diff --git a/python/paddle/dataset/conll05.py b/python/paddle/dataset/conll05.py
index 25623feabb89617e4272789a8ddd11da3409811c..724202b95694c727ebe5408f37fabd69d0c65b75 100644
--- a/python/paddle/dataset/conll05.py
+++ b/python/paddle/dataset/conll05.py
@@ -24,7 +24,7 @@ import tarfile
 import gzip
 import itertools
 import paddle.dataset.common
-from six.moves import zip
+from six.moves import zip, range
 
 __all__ = ['test, get_dict', 'get_embedding', 'convert']
 
diff --git a/python/paddle/dataset/imdb.py b/python/paddle/dataset/imdb.py
index e7fe4e0b7e5832c2bc7ca1307725936a70292c39..39fc29fdac8aa907ac7d1de4b84aa8b3640d903b 100644
--- a/python/paddle/dataset/imdb.py
+++ b/python/paddle/dataset/imdb.py
@@ -25,6 +25,7 @@ import collections
 import tarfile
 import re
 import string
+from six.moves import range
 
 __all__ = ['build_dict', 'train', 'test', 'convert']
 
@@ -66,7 +67,7 @@ def build_dict(pattern, cutoff):
 
     dictionary = sorted(word_freq, key=lambda x: (-x[1], x[0]))
     words, _ = list(zip(*dictionary))
-    word_idx = dict(list(zip(words, list(range(len(words))))))
+    word_idx = dict(list(zip(words, range(len(words)))))
     word_idx['<unk>'] = len(words)
     return word_idx
 
diff --git a/python/paddle/dataset/imikolov.py b/python/paddle/dataset/imikolov.py
index bc007c9d3c8e2f1e4ff091f7c2c93eacbbe8d0e0..bfb087ff381a00a9e6de3a2654ef53aa4c4fdce6 100644
--- a/python/paddle/dataset/imikolov.py
+++ b/python/paddle/dataset/imikolov.py
@@ -14,13 +14,14 @@
 """
 imikolov's simple dataset.
 
-This module will download dataset from 
+This module will download dataset from
 http://www.fit.vutbr.cz/~imikolov/rnnlm/ and parse training set and test set
 into paddle reader creators.
 """
 import paddle.dataset.common
 import collections
 import tarfile
+from six.moves import range
 
 __all__ = ['train', 'test', 'build_dict', 'convert']
 
@@ -68,7 +69,7 @@ def build_dict(min_word_freq=50):
 
         word_freq_sorted = sorted(word_freq, key=lambda x: (-x[1], x[0]))
         words, _ = list(zip(*word_freq_sorted))
-        word_idx = dict(list(zip(words, list(range(len(words))))))
+        word_idx = dict(list(zip(words, range(len(words)))))
         word_idx['<unk>'] = len(words)
 
     return word_idx
diff --git a/python/paddle/dataset/mnist.py b/python/paddle/dataset/mnist.py
index ffa9008c80129b80b3807dbab37bc198e59cf5a2..55e82fa755531f65a2f76bcb0fd71b3f4955e775 100644
--- a/python/paddle/dataset/mnist.py
+++ b/python/paddle/dataset/mnist.py
@@ -21,6 +21,7 @@ import paddle.dataset.common
 import subprocess
 import numpy
 import platform
+from six.moves import range
 __all__ = ['train', 'test', 'convert']
 
 URL_PREFIX = 'http://yann.lecun.com/exdb/mnist/'
diff --git a/python/paddle/dataset/tests/common_test.py b/python/paddle/dataset/tests/common_test.py
index 777cd06a19726f8ad73774c958c8cb512808a3aa..ede3d593ebff6a9f4fc8de0680965686a4c40279 100644
--- a/python/paddle/dataset/tests/common_test.py
+++ b/python/paddle/dataset/tests/common_test.py
@@ -16,6 +16,7 @@ import paddle.dataset.common
 import unittest
 import tempfile
 import glob
+from six.moves import range
 
