This is an automated email from the ASF dual-hosted git repository.

jxie 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 cfaca3d  Fix minor math equation rendering format in NDArray API doc 
strings (#10444)
cfaca3d is described below

commit cfaca3d2b0ac19245d4f40a8116a92264aa67aed
Author: ImSheridan <xiaoyudong0...@gmail.com>
AuthorDate: Tue Apr 10 01:24:26 2018 +0800

    Fix minor math equation rendering format in NDArray API doc strings (#10444)
---
 src/operator/contrib/krprod.cc    | 2 +-
 src/operator/optimizer_op.cc      | 7 ++++---
 src/operator/regression_output.cc | 2 +-
 3 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/src/operator/contrib/krprod.cc b/src/operator/contrib/krprod.cc
index b5f9117..8fc7661 100644
--- a/src/operator/contrib/krprod.cc
+++ b/src/operator/contrib/krprod.cc
@@ -85,7 +85,7 @@ the (column-wise) Khatri-Rao product is defined as the matrix,
 .. math::
    X = A_1 \otimes \cdots \otimes A_n \in \mathbb{R}^{(M_1 \cdots M_n) \times 
N},
 
-where the :math:`k`th column is equal to the column-wise outer product
+where the :math:`k` th column is equal to the column-wise outer product
 :math:`{A_1}_k \otimes \cdots \otimes {A_n}_k` where :math:`{A_i}_k` is the kth
 column of the ith matrix.
 
diff --git a/src/operator/optimizer_op.cc b/src/operator/optimizer_op.cc
index 7d87e2c..fe0be9d 100644
--- a/src/operator/optimizer_op.cc
+++ b/src/operator/optimizer_op.cc
@@ -42,10 +42,11 @@ DMLC_REGISTER_PARAMETER(AdagradParam);
 
 NNVM_REGISTER_OP(signsgd_update)
 .describe(R"code(Update function for SignSGD optimizer.
+
 .. math::
 
  g_t = \nabla J(W_{t-1})\\
- W_t = W_{t-1} - \eta_t \text{sign}(g_t)}
+ W_t = W_{t-1} - \eta_t \text{sign}(g_t)
 
 It updates the weights using::
 
@@ -72,7 +73,7 @@ NNVM_REGISTER_OP(signum_update)
 
  g_t = \nabla J(W_{t-1})\\
  m_t = \beta m_{t-1} + (1 - \beta) g_t\\
- W_t = W_{t-1} - \eta_t \text{sign}(m_t)}
+ W_t = W_{t-1} - \eta_t \text{sign}(m_t)
 
 It updates the weights using::
  state = momentum * state + (1-momentum) * gradient
@@ -398,7 +399,7 @@ available at 
http://proceedings.mlr.press/v70/zheng17a/zheng17a.pdf.
 
  g_t = \nabla J(W_{t-1})\\
  v_t = \beta_2 v_{t-1} + (1 - \beta_2) g_t^2\\
- d_t = \frac{ (1 - \beta_1^t) }{ \eta_t } (\sqrt{ \frac{ v_t }{ 1 - \beta_2^t 
} } + \epsilon)
+ d_t = \frac{ 1 - \beta_1^t }{ \eta_t } (\sqrt{ \frac{ v_t }{ 1 - \beta_2^t } 
} + \epsilon)
  \sigma_t = d_t - \beta_1 d_{t-1}
  z_t = \beta_1 z_{ t-1 } + (1 - \beta_1^t) g_t - \sigma_t W_{t-1}
  W_t = - \frac{ z_t }{ d_t }
diff --git a/src/operator/regression_output.cc 
b/src/operator/regression_output.cc
index 9539a15..07122d4 100644
--- a/src/operator/regression_output.cc
+++ b/src/operator/regression_output.cc
@@ -131,7 +131,7 @@ The logistic function, also known as the sigmoid function, 
is computed as
 :math:`\frac{1}{1+exp(-\textbf{x})}`.
 
 Commonly, the sigmoid is used to squash the real-valued output of a linear 
model
-:math:wTx+b into the [0,1] range so that it can be interpreted as a 
probability.
+:math:`wTx+b` into the [0,1] range so that it can be interpreted as a 
probability.
 It is suitable for binary classification or probability prediction tasks.
 
 .. note::

-- 
To stop receiving notification emails like this one, please contact
j...@apache.org.

Reply via email to