Upgrade string literals to raw string (#28989)

* upgrade comment string to raw string

* fix string in

* fix string with ' '

* revert update on comments

* upgrade only necessary

* fix sample code checker

* fix comments with '''

paddle/scripts/conda_build.py

@@ -1,4 +1,19 @@
 #!/bin/python
+
+# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+# 
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+# 
+#     http://www.apache.org/licenses/LICENSE-2.0
+# 
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 #
 import platform
 from sys import argv
@@ -120,7 +135,7 @@ python setup.py install
         self.py_str = ["py27", "py35", "py36", "py37"]
         self.pip_end = ".whl --no-deps"
         self.pip_prefix_linux = "pip install /package/paddlepaddle"
-        self.pip_prefix_windows = "pip install C:\package\paddlepaddle"
+        self.pip_prefix_windows = r"pip install C:\package\paddlepaddle"
         self.pip_gpu = "_gpu-"
         self.pip_cpu = "-"
         self.mac_pip = [
@@ -216,7 +231,7 @@ package:
     - matplotlib"""
     if not (cuda_str == None):
         meta_str = meta_str + cuda_str
-    
+
     blt_str = var.blt_const + blt_var
     if (python_str == var.python27):
         blt_str = blt_str + """
@@ -224,7 +239,7 @@ package:
     else:
         meta_str = meta_str + """
     - opencv>=3.4.2"""
-    
+
     meta_str = meta_str + var.test + var.about
     meta_filename = "meta.yaml"
     build_filename = "bld.bat"

python/paddle/dataset/imdb.py

@@ -116,8 +116,8 @@ def train(word_idx):
     :rtype: callable
     """
     return reader_creator(
-        re.compile("aclImdb/train/pos/.*\.txt$"),
-        re.compile("aclImdb/train/neg/.*\.txt$"), word_idx)
+        re.compile(r"aclImdb/train/pos/.*\.txt$"),
+        re.compile(r"aclImdb/train/neg/.*\.txt$"), word_idx)
 
 
 @deprecated(
@@ -137,8 +137,8 @@ def test(word_idx):
     :rtype: callable
     """
     return reader_creator(
-        re.compile("aclImdb/test/pos/.*\.txt$"),
-        re.compile("aclImdb/test/neg/.*\.txt$"), word_idx)
+        re.compile(r"aclImdb/test/pos/.*\.txt$"),
+        re.compile(r"aclImdb/test/neg/.*\.txt$"), word_idx)
 
 
 @deprecated(
@@ -153,7 +153,7 @@ def word_dict():
     :rtype: dict
     """
     return build_dict(
-        re.compile("aclImdb/((train)|(test))/((pos)|(neg))/.*\.txt$"), 150)
+        re.compile(r"aclImdb/((train)|(test))/((pos)|(neg))/.*\.txt$"), 150)
 
 
 @deprecated(

python/paddle/dataset/tests/imdb_test.py

@@ -18,13 +18,13 @@ import paddle.dataset.imdb
 import unittest
 import re
 
-TRAIN_POS_PATTERN = re.compile("aclImdb/train/pos/.*\.txt$")
-TRAIN_NEG_PATTERN = re.compile("aclImdb/train/neg/.*\.txt$")
-TRAIN_PATTERN = re.compile("aclImdb/train/.*\.txt$")
+TRAIN_POS_PATTERN = re.compile(r"aclImdb/train/pos/.*\.txt$")
+TRAIN_NEG_PATTERN = re.compile(r"aclImdb/train/neg/.*\.txt$")
+TRAIN_PATTERN = re.compile(r"aclImdb/train/.*\.txt$")
 
-TEST_POS_PATTERN = re.compile("aclImdb/test/pos/.*\.txt$")
-TEST_NEG_PATTERN = re.compile("aclImdb/test/neg/.*\.txt$")
-TEST_PATTERN = re.compile("aclImdb/test/.*\.txt$")
+TEST_POS_PATTERN = re.compile(r"aclImdb/test/pos/.*\.txt$")
+TEST_NEG_PATTERN = re.compile(r"aclImdb/test/neg/.*\.txt$")
+TEST_PATTERN = re.compile(r"aclImdb/test/.*\.txt$")
 
 
 class TestIMDB(unittest.TestCase):

python/paddle/distributed/fleet/base/distributed_strategy.py

@@ -862,7 +862,7 @@ class DistributedStrategy(object):
 
