- 18 12月, 2019 4 次提交
-
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 286103069
-
由 David Chen 提交于
PiperOrigin-RevId: 286087529
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 286061591
-
由 David Chen 提交于
PiperOrigin-RevId: 285991432
-
- 17 12月, 2019 6 次提交
-
-
由 Chris Jones 提交于
PiperOrigin-RevId: 285974579
-
由 Yuqi Li 提交于
PiperOrigin-RevId: 285967157
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285930545
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285905996
-
由 Hongkun Yu 提交于
Remove not maintained code path. PiperOrigin-RevId: 285869559
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 285844156
-
- 16 12月, 2019 1 次提交
-
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285765110
-
- 15 12月, 2019 3 次提交
-
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285618209
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285613648
-
由 Zongwei Zhou 提交于
PiperOrigin-RevId: 285568490
-
- 14 12月, 2019 4 次提交
-
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 285533511
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 285503670
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 285498103
-
由 Allen Wang 提交于
PiperOrigin-RevId: 285422968
-
- 13 12月, 2019 5 次提交
-
-
由 Gonzalo Gasca Meza 提交于
Nit, needs new line
-
由 David Chen 提交于
PiperOrigin-RevId: 285322154
-
由 Chen Chen 提交于
PiperOrigin-RevId: 285285388
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 285234431
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 285223338
-
- 12 12月, 2019 1 次提交
-
-
由 A. Unique TensorFlower 提交于
enable_eager, tf_gpu_thread_mode and datasets_num_private_threads PiperOrigin-RevId: 285157129
-
- 11 12月, 2019 4 次提交
-
-
由 David Chen 提交于
PiperOrigin-RevId: 284874717
-
由 Pengchong Jin 提交于
PiperOrigin-RevId: 284854952
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 284805626
-
由 Chen Chen 提交于
PiperOrigin-RevId: 284792715
-
- 10 12月, 2019 1 次提交
-
-
由 Chen Chen 提交于
PiperOrigin-RevId: 284624714
-
- 08 12月, 2019 1 次提交
-
-
由 Hongkun Yu 提交于
Fix link.
-
- 07 12月, 2019 4 次提交
-
-
由 Chen Chen 提交于
PiperOrigin-RevId: 284278191
-
由 Skirmantas Kligys 提交于
* Manually quantize selected inputs before summing them up. PiperOrigin-RevId: 283653640 * Fix messy variable paths in LSTD head. With a VariableScope passed as parameter into tf.variable_scope(), TF gets all confused and starts placing variables into scopes different from operations (variable_scope != name_scope). That leads to very messy TF graphs. Before CL: FeatureExtractor/LSTM/conv_lstm_cell/bottleneck_0/depthwise_weights FeatureExtractor/LSTM/LSTM/conv_lstm_cell/bottleneck_0/separable_conv2d/depthwise After CL: FeatureExtractor/LSTM/conv_lstm_cell/bottleneck_0/depthwise_weights FeatureExtractor/LSTM/conv_lstm_cell/bottleneck_0/separable_conv2d/depthwise PiperOrigin-RevId: 284011539
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 284259997
-
由 Chen Chen 提交于
PiperOrigin-RevId: 284203261
-
- 06 12月, 2019 4 次提交
-
-
由 Haoyu Zhang 提交于
Remove code that disables pin-to-host optimization. It's disabled by default and we already have XLA presubmit that will catch any issues when the optimization is reenabled. PiperOrigin-RevId: 284078252
-
由 A. Unique TensorFlower 提交于
Also adding support for CTL mode in the borg file. PiperOrigin-RevId: 284075404
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 284016229
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 283987800
-
- 05 12月, 2019 2 次提交
-
-
由 Jaehong Kim 提交于
PiperOrigin-RevId: 283962490
-
由 Haoyu Zhang 提交于
No need to test explicit_gpu_placement + no_dist_strat + function case, since the performance issue is already fixed by unifying the Keras codepaths. However, there is still performance gap using the explicit_gpu_placement flag when run_eagerly. PiperOrigin-RevId: 283890008
-