Merge pull request #3845 from lcy-seso/rename_mse_to_square_error

rename mse_cost into square_error_cost.

doc/api/v2/config/layer.rst

@@ -434,9 +434,9 @@ lambda_cost
 ..  autoclass:: paddle.v2.layer.lambda_cost
     :noindex:
 
-mse_cost
+square_error_cost
 --------
-..  autoclass:: paddle.v2.layer.mse_cost
+..  autoclass:: paddle.v2.layer.square_error_cost
     :noindex:
 
 rank_cost

doc/getstarted/basic_usage/index_cn.rst

@@ -55,7 +55,7 @@ PaddlePaddle是源于百度的一个深度学习平台。这份简短的介绍
     # 线性计算网络层: ȳ = wx + b
     ȳ = fc_layer(input=x, param_attr=ParamAttr(name='w'), size=1, act=LinearActivation(), bias_attr=ParamAttr(name='b'))
     # 计算误差函数，即  ȳ 和真实 y 之间的距离
-    cost = mse_cost(input= ȳ, label=y)
+    cost = square_error_cost(input= ȳ, label=y)
     outputs(cost)
 
 
@@ -69,7 +69,7 @@ PaddlePaddle是源于百度的一个深度学习平台。这份简短的介绍
     
     - **数据层**：数据层 `data_layer` 是神经网络的入口，它读入数据并将它们传输到接下来的网络层。这里数据层有两个，分别对应于变量 `x` 和 `y`。
     - **全连接层**：全连接层 `fc_layer` 是基础的计算单元，这里利用它建模变量之间的线性关系。计算单元是神经网络的核心，PaddlePaddle支持大量的计算单元和任意深度的网络连接，从而可以拟合任意的函数来学习复杂的数据关系。
-    - **回归误差代价层**：回归误差代价层 `mse_cost` 是众多误差代价函数层的一种，它们在训练过程作为网络的出口，用来计算模型的误差，是模型参数优化的目标函数。
+    - **回归误差代价层**：回归误差代价层 `square_error_cost` 是众多误差代价函数层的一种，它们在训练过程作为网络的出口，用来计算模型的误差，是模型参数优化的目标函数。
 
 定义了网络结构并保存为 `trainer_config.py` 之后，运行以下训练命令：
 

doc/getstarted/basic_usage/index_en.rst

@@ -49,7 +49,7 @@ To recover this relationship between ``X`` and ``Y``, we use a neural network wi
         x = data_layer(name='x', size=1)
         y = data_layer(name='y', size=1)
         y_predict = fc_layer(input=x, param_attr=ParamAttr(name='w'), size=1, act=LinearActivation(), bias_attr=ParamAttr(name='b'))
-        cost = mse_cost(input=y_predict, label=y)
+        cost = square_error_cost(input=y_predict, label=y)
         outputs(cost)
 
 Some of the most fundamental usages of PaddlePaddle are demonstrated:

doc/getstarted/concepts/src/train.py

@@ -8,7 +8,7 @@ paddle.init(use_gpu=False)
 x = paddle.layer.data(name='x', type=paddle.data_type.dense_vector(2))
 y_predict = paddle.layer.fc(input=x, size=1, act=paddle.activation.Linear())
 y = paddle.layer.data(name='y', type=paddle.data_type.dense_vector(1))
-cost = paddle.layer.mse_cost(input=y_predict, label=y)
+cost = paddle.layer.square_error_cost(input=y_predict, label=y)
 
 # create parameters
 parameters = paddle.parameters.create(cost)

doc/getstarted/concepts/use_concepts_cn.rst

@@ -81,9 +81,9 @@ PaddlePaddle支持不同类型的输入数据，主要包括四种类型，和
 ..	code-block:: bash
 
     y_predict = paddle.layer.fc(input=x, size=1, act=paddle.activation.Linear())
-    cost = paddle.layer.mse_cost(input=y_predict, label=y)
+    cost = paddle.layer.square_error_cost(input=y_predict, label=y)
 
