Jiri Simsa 提交于
PiperOrigin-RevId: 216627219
-
由 Kay Zhu 提交于
[XLA::GPU] Explicitly use {lhs,rhs}_contracting_dimension from DotDimensionNumbers as the reduction dimension for Dot. PiperOrigin-RevId: 216627100
-
由 A. Unique TensorFlower 提交于
- Array fields minmax and quantization_params are pointers. (so the current check could have false negatives as identical objects have different addresses) - also compare narrow_range. (so the current check could have false positives --- my bad, I added narrow_range later and forgot to update this code). PiperOrigin-RevId: 216626868
-
由 Benjamin Kramer 提交于
We fuse everything into the scatter now, and emit two kernels. The first kernel fills the output buffer with the computation fused into the scatter operand. The second kernel is a regular scatter, which also contains the fused operations from the updates and scatter_indices inputs. PiperOrigin-RevId: 216624225
-
由 Rohan Jain 提交于
PiperOrigin-RevId: 216624182
-
由 Justin Lebar 提交于
- Make the channel-dims-equal-to-3 heuristic more specific. The relevant cudnn optimization only occurs with input-channels equal to 3 and output-channels equal to 32 or 64. - Simplify the logic that decided which shapes' dimensions to pad by using a clever (if I do say so myself) trick. Really all we need is a new layer of indirection which allows us to say "modify the input shape's feature dim" and have that be reflected in the lhs shape's feature dim. - Rename it to CudnnConvPadForTensorCores. I initially thought that the 3-input-channels optimization applied to Pascal, but experimentally this seems to be incorrect. PiperOrigin-RevId: 216623854
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216623201
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216622520
-
由 Jared Duke 提交于
INFINITY define requires the cmath header. PiperOrigin-RevId: 216621867
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 216619617
-
由 Justin Lebar 提交于
Make them shorter and more consistent. - CudnnConvolutionFoo -> CudnnConvFoo - PadInsertion -> CudnnConvPaddingLegalization - PadForTensorCores -> CudnnConvPadForSpeed (padding channel dimensions from 3 -> 4 is not a tensor-cores-related optimization and ideally should be run on P100s as well). PiperOrigin-RevId: 216618934
-
由 Russell Power 提交于
PiperOrigin-RevId: 216618378
-