     @property
     def dgc_configs(self):
-        """
+        r"""
         Set Deep Gradient Compression training configurations. In general, dgc has serveral configurable
         settings that can be configured through a dict.
 

python/paddle/distributed/fleet/launch.py

@@ -11,7 +11,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""
+r"""
 fleetrun is a module that spawns multiple distributed
 process on each training node for gpu training and cpu training.
 Usage:

python/paddle/distributed/fleet/meta_optimizers/parameter_server_optimizer.py

@@ -158,13 +158,13 @@ class ParameterServerOptimizer(MetaOptimizerBase):
                     ['vm_stat'], stdout=subprocess.PIPE).communicate()[0]
                 # Process vm_stat
                 vmLines = vm.split('\n')
-                sep = re.compile(':[\s]+')
+                sep = re.compile(r':[\s]+')
                 vmStats = {}
                 for row in range(1, len(vmLines) - 2):
                     rowText = vmLines[row].strip()
                     rowElements = sep.split(rowText)
                     vmStats[(rowElements[0]
-                             )] = int(rowElements[1].strip('\.')) * 4096
+                             )] = int(rowElements[1].strip(r'\.')) * 4096
                 return vmStats["Pages free"]
             elif platform.system() == "Linux":
                 mems = {}

python/paddle/distributed/launch.py

@@ -11,7 +11,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""
+r"""
 paddle.distributed.launch is a module that spawns multiple distributed 
 process on each training node for gpu training.
 Usage:

python/paddle/distribution.py

@@ -166,7 +166,7 @@ class Distribution(object):
 
 
 class Uniform(Distribution):
-    """Uniform distribution with `low` and `high` parameters.
+    r"""Uniform distribution with `low` and `high` parameters.
 
     Mathematical Details
 
@@ -374,7 +374,7 @@ class Uniform(Distribution):
         return elementwise_div((lb * ub), (self.high - self.low), name=name)
 
     def entropy(self):
-        """Shannon entropy in nats.
+        r"""Shannon entropy in nats.
 
         The entropy is
 
@@ -391,7 +391,7 @@ class Uniform(Distribution):
 
 
 class Normal(Distribution):
-    """The Normal distribution with location `loc` and `scale` parameters.
+    r"""The Normal distribution with location `loc` and `scale` parameters.
 
     Mathematical details
 
@@ -534,7 +534,7 @@ class Normal(Distribution):
                 return output
 
     def entropy(self):
-        """Shannon entropy in nats.
+        r"""Shannon entropy in nats.
 
         The entropy is
 
@@ -599,7 +599,7 @@ class Normal(Distribution):
             name=name)
 
     def kl_divergence(self, other):
-        """The KL-divergence between two normal distributions.
+        r"""The KL-divergence between two normal distributions.
 
         The probability density function (pdf) is
 
@@ -644,7 +644,7 @@ class Normal(Distribution):
 
 
 class Categorical(Distribution):
-    """
+    r"""
     Categorical distribution is a discrete probability distribution that 
     describes the possible results of a random variable that can take on 
     one of K possible categories, with the probability of each category 

python/paddle/fluid/clip.py

@@ -40,7 +40,7 @@ class BaseErrorClipAttr(object):
 
 
 class ErrorClipByValue(BaseErrorClipAttr):
-    """
+    r"""
     Clips tensor values to the range [min, max].
 
     Given a tensor ``t`` (see Examples below), this operation clips its value \
@@ -241,7 +241,7 @@ class ClipGradByValue(ClipGradBase):
 
 
 class ClipGradByNorm(ClipGradBase):
-    """
+    r"""
     Limit the l2 norm of multi-dimensional Tensor :math:`X` to ``clip_norm`` .
     