-其中，x与y为之前描述的输入层；而y_predict是接收x作为输入，接上一个全连接层；cost接收y_predict与y作为输入，接上均方误差层。
+其中，x与y为之前描述的输入层；而y_predict是接收x作为输入，接上一个全连接层；cost接收y_predict与y作为输入，接上平方误差层。
 
 最后一层cost中记录了神经网络的所有拓扑结构，通过组合不同的layer，我们即可完成神经网络的搭建。
 
@@ -147,4 +147,4 @@ PaddlePaddle支持不同类型的输入数据，主要包括四种类型，和
 ..  literalinclude:: src/train.py
     :linenos:
 
-有关线性回归的实际应用，可以参考PaddlePaddle book的 `第一章节 <http://book.paddlepaddle.org/index.html>`_。
\ No newline at end of file
+有关线性回归的实际应用，可以参考PaddlePaddle book的 `第一章节 <http://book.paddlepaddle.org/index.html>`_。

doc/howto/usage/k8s/k8s_distributed_cn.md

@@ -213,7 +213,7 @@ I1116 09:10:17.123440    50 Util.cpp:130] Calling runInitFunctions
 I1116 09:10:17.123764    50 Util.cpp:143] Call runInitFunctions done.
 [WARNING 2016-11-16 09:10:17,227 default_decorators.py:40] please use keyword arguments in paddle config.
 [INFO 2016-11-16 09:10:17,239 networks.py:1282] The input order is [movie_id, title, genres, user_id, gender, age, occupation, rating]
-[INFO 2016-11-16 09:10:17,239 networks.py:1289] The output order is [__mse_cost_0__]
+[INFO 2016-11-16 09:10:17,239 networks.py:1289] The output order is [__square_error_cost_0__]
 I1116 09:10:17.392917    50 Trainer.cpp:170] trainer mode: Normal
 I1116 09:10:17.613910    50 PyDataProvider2.cpp:257] loading dataprovider dataprovider::process
 I1116 09:10:17.680917    50 PyDataProvider2.cpp:257] loading dataprovider dataprovider::process

python/paddle/trainer_config_helpers/layers.py

@@ -53,7 +53,7 @@ __all__ = [
     'cos_sim',
     'hsigmoid',
     'conv_projection',
-    'mse_cost',
+    'square_error_cost',
     'regression_cost',
     'classification_cost',
     'LayerOutput',
@@ -4238,13 +4238,18 @@ def __cost_input__(input, label, weight=None):
 
 @wrap_name_default()
 @layer_support()
-def mse_cost(input, label, weight=None, name=None, coeff=1.0, layer_attr=None):
+def square_error_cost(input,
+                      label,
+                      weight=None,
+                      name=None,
+                      coeff=1.0,
+                      layer_attr=None):
     """
-    mean squared error cost:
+    sum of square error cost:
 
     ..  math::
 
-        \\frac{1}{N}\sum_{i=1}^N(t_i-y_i)^2
+        cost = \\sum_{i=1}^N(t_i-y_i)^2
 
     :param name: layer name.
     :type name: basestring
@@ -4273,7 +4278,7 @@ def mse_cost(input, label, weight=None, name=None, coeff=1.0, layer_attr=None):
     return LayerOutput(name, LayerType.COST, parents=parents, size=1)
 
 
-regression_cost = mse_cost
+regression_cost = square_error_cost
 
 
 @wrap_name_default("cost")
@@ -5798,9 +5803,9 @@ def huber_regression_cost(input,
                           coeff=1.0,
                           layer_attr=None):
     """
-    In statistics, the Huber loss is a loss function used in robust regression, 
-    that is less sensitive to outliers in data than the squared error loss. 
-    Given a prediction f(x), a label y and :math:`\delta`, the loss function 
+    In statistics, the Huber loss is a loss function used in robust regression,
+    that is less sensitive to outliers in data than the squared error loss.
+    Given a prediction f(x), a label y and :math:`\delta`, the loss function
     is defined as:
 
