From 296e756b509cbb48940e90f08ec5d051ebfcf5e8 Mon Sep 17 00:00:00 2001
From: wangshuide2020 <wangshuide@huawei.com>
Date: Thu, 3 Jun 2021 15:23:42 +0800
Subject: [PATCH] fix the reduction args of Loss operator, line too long and
 other warning problems.

---
 mindspore/nn/loss/loss.py             | 4 +---
 mindspore/nn/optim/lars.py            | 5 +++--
 mindspore/nn/optim/lazyadam.py        | 3 ++-
 mindspore/ops/_grad/grad_quant_ops.py | 1 +
 4 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/mindspore/nn/loss/loss.py b/mindspore/nn/loss/loss.py
index 1c70ba34c69..0d62e9ca8cd 100644
--- a/mindspore/nn/loss/loss.py
+++ b/mindspore/nn/loss/loss.py
@@ -40,11 +40,9 @@ class Loss(Cell):
     """
     def __init__(self, reduction='mean'):
         super(Loss, self).__init__()
-        if reduction is None:
-            reduction = 'none'
 
         if reduction not in ('mean', 'sum', 'none'):
-            raise ValueError(f"reduction method for {reduction.lower()} is not supported")
+            raise ValueError(f"reduction method for {reduction} is not supported")
 
         self.average = True
         self.reduce = True
diff --git a/mindspore/nn/optim/lars.py b/mindspore/nn/optim/lars.py
index f19992fcb75..0dcf543c7db 100755
--- a/mindspore/nn/optim/lars.py
+++ b/mindspore/nn/optim/lars.py
@@ -58,7 +58,8 @@ class LARS(Optimizer):
     .. math::
 
         \begin{array}{ll} \\
-            \lambda  = \frac{\theta  \text{ * } || \omega  ||  }{|| g_{t} || \text{ + } \delta \text{ * } || \omega  || }  \\
+            \lambda  = \frac{\theta  \text{ * } || \omega  ||  } \\
+                            {|| g_{t} || \text{ + } \delta \text{ * } || \omega  || }  \\
             \lambda  =
             \begin{cases}
                 \min(\frac{\lambda}{\alpha }, 1)
@@ -70,7 +71,7 @@ class LARS(Optimizer):
         \end{array}
 
     :math:`\theta` represents `coefficient`, :math:`\omega` represents `parameters`, :math:`g` represents `gradients`,
-    :math:`t` represents updateing step, :math:`\delta` represents `weight_decay`,
+    :math:`t` represents updating step, :math:`\delta` represents `weight_decay`,
     :math:`\alpha` represents `learning_rate`, :math:`clip` represents `use_clip`.
 
     Args:
diff --git a/mindspore/nn/optim/lazyadam.py b/mindspore/nn/optim/lazyadam.py
index 8eca2300582..49141009623 100644
--- a/mindspore/nn/optim/lazyadam.py
+++ b/mindspore/nn/optim/lazyadam.py
@@ -107,7 +107,8 @@ class LazyAdam(Optimizer):
     r"""
     This optimizer will apply a lazy adam algorithm when gradient is sparse.
 
-    The original adam algorithm is proposed in `Adam: A Method for Stochastic Optimization <https://arxiv.org/abs/1412.6980>`_.
+    The original adam algorithm is proposed in
+    `Adam: A Method for Stochastic Optimization <https://arxiv.org/abs/1412.6980>`_.
 
     The updating formulas are as follows,
 
diff --git a/mindspore/ops/_grad/grad_quant_ops.py b/mindspore/ops/_grad/grad_quant_ops.py
index 8809c7ac2fe..57c2829a2e0 100644
--- a/mindspore/ops/_grad/grad_quant_ops.py
+++ b/mindspore/ops/_grad/grad_quant_ops.py
@@ -138,6 +138,7 @@ def get_bprop_BatchNormFold(self):
 
 @bprop_getters.register(P.BNTrainingReduce)
 def get_bprop_BNTrainingReduce(self):
+    """Generate bprop for BNTrainingReduce for Ascend"""
     def bprop(x, out, dout):
         return (zeros_like(x),)