This is an automated email from the ASF dual-hosted git repository. cjolivier01 pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/master by this push: new 81dc02e add missed optimizer docs (#9779) 81dc02e is described below commit 81dc02e517e0cd2be3148237e5bbb0482dc1f09f Author: Sheng Zha <s...@users.noreply.github.com> AuthorDate: Tue Feb 13 08:01:14 2018 -0800 add missed optimizer docs (#9779) --- docs/api/python/optimization/optimization.md | 6 ++++++ python/mxnet/optimizer.py | 4 ++-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/docs/api/python/optimization/optimization.md b/docs/api/python/optimization/optimization.md index e333b00..7d6276d 100644 --- a/docs/api/python/optimization/optimization.md +++ b/docs/api/python/optimization/optimization.md @@ -96,8 +96,14 @@ implements one weight updating function. Adam AdaGrad AdaDelta + Adamax + Nadam DCASGD SGLD + Signum + FTML + LBSGD + Ftrl ``` ## The ``mxnet.lr_scheduler`` package diff --git a/python/mxnet/optimizer.py b/python/mxnet/optimizer.py index 0652772..065c08c 100644 --- a/python/mxnet/optimizer.py +++ b/python/mxnet/optimizer.py @@ -68,8 +68,8 @@ class Optimizer(object): Flag to control the internal precision of the optimizer. ``False`` results in using the same precision as the weights (default), ``True`` makes internal 32-bit copy of the weights and applies gradients - in 32-bit precision even if actual weights used in the model have lower precision. - Turning this on can improve convergence and accuracy when training with float16. + in 32-bit precision even if actual weights used in the model have lower precision. + Turning this on can improve convergence and accuracy when training with float16. Properties ---------- -- To stop receiving notification emails like this one, please contact cjolivie...@apache.org.