experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Added TPU code for FTRL with multiply_linear_by_lr setting (CPU and GPU

Browse Source 
versions already added).

PiperOrigin-RevId: 305428489
Change-Id: Ib2bb5b26ba06c535b96222aa1488a018d4a8f343
This commit is contained in:
A. Unique TensorFlower 2020-04-08 01:11:22 -07:00 committed by TensorFlower Gardener
parent 9a273849b2
commit 72e0b46788
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
tensorflow/core/protobuf/tpu/optimization_parameters.proto
View File

@ -86,6 +86,7 @@ message FtrlParameters {
float lr_power = 3;
float initial_accum = 4;
float initial_linear = 5;
bool multiply_linear_by_lr = 6;
}
// The Adam optimizer does not implement hyper-parameter update; use the dynamic