 
 class TestCommon(unittest.TestCase):
diff --git a/python/paddle/dataset/uci_housing.py b/python/paddle/dataset/uci_housing.py
index 410ca7af0d6d1dc26acbf92fce5e49fce7d3a3bb..cc946762da466cd2b5142ff7508a5b1c0976c740 100644
--- a/python/paddle/dataset/uci_housing.py
+++ b/python/paddle/dataset/uci_housing.py
@@ -22,6 +22,7 @@ parse training set and test set into paddle reader creators.
 import os
 
 import numpy as np
+import six
 import tempfile
 import tarfile
 import os
@@ -74,7 +75,7 @@ def load_data(filename, feature_num=14, ratio=0.8):
     maximums, minimums, avgs = data.max(axis=0), data.min(axis=0), data.sum(
         axis=0) / data.shape[0]
     feature_range(maximums[:-1], minimums[:-1])
-    for i in range(feature_num - 1):
+    for i in six.moves.range(feature_num - 1):
         data[:, i] = (data[:, i] - avgs[i]) / (maximums[i] - minimums[i])
     offset = int(data.shape[0] * ratio)
     UCI_TRAIN_DATA = data[:offset]
@@ -137,7 +138,7 @@ def predict_reader():
     It returns just one tuple data to do inference.
 
     :return: one tuple data
-    :rtype: tuple 
+    :rtype: tuple
     """
     global UCI_TEST_DATA
     load_data(paddle.dataset.common.download(URL, 'uci_housing', MD5))
diff --git a/python/paddle/fluid/executor.py b/python/paddle/fluid/executor.py
index 35da1d06a2c1da8ba663ea0f0b9a0e58ea7c4470..8437a9f20f924ccbff51a4009e110381fb975e51 100644
--- a/python/paddle/fluid/executor.py
+++ b/python/paddle/fluid/executor.py
@@ -346,7 +346,7 @@ class Executor(object):
     def _fetch_data(self, fetch_list, fetch_var_name, scope):
         outs = [
             core.get_fetch_variable(scope, fetch_var_name, i)
-            for i in range(len(fetch_list))
+            for i in six.moves.range(len(fetch_list))
         ]
         return outs
 
diff --git a/python/paddle/fluid/framework.py b/python/paddle/fluid/framework.py
index f0653a43ce607b2f321bb5c4ac49a8fa56c81e99..9a2c8adc034d028ea56ec35862685fb65c9fa82f 100644
--- a/python/paddle/fluid/framework.py
+++ b/python/paddle/fluid/framework.py
@@ -1497,7 +1497,9 @@ class Program(object):
         else:
             p = Program()
             p.desc = core.ProgramDesc(self.desc)
-            p.blocks = [Block(p, i) for i in range(self.desc.num_blocks())]
+            p.blocks = [
+                Block(p, i) for i in six.moves.range(self.desc.num_blocks())
+            ]
             p._sync_with_cpp()
 
         p._copy_param_info_from(self)
@@ -1549,7 +1551,9 @@ class Program(object):
             targets_idx.append([t.block.idx, t.idx])
         res = Program()
         res.desc = core.prune(self.desc, targets_idx)
-        res.blocks = [Block(res, i) for i in range(res.desc.num_blocks())]
+        res.blocks = [
+            Block(res, i) for i in six.moves.range(res.desc.num_blocks())
+        ]
         res._sync_with_cpp()
         return res
 
@@ -1590,13 +1594,15 @@ class Program(object):
                 root_block._remove_var(var.name())
 
         # change all `is_test` attributes to True
-        for i in range(res.desc.num_blocks()):
+        for i in six.moves.range(res.desc.num_blocks()):
             block = res.desc.block(i)
-            for j in range(block.op_size()):
+            for j in six.moves.range(block.op_size()):
                 op = block.op(j)
                 if op.has_attr('is_test'):
                     op.set_attr('is_test', True)
-        res.blocks = [Block(res, i) for i in range(res.desc.num_blocks())]
+        res.blocks = [
+            Block(res, i) for i in six.moves.range(res.desc.num_blocks())
+        ]
         res._sync_with_cpp()
         return res
 