     - If the l2 norm of :math:`X` is greater than ``clip_norm`` , :math:`X` will be compressed by a ratio.
@@ -343,7 +343,7 @@ class ClipGradByNorm(ClipGradBase):
 
 
 class ClipGradByGlobalNorm(ClipGradBase):
-    """
+    r"""
     Given a list of Tensor :math:`t\_list` , calculate the global norm for the elements of all tensors in 
     :math:`t\_list` , and limit it to ``clip_norm`` .
     

python/paddle/fluid/contrib/layers/nn.py

@@ -137,7 +137,7 @@ def var_conv_2d(input,
                 act=None,
                 dtype='float32',
                 name=None):
-    """
+    r"""
     The var_conv_2d layer calculates the output base on the :attr:`input` with variable length,
     row, col, input channel, filter size and strides. Both :attr:`input`, :attr:`row`,
     and :attr:`col` are 1-level LodTensor. The convolution operation is same as conv2d layer with
@@ -477,7 +477,7 @@ def fused_embedding_seq_pool(input,
                              combiner='sum',
                              param_attr=None,
                              dtype='float32'):
-    """
+    r"""
     **Embedding Sequence pool**
 
     This layer is the fusion of lookup table and sequence_pool.
@@ -1442,7 +1442,7 @@ def batch_fc(input, param_size, param_attr, bias_size, bias_attr, act=None):
 
 
 def _pull_box_extended_sparse(input, size, extend_size=64, dtype='float32'):
-    """
+    r"""
     **Pull Box Extended Sparse Layer**
     This layer is used to lookup embeddings of IDs, provided by :attr:`input`, in
     BoxPS lookup table. The result of this lookup is the embedding of each ID in the
@@ -1640,7 +1640,7 @@ def fused_bn_add_act(x,
                      moving_variance_name=None,
                      act=None,
                      name=None):
-    """
+    r"""
     This Op performs batch norm on input x, and adds the result to input y. Then
     it performs activation on the sum. The data format of inputs must be NHWC
     `[batch, in_height, in_width, in_channels]`.

python/paddle/fluid/contrib/layers/rnn_impl.py

@@ -175,7 +175,7 @@ def basic_gru(input,
               activation=None,
               dtype='float32',
               name='basic_gru'):
-    """
+    r"""
     GRU implementation using basic operator, supports multiple layers and bidirectional gru.
 
     .. math::
@@ -418,7 +418,7 @@ def basic_lstm(input,
                forget_bias=1.0,
                dtype='float32',
                name='basic_lstm'):
-    """
+    r"""
     LSTM implementation using basic operators, supports multiple layers and bidirectional LSTM.
 
     .. math::
@@ -697,7 +697,7 @@ def basic_lstm(input,
 
 
 class BasicLSTMUnit(Layer):
-    """
+    r"""
     ****
     BasicLSTMUnit class, Using basic operator to build LSTM
     The algorithm can be described as the code below.

python/paddle/fluid/contrib/memory_usage_calc.py

@@ -44,7 +44,7 @@ DEBUG = False
 
 
 def memory_usage(program, batch_size):
-    """
+    r"""
     Get the estimate memory usage of program with input batch size.
 
     Args:

python/paddle/fluid/contrib/slim/quantization/imperative/qat.py

@@ -64,7 +64,7 @@ class ImperativeQuantAware(object):
                  act_preprocess_layer=None,
                  weight_quantize_layer=None,
                  act_quantize_layer=None):
-        """
+        r"""
         The constructor for ImperativeQuantAware.
 
         Args:

python/paddle/fluid/contrib/slim/quantization/imperative/quant_nn.py

@@ -30,7 +30,7 @@ __all__ = [
 
 
 class FakeQuantMovingAverage(layers.Layer):
-    """
+    r"""
     FakeQuantMovingAverage layer does the moving_average_abs_max quant and then dequant.
     Its computational formula is described as below:
 
@@ -128,7 +128,7 @@ class FakeQuantMovingAverage(layers.Layer):
 
 
 class FakeQuantAbsMax(layers.Layer):
-    """
+    r"""
     FakeQuantAbsMax layer does the abs_max quant and then dequant.
     Its computational formula is described as below:
 
