- 14 12月, 2018 5 次提交
-
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225377288
-
由 Shashi Shekhar 提交于
Add --cxxopt='--std=c++11' to instructions. PiperOrigin-RevId: 225376880
-
由 Dan Moldovan 提交于
Use weak refs when caching symbols in the namespace, to avoid tripping circular reference detectors in tests. PiperOrigin-RevId: 225376019
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225372823
-
由 A. Unique TensorFlower 提交于
This CL adds a fast-path for R2 dot. For now the fast implementation has certain limitations: 1. Only operands with default layout, and 2. float type It uses the XLA's CPU runtime functions which invoke eigen. PiperOrigin-RevId: 225372611
-
- 13 12月, 2018 35 次提交
-
-
由 Skye Wanderman-Milne 提交于
These are required for cond_v2 to work properly on GPU. The fix in optional_ops.cc is subtle, but I can't think of a way to more generally fix this lifetime issue. PiperOrigin-RevId: 225369355
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225366627
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225364007
-
由 A. Unique TensorFlower 提交于
.bat files have an 8k limit on the command line length which we are now hitting. PiperOrigin-RevId: 225355646
-
由 A. Unique TensorFlower 提交于
It contains the same source files as fused_batch_norm_op, leading to duplicate symbol errors in some configurations. PiperOrigin-RevId: 225350457
-
由 James Keeling 提交于
I pull the logic to convert a *C.TF_DeviceList to a []Device into its own function `deviceSliceFromDeviceList`. This is so I can reuse it when wrapping the Eager C API. PiperOrigin-RevId: 225344064
-
由 James Keeling 提交于
This is useful when printing devices. PiperOrigin-RevId: 225343692
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225327770
-
由 Michael Reneer 提交于
PiperOrigin-RevId: 225323136
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225322589
-
由 Anna R 提交于
PiperOrigin-RevId: 225321405
-
由 Guangda Lai 提交于
PiperOrigin-RevId: 225318469
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225310802
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225307980
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225299240
-
由 Youlong Cheng 提交于
PiperOrigin-RevId: 225298488
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225296419
-
由 TensorFlower Gardener 提交于
PiperOrigin-RevId: 225296218
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225294738
-
由 Scott Zhu 提交于
Also increase the shard to speed up the tests. PiperOrigin-RevId: 225293834
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225292379
-
由 Scott Zhu 提交于
Test that can only run in v1 has the proper bug attached. PiperOrigin-RevId: 225291768
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225289423
-
由 Francois Chollet 提交于
PiperOrigin-RevId: 225287527
-
由 Dan Moldovan 提交于
PiperOrigin-RevId: 225283401
-
由 Francois Chollet 提交于
PiperOrigin-RevId: 225283392
-
由 A. Unique TensorFlower 提交于
Adds unicode_decode and unicode_decode_with_offsets ops, which decode strings into unicode codepoints. Adds unicode_split and unicode_split_with_offset ops, which split strings into unicode characters. RELNOTES: Adds unicode_decode, unicode_decode_with_offsets, unicode_split, and unicode_split_with_offset ops. PiperOrigin-RevId: 225281768
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 225281253
-
由 Trevor Morris 提交于
-
由 Trevor Morris 提交于
Fix bug with masking and undefined batch dims. Masking needs to take place inside of GetStridedSliceBound
-
由 Trevor Morris 提交于
-
由 Trevor Morris 提交于
-
由 Trevor Morris 提交于
-
由 Trevor Morris 提交于
-
由 Trevor Morris 提交于
-