From 168bbd6d3c310f284a1c8ad397846bd845495890 Mon Sep 17 00:00:00 2001
From: juncaipeng <52520497+juncaipeng@users.noreply.github.com>
Date: Wed, 25 Sep 2019 16:48:00 +0800
Subject: [PATCH] Update L1Decay and L2Decay (#1367)

* modify L1DecayRegularizer, L1Decay, L2DecayRegularizer and L2Decay, test=develop
---
 .../regularizer_cn/L1DecayRegularizer_cn.rst  | 30 ++-----------------
 .../api_cn/regularizer_cn/L1Decay_cn.rst      |  2 --
 .../regularizer_cn/L2DecayRegularizer_cn.rst  | 29 ++----------------
 .../api_cn/regularizer_cn/L2Decay_cn.rst      |  2 --
 4 files changed, 4 insertions(+), 59 deletions(-)

diff --git a/doc/fluid/api_cn/regularizer_cn/L1DecayRegularizer_cn.rst b/doc/fluid/api_cn/regularizer_cn/L1DecayRegularizer_cn.rst
index 50d65fa4b..720676825 100644
--- a/doc/fluid/api_cn/regularizer_cn/L1DecayRegularizer_cn.rst
+++ b/doc/fluid/api_cn/regularizer_cn/L1DecayRegularizer_cn.rst
@@ -5,33 +5,7 @@ L1DecayRegularizer
 
 .. py:class:: paddle.fluid.regularizer.L1DecayRegularizer(regularization_coeff=0.0)
 
-L1DecayRegularizer实现L1权重衰减正则化，用于模型训练，使得权重矩阵稀疏。
+**注意：paddle.fluid.regularizer.L1DecayRegularizer是paddle.fluid.regularizer.L1Decay的别名，推荐使用paddle.fluid.regularizer.L1Decay。**
 
-具体实现中，L1权重衰减正则化的计算公式如下：
+详见 :ref:`cn_api_fluid_regularizer_L1Decay` 接口的使用文档。
 
-.. math::
-            \\L1WeightDecay=reg\_coeff∗sign(parameter)\\
-
-参数：
-  - **regularization_coeff** (float) – L1正则化系数，默认值为0.0。
-
-**代码示例**
-
-.. code-block:: python
-
-    import paddle.fluid as fluid
-
-    main_prog = fluid.Program()
-    startup_prog = fluid.Program()
-    with fluid.program_guard(main_prog, startup_prog):
-        data = fluid.layers.data(name='image', shape=[3, 28, 28], dtype='float32')
-        label = fluid.layers.data(name='label', shape=[1], dtype='int64')
-        hidden = fluid.layers.fc(input=data, size=128, act='relu')
-        prediction = fluid.layers.fc(input=hidden, size=10, act='softmax')
-        loss = fluid.layers.cross_entropy(input=prediction, label=label)
-        avg_loss = fluid.layers.mean(loss)
-    optimizer = fluid.optimizer.Adagrad(
-        learning_rate=1e-4,
-        regularization=fluid.regularizer.L1DecayRegularizer(
-            regularization_coeff=0.1))
-    optimizer.minimize(avg_loss)
diff --git a/doc/fluid/api_cn/regularizer_cn/L1Decay_cn.rst b/doc/fluid/api_cn/regularizer_cn/L1Decay_cn.rst
index ad0bf8869..9cf0766e5 100644
--- a/doc/fluid/api_cn/regularizer_cn/L1Decay_cn.rst
+++ b/doc/fluid/api_cn/regularizer_cn/L1Decay_cn.rst
@@ -6,8 +6,6 @@ L1Decay
 
 .. py:attribute::   paddle.fluid.regularizer.L1Decay(regularization_coeff=0.0)
 
-``L1Decay`` 是 ``L1DecayRegularizer`` 的别名。
-
 L1Decay实现L1权重衰减正则化，用于模型训练，使得权重矩阵稀疏。
 
 具体实现中，L1权重衰减正则化的计算公式如下：
diff --git a/doc/fluid/api_cn/regularizer_cn/L2DecayRegularizer_cn.rst b/doc/fluid/api_cn/regularizer_cn/L2DecayRegularizer_cn.rst
index f2961c26a..4c6d2a4d2 100644
--- a/doc/fluid/api_cn/regularizer_cn/L2DecayRegularizer_cn.rst
+++ b/doc/fluid/api_cn/regularizer_cn/L2DecayRegularizer_cn.rst
@@ -5,33 +5,8 @@ L2DecayRegularizer
 
 .. py:class:: paddle.fluid.regularizer.L2DecayRegularizer(regularization_coeff=0.0)
 
-L2DecayRegularizer实现L2权重衰减正则化，用于模型训练，有助于防止模型对训练数据过拟合。
+**注意：paddle.fluid.regularizer.L2DecayRegularizer是paddle.fluid.regularizer.L2Decay的别名，推荐使用paddle.fluid.regularizer.L2Decay。**
 
-具体实现中，L2权重衰减正则化的计算公式如下：
+详见 :ref:`cn_api_fluid_regularizer_L2Decay` 接口的使用文档。
 
-.. math::
-            \\L2WeightDecay=reg\_coeff*parameter\\
 
-参数:
-  - **regularization_coeff** (float) – 正则化系数，默认值为0.0。
-  
-**代码示例**
-
-.. code-block:: python
-
-    import paddle.fluid as fluid
-
-    main_prog = fluid.Program()
-    startup_prog = fluid.Program()
-    with fluid.program_guard(main_prog, startup_prog):
-        data = fluid.layers.data(name='image', shape=[3, 28, 28], dtype='float32')
-        label = fluid.layers.data(name='label', shape=[1], dtype='int64')
-        hidden = fluid.layers.fc(input=data, size=128, act='relu')
-        prediction = fluid.layers.fc(input=hidden, size=10, act='softmax')
-        loss = fluid.layers.cross_entropy(input=prediction, label=label)
-        avg_loss = fluid.layers.mean(loss)
-    optimizer = fluid.optimizer.Adagrad(
-        learning_rate=1e-4,
-        regularization=fluid.regularizer.L2DecayRegularizer(
-            regularization_coeff=0.1))
-    optimizer.minimize(avg_loss)
diff --git a/doc/fluid/api_cn/regularizer_cn/L2Decay_cn.rst b/doc/fluid/api_cn/regularizer_cn/L2Decay_cn.rst
index 7966e36c6..1de52411d 100644
--- a/doc/fluid/api_cn/regularizer_cn/L2Decay_cn.rst
+++ b/doc/fluid/api_cn/regularizer_cn/L2Decay_cn.rst
@@ -5,8 +5,6 @@ L2Decay
 
 .. py:attribute::   paddle.fluid.regularizer.L2Decay
 
-``L2Decay`` 是 ``L2DecayRegularizer`` 的别名。
-
 L2Decay实现L2权重衰减正则化，用于模型训练，有助于防止模型对训练数据过拟合。
 
 具体实现中，L2权重衰减正则化的计算公式如下：
-- 
GitLab