@@ -545,7 +545,7 @@ class QuantizedLinear(layers.Layer):
 
 class MovingAverageAbsMaxScale(layers.Layer):
     def __init__(self, name=None, moving_rate=0.9, dtype='float32'):
-        """
+        r"""
         MovingAverageMaxScale layer is used to calculating the output quantization scale of Layer.
         Its computational formula is described as below:
 

python/paddle/fluid/contrib/slim/quantization/quant_int8_mkldnn_pass.py

@@ -37,7 +37,7 @@ class QuantInt8MkldnnPass(object):
     """
 
     def __init__(self, _scope=None, _place=None):
-        """
+        r"""
         Args:
             scope(fluid.Scope): scope is used to initialize the new parameters.
             place(fluid.CPUPlace): place is used to initialize the new parameters.

python/paddle/fluid/contrib/slim/quantization/quantization_pass.py

@@ -239,7 +239,7 @@ class QuantizationTransformPass(object):
                  act_preprocess_func=None,
                  optimizer_func=None,
                  executor=None):
-        """
+        r"""
         Constructor.
 
         Args:

python/paddle/fluid/contrib/utils/hdfs_utils.py

@@ -33,7 +33,7 @@ _logger = get_logger(
 
 
 class HDFSClient(object):
-    """
+    r"""
     A tool of HDFS 
 
     Args:
@@ -376,7 +376,7 @@ class HDFSClient(object):
             _logger.info("HDFS list path: {} successfully".format(hdfs_path))
 
             ret_lines = []
-            regex = re.compile('\s+')
+            regex = re.compile(r'\s+')
             out_lines = output.strip().split("\n")
             for line in out_lines:
                 re_line = regex.split(line)
@@ -418,7 +418,7 @@ class HDFSClient(object):
             _logger.info("HDFS list all files: {} successfully".format(
                 hdfs_path))
             lines = []
-            regex = re.compile('\s+')
+            regex = re.compile(r'\s+')
             out_lines = output.strip().split("\n")
             for line in out_lines:
                 re_line = regex.split(line)

python/paddle/fluid/core.py

@@ -224,7 +224,7 @@ def less_than_ver(a, b):
     import operator
 
     def to_list(s):
-        s = re.sub('(\.0+)+$', '', s)
+        s = re.sub(r'(\.0+)+$', '', s)
         return [int(x) for x in s.split('.')]
 
     return operator.lt(to_list(a), to_list(b))

python/paddle/fluid/dataloader/dataloader_iter.py

@@ -101,10 +101,11 @@ class _DatasetKind(object):
     ITER = 1
 
     @staticmethod
-    def create_fetcher(kind, dataset, auto_collate_batch, collate_fn, drop_last):
+    def create_fetcher(kind, dataset, auto_collate_batch, collate_fn,
+                       drop_last):
         if kind == _DatasetKind.MAP:
-            return _MapDatasetFetcher(dataset, auto_collate_batch,
-                                      collate_fn, drop_last)
+            return _MapDatasetFetcher(dataset, auto_collate_batch, collate_fn,
+                                      drop_last)
         elif kind == _DatasetKind.ITER:
             return _IterableDatasetFetcher(dataset, auto_collate_batch,
                                            collate_fn, drop_last)
@@ -240,7 +241,8 @@ class _DataLoaderIterBase(object):
             if self._dataset_kind == _DatasetKind.MAP:
                 self._sampler_iter = iter(list(range(len(self._dataset))))
             else:
-                self._sampler_iter = iter(_InfiniteIterableSampler(self._dataset, 1))
+                self._sampler_iter = iter(
+                    _InfiniteIterableSampler(self._dataset, 1))
             self._collate_fn = loader.collate_fn
 
         # LoDTensorBlockingQueue instance for create_py_reader and a thread
@@ -380,8 +382,8 @@ class _DataLoaderIterSingleProcess(_DataLoaderIterBase):
 
 # NOTE(chenweihang): _worker_loop must be top level method to be pickled
 def _worker_loop(dataset, dataset_kind, indices_queue, out_queue, done_event,
-                 auto_collate_batch, collate_fn, init_fn, worker_id, num_workers,
-                 use_shared_memory):
+                 auto_collate_batch, collate_fn, init_fn, worker_id,
+                 num_workers, use_shared_memory):
     try:
         # NOTE: [ mmap files clear ] When the child process exits unexpectedly,
         # some shared memory objects may have been applied for but have not yet
@@ -400,8 +402,8 @@ def _worker_loop(dataset, dataset_kind, indices_queue, out_queue, done_event,
         try:
             if init_fn is not None:
                 init_fn(worker_id)
-            fetcher = _DatasetKind.create_fetcher(dataset_kind, dataset,
-                                    auto_collate_batch, collate_fn, True)
+            fetcher = _DatasetKind.create_fetcher(
+                dataset_kind, dataset, auto_collate_batch, collate_fn, True)
         except:
             init_exception = Exception("init_fn failed in worker {}: " \
                                     "{}".format(worker_id, sys.exc_info()))