@@ -1616,7 +1622,7 @@ class Program(object):
         """
         p = Program()
         p.desc = core.ProgramDesc(binary_str)
-        p.blocks = [Block(p, i) for i in range(p.desc.num_blocks())]
+        p.blocks = [Block(p, i) for i in six.moves.range(p.desc.num_blocks())]
         p._sync_with_cpp()
         return p
 
diff --git a/python/paddle/fluid/layer_helper.py b/python/paddle/fluid/layer_helper.py
index 64337465ed0778edf89c5c4fda06cd27b3f386f2..5f66f54cf743c5fb8ec939d6ef29c3c61428be48 100644
--- a/python/paddle/fluid/layer_helper.py
+++ b/python/paddle/fluid/layer_helper.py
@@ -85,7 +85,7 @@ class LayerHelper(object):
             raise ValueError("parameter number mismatch")
         elif len(param_attr) == 1 and length != 1:
             tmp = [None] * length
-            for i in range(length):
+            for i in six.moves.range(length):
                 tmp[i] = copy.deepcopy(param_attr[0])
             param_attr = tmp
         return param_attr
diff --git a/python/paddle/fluid/layers/detection.py b/python/paddle/fluid/layers/detection.py
index 9fae96d9bc7f5e6e89f8ee3894d28ae306e10ca9..c11455b7a6ecf467e64ad20a9499623114c09de3 100644
--- a/python/paddle/fluid/layers/detection.py
+++ b/python/paddle/fluid/layers/detection.py
@@ -21,6 +21,7 @@ from ..layer_helper import LayerHelper
 from . import tensor
 from . import nn
 import math
+import six
 from functools import reduce
 
 __all__ = [
@@ -102,7 +103,7 @@ def rpn_target_assign(loc,
             examples.
 
     Returns:
-        tuple: 
+        tuple:
                A tuple(predicted_scores, predicted_location, target_label,
                target_bbox) is returned. The predicted_scores and
                predicted_location is the predicted result of the RPN.
@@ -113,7 +114,7 @@ def rpn_target_assign(loc,
                anchors. The predicted_scores is a 2D Tensor with shape
                [F + B, 1], and the shape of target_label is same as the shape
                of the predicted_scores, B is the number of the background
-               anchors, the F and B is depends on the input of this operator. 
+               anchors, the F and B is depends on the input of this operator.
 
     Examples:
         .. code-block:: python
@@ -230,8 +231,8 @@ def detection_output(loc,
         nms_eta(float): The parameter for adaptive NMS.
 
     Returns:
-        Variable: 
-        
+        Variable:
+
             The detection outputs is a LoDTensor with shape [No, 6].
             Each row has six values: [label, confidence, xmin, ymin, xmax, ymax].
             `No` is the total number of detections in this mini-batch. For each
@@ -501,7 +502,7 @@ def target_assign(input,
 
     Assumed that the row offset for each instance in `neg_indices` is called neg_lod,
     for i-th instance and each `id` of neg_indices in this instance:
-    
+
     .. code-block:: text
 
         out[i][id][0 : K] = {mismatch_value, mismatch_value, ...}
@@ -519,11 +520,11 @@ def target_assign(input,
        mismatch_value (float32): Fill this value to the mismatched location.
 
     Returns:
-        tuple: 
-               A tuple(out, out_weight) is returned. out is a 3D Tensor with 
-               shape [N, P, K], N and P is the same as they are in 
-               `neg_indices`, K is the same as it in input of X. If 
-               `match_indices[i][j]`. out_weight is the weight for output with 
+        tuple:
+               A tuple(out, out_weight) is returned. out is a 3D Tensor with
+               shape [N, P, K], N and P is the same as they are in
+               `neg_indices`, K is the same as it in input of X. If
+               `match_indices[i][j]`. out_weight is the weight for output with
                the shape of [N, P, 1].
 
     Examples:
@@ -822,7 +823,7 @@ def prior_box(input,
        offset(float): Prior boxes center offset. Default: 0.5
        name(str): Name of the prior box op. Default: None.
        min_max_aspect_ratios_order(bool): If set True, the output prior box is
-            in order of [min, max, aspect_ratios], which is consistent with 
+            in order of [min, max, aspect_ratios], which is consistent with
             Caffe. Please note, this order affects the weights order of
             convolution layer followed by