"git@developer.sourcefind.cn:zhaoyu6/sglang.git" did not exist on "3602692c7ca7c3757cc3d2b5dfc829209205731a"
- 08 May, 2021 1 commit
-
-
Yuexin Wu authored
PiperOrigin-RevId: 372675330
-
- 30 Apr, 2021 2 commits
-
-
Abdullah Rashwan authored
PiperOrigin-RevId: 371418693
-
Abdullah Rashwan authored
PiperOrigin-RevId: 371389966
-
- 26 Apr, 2021 1 commit
-
-
Hao Wu authored
PiperOrigin-RevId: 370551493
-
- 21 Apr, 2021 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 369588823
-
- 20 Apr, 2021 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 369375299
-
- 14 Apr, 2021 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 368535232
-
Jaehong Kim authored
This CL changes ResidualBlock and InvertedBottleneckBlock. PiperOrigin-RevId: 368383954
-
- 12 Apr, 2021 1 commit
-
-
Reed Wanderman-Milne authored
For all modified calls to set_mixed_precision_policy(), the loss_scale argument was removed, as it cannot be passed if the nonexperimental API is used. For all such callers, the loss_scale is later used to explicitly create a LossScaleOptimizer, so removing the argument has no impact. Switching to the non-experimental LossScaleOptimizer has no effect, as it has near identical behavior and all isinstance checks within the official models check for the non-experimental version. PiperOrigin-RevId: 368101975
-
- 06 Apr, 2021 3 commits
-
-
Reed Wanderman-Milne authored
PiperOrigin-RevId: 367105004
-
Reed Wanderman-Milne authored
This has no functional impact since the default is currently True, but I plan on changing the default to False soon. PiperOrigin-RevId: 367049942
-
Reed Wanderman-Milne authored
The function `tf.train.experimental.enable_mixed_precision_graph_rewrite` will be removed from the TF2 namespace soon, at which point it will only be accessible under tf.compat.v1. PiperOrigin-RevId: 367046393
-
- 30 Mar, 2021 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 365713370
-
- 25 Mar, 2021 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 365085378
-
- 18 Mar, 2021 2 commits
-
-
Frederick Liu authored
PiperOrigin-RevId: 363595377
-
Frederick Liu authored
PiperOrigin-RevId: 363595377
-
- 06 Mar, 2021 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 361348924
-
Hongkun Yu authored
PiperOrigin-RevId: 361348924
-
- 03 Mar, 2021 4 commits
-
-
Reed Wanderman-Milne authored
The default is True, but I plan on changing it to False soon. After that, I plan on removing the argument and never using the experimental API. PiperOrigin-RevId: 360724698
-
Reed Wanderman-Milne authored
The default is True, but I plan on changing it to False soon. After that, I plan on removing the argument and never using the experimental API. PiperOrigin-RevId: 360724698
-
Le Hou authored
PiperOrigin-RevId: 360719881
-
Le Hou authored
PiperOrigin-RevId: 360719881
-
- 02 Mar, 2021 2 commits
- 01 Mar, 2021 2 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 360256877
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 360256877
-
- 26 Feb, 2021 2 commits
- 17 Feb, 2021 2 commits
-
-
Frederick Liu authored
[core] Use the full model instead of checkpoint_items when constructing Checkpoint in multitask evaluator. PiperOrigin-RevId: 357962526
-
Frederick Liu authored
[core] Use the full model instead of checkpoint_items when constructing Checkpoint in multitask evaluator. PiperOrigin-RevId: 357962526
-
- 15 Feb, 2021 2 commits
-
-
Frederick Liu authored
[core] Update BestCheckpointExporter to support dictionary with arbitrary depth (possibly coming from multitask eval). Also added best checkpointexporter support to multitask. PiperOrigin-RevId: 357584596
-
Frederick Liu authored
[core] Update BestCheckpointExporter to support dictionary with arbitrary depth (possibly coming from multitask eval). Also added best checkpointexporter support to multitask. PiperOrigin-RevId: 357584596
-
- 07 Feb, 2021 2 commits
- 22 Jan, 2021 6 commits
-
-
Le Hou authored
PiperOrigin-RevId: 353327814
-
Le Hou authored
PiperOrigin-RevId: 353327814
-
Hongkun Yu authored
PiperOrigin-RevId: 353315689
-
Hongkun Yu authored
PiperOrigin-RevId: 353315689
-
-
-