python/paddle/fluid/distributed/downpour.py

@@ -22,7 +22,7 @@ from google.protobuf import text_format
 
 
 class DownpourSGD(object):
-    """
+    r"""
     Distributed optimizer of downpour stochastic gradient descent
     Standard implementation of Google's Downpour SGD
     in Large Scale Distributed Deep Networks

python/paddle/fluid/distributed/node.py

@@ -52,7 +52,7 @@ class DownpourServer(Server):
 
     def add_sparse_table(self, table_id, learning_rate, slot_key_vars,
                          slot_value_var):
-        """
+        r"""
         Args:
             table_id(int): id of sparse params table
             learning_rate(float): the learning rate used to update parameters. \
@@ -84,7 +84,7 @@ class DownpourServer(Server):
         table.accessor.downpour_accessor_param.delete_threshold = 0.8
 
     def add_dense_table(self, table_id, learning_rate, param_var, grad_var):
-        """
+        r"""
         Args:
             table_id(int): id of sparse params table
             learning_rate(float): the learning rate used to update parameters. \
@@ -135,7 +135,7 @@ class DownpourWorker(Worker):
 
     def add_sparse_table(self, table_id, learning_rate, slot_key_vars,
                          slot_value_vars):
-        """
+        r"""
         Args:
             table_id(int): id of sparse params table
             learning_rate(float): the learning rate used to update parameters. \
@@ -153,7 +153,7 @@ class DownpourWorker(Worker):
             [var.name + "@GRAD" for var in slot_value_vars])
 
     def add_dense_table(self, table_id, learning_rate, param_vars, grad_vars):
-        """
+        r"""
         Args:
             table_id(int): id of sparse params table
             learning_rate(float): the learning rate used to update parameters. \

python/paddle/fluid/dygraph/base.py

@@ -593,7 +593,7 @@ def grad(outputs,
 
 @framework.dygraph_only
 def to_variable(value, name=None, zero_copy=None, dtype=None):
-    """
+    r"""
     :api_attr: imperative
 
     The API will create a ``Variable`` or ``ComplexVariable`` object from 

python/paddle/fluid/dygraph/learning_rate_scheduler.py

@@ -183,7 +183,7 @@ class PiecewiseDecay(LearningRateDecay):
 
 
 class NaturalExpDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies natural exponential decay to the initial learning rate.
@@ -266,7 +266,7 @@ class NaturalExpDecay(LearningRateDecay):
 
 
 class ExponentialDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies exponential decay to the learning rate.
@@ -348,7 +348,7 @@ class ExponentialDecay(LearningRateDecay):
 
 
 class InverseTimeDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies inverse time decay to the initial learning rate.
@@ -426,7 +426,7 @@ class InverseTimeDecay(LearningRateDecay):
 
 
 class PolynomialDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies polynomial decay to the initial learning rate.
@@ -520,7 +520,7 @@ class PolynomialDecay(LearningRateDecay):
 
 
 class CosineDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies cosine decay to the learning rate.
@@ -578,7 +578,7 @@ class CosineDecay(LearningRateDecay):
 
 
 class NoamDecay(LearningRateDecay):
-    """
+    r"""
     :api_attr: imperative
 
     Applies Noam decay to the initial learning rate. 

python/paddle/fluid/dygraph/nn.py

@@ -42,7 +42,7 @@ __all__ = [
 
 
 class Conv2D(layers.Layer):
-    """
+    r"""
     This interface is used to construct a callable object of the ``Conv2D`` class.
     For more details, refer to code examples.
     The convolution2D layer calculates the output based on the input, filter
@@ -282,7 +282,7 @@ class Conv2D(layers.Layer):
 
 
 class Conv3D(layers.Layer):
-    """
+    r"""
     **Convlution3D Layer**
 
     The convolution3D layer calculates the output based on the input, filter
@@ -484,7 +484,7 @@ class Conv3D(layers.Layer):
 
 
 class Conv3DTranspose(layers.Layer):
-    """
+    r"""
     **Convlution3D transpose layer**
 