     .. math:
@@ -5848,13 +5853,13 @@ def huber_classification_cost(input,
                               coeff=1.0,
                               layer_attr=None):
     """
-    For classification purposes, a variant of the Huber loss called modified Huber 
-    is sometimes used. Given a prediction f(x) (a real-valued classifier score) and 
-    a true binary class label :math:`y\in \left \{-1, 1 \right \}`, the modified Huber 
+    For classification purposes, a variant of the Huber loss called modified Huber
+    is sometimes used. Given a prediction f(x) (a real-valued classifier score) and
+    a true binary class label :math:`y\in \left \{-1, 1 \right \}`, the modified Huber
     loss is defined as:
 
     .. math:
-       loss = \max \left ( 0, 1-yf(x) \right )^2, yf(x)\geq 1 
+       loss = \max \left ( 0, 1-yf(x) \right )^2, yf(x)\geq 1
        loss = -4yf(x), \text{otherwise}
 
     The example usage is:

python/paddle/trainer_config_helpers/tests/configs/protostr/test_cost_layers_with_weight.protostr

@@ -45,7 +45,7 @@ layers {
   coeff: 1.0
 }
 layers {
-  name: "__mse_cost_0__"
+  name: "__square_error_cost_0__"
   type: "square_error"
   size: 1
   active_type: ""
@@ -130,7 +130,7 @@ input_layer_names: "label"
 input_layer_names: "weight"
 input_layer_names: "multi_class_label"
 output_layer_names: "__cost_0__"
-output_layer_names: "__mse_cost_0__"
+output_layer_names: "__square_error_cost_0__"
 output_layer_names: "__nce_layer_0__"
 evaluators {
   name: "classification_error_evaluator"
@@ -146,7 +146,7 @@ sub_models {
   layer_names: "weight"
   layer_names: "__fc_layer_0__"
   layer_names: "__cost_0__"
-  layer_names: "__mse_cost_0__"
+  layer_names: "__square_error_cost_0__"
   layer_names: "multi_class_label"
   layer_names: "__nce_layer_0__"
   input_layer_names: "input"
@@ -154,7 +154,7 @@ sub_models {
   input_layer_names: "weight"
   input_layer_names: "multi_class_label"
   output_layer_names: "__cost_0__"
-  output_layer_names: "__mse_cost_0__"
+  output_layer_names: "__square_error_cost_0__"
   output_layer_names: "__nce_layer_0__"
   evaluator_names: "classification_error_evaluator"
   is_recurrent_layer_group: false

python/paddle/trainer_config_helpers/tests/configs/test_cost_layers_with_weight.py

@@ -10,7 +10,7 @@ fc = fc_layer(input=data, size=10, act=SoftmaxActivation())
 outputs(
     classification_cost(
         input=fc, label=lbl, weight=wt),
-    mse_cost(
+    square_error_cost(
         input=fc, label=lbl, weight=wt),
     nce_layer(
         input=fc,

python/paddle/v2/tests/test_layer.py

@@ -134,8 +134,9 @@ class CostLayerTest(unittest.TestCase):
         cost3 = layer.cross_entropy_cost(input=inference, label=label)
         cost4 = layer.cross_entropy_with_selfnorm_cost(
             input=inference, label=label)
-        cost5 = layer.mse_cost(input=inference, label=label)
-        cost6 = layer.mse_cost(input=inference, label=label, weight=weight)
+        cost5 = layer.square_error_cost(input=inference, label=label)
+        cost6 = layer.square_error_cost(
+            input=inference, label=label, weight=weight)
         cost7 = layer.multi_binary_label_cross_entropy_cost(
             input=inference, label=label)
         cost8 = layer.rank_cost(left=score, right=score, label=score)