diff options
| author |  Yuefeng Zhou <yuefengz@google.com> | 2018-08-29 11:18:03 -0700 |
|---|---|---|
| committer |  TensorFlower Gardener <gardener@tensorflow.org> | 2018-08-29 11:22:09 -0700 |
| commit | 0a4353e835f4c32e663488d6d1396dcc34c833a7 (patch) | |
| tree | 139e0df944c135534e9ee8b2071f548914e189d1 /tensorflow/contrib/lite/mmap_allocation_disabled.cc | |
| parent | a2d2ac24a2ca2a50524f0acab1643b7d186e5074 (diff) | |
Fixed that in collective all reduce cross tower ops that batch all reduce is not properly used.
PiperOrigin-RevId: 210755447
Diffstat (limited to 'tensorflow/contrib/lite/mmap_allocation_disabled.cc')
0 files changed, 0 insertions, 0 deletions