     The convolution3D transpose layer calculates the output based on the input,
@@ -701,7 +701,7 @@ class Conv3DTranspose(layers.Layer):
 
 
 class Pool2D(layers.Layer):
-    """
+    r"""
 
     This interface is used to construct a callable object of the ``Pool2D`` class.
     For more details, refer to code examples.
@@ -1009,7 +1009,7 @@ class Linear(layers.Layer):
 
 
 class InstanceNorm(layers.Layer):
-    """
+    r"""
     This interface is used to construct a callable object of the ``InstanceNorm`` class.
     For more details, refer to code examples.
 
@@ -1143,7 +1143,7 @@ class InstanceNorm(layers.Layer):
 
 
 class BatchNorm(layers.Layer):
-    """
+    r"""
     :alias_main: paddle.nn.BatchNorm
 	:alias: paddle.nn.BatchNorm,paddle.nn.layer.BatchNorm,paddle.nn.layer.norm.BatchNorm
 	:old_api: paddle.fluid.dygraph.BatchNorm
@@ -1492,7 +1492,7 @@ class Dropout(layers.Layer):
 
 
 class Embedding(layers.Layer):
-    """
+    r"""
     :alias_main: paddle.nn.Embedding
 	:alias: paddle.nn.Embedding,paddle.nn.layer.Embedding,paddle.nn.layer.common.Embedding
 	:old_api: paddle.fluid.dygraph.Embedding
@@ -1652,7 +1652,7 @@ class Embedding(layers.Layer):
 
 
 class LayerNorm(layers.Layer):
-    """
+    r"""
     :alias_main: paddle.nn.LayerNorm
 	:alias: paddle.nn.LayerNorm,paddle.nn.layer.LayerNorm,paddle.nn.layer.norm.LayerNorm
 	:old_api: paddle.fluid.dygraph.LayerNorm
@@ -2242,7 +2242,7 @@ class NCE(layers.Layer):
 
 
 class PRelu(layers.Layer):
-    """
+    r"""
     This interface is used to construct a callable object of the ``PRelu`` class.
     For more details, refer to code examples.
     It implements three activation methods of the ``PRelu`` activation function.
@@ -2350,7 +2350,7 @@ class PRelu(layers.Layer):
 
 
 class BilinearTensorProduct(layers.Layer):
-    """
+    r"""
 
     **Add Bilinear Tensor Product Layer**
 
@@ -2467,7 +2467,7 @@ class BilinearTensorProduct(layers.Layer):
 
 
 class Conv2DTranspose(layers.Layer):
-    """
+    r"""
     This interface is used to construct a callable object of the ``Conv2DTranspose`` class.
     For more details, refer to code examples.
     The convolution2D transpose layer calculates the output based on the input,
@@ -2979,7 +2979,7 @@ class GroupNorm(layers.Layer):
 
 
 class SpectralNorm(layers.Layer):
-    """
+    r"""
     :alias_main: paddle.nn.SpectralNorm
 	:alias: paddle.nn.SpectralNorm,paddle.nn.layer.SpectralNorm,paddle.nn.layer.norm.SpectralNorm
 	:old_api: paddle.fluid.dygraph.SpectralNorm

python/paddle/fluid/dygraph/rnn.py

@@ -20,7 +20,7 @@ __all__ = ['LSTMCell', 'GRUCell']
 
 
 class LSTMCell(Layer):
-    """
+    r"""
     LSTMCell implementation using basic operators.
     There are two LSTMCell version, the default one is compatible with CUDNN LSTM implementation.
     The algorithm can be described as the equations below.
@@ -236,7 +236,7 @@ class LSTMCell(Layer):
 
 
 class GRUCell(Layer):
-    """
+    r"""
     GRU implementation using basic operators.
     There are two GRUCell version, the default one is compatible with CUDNN GRU implementation.
     The algorithm can be described as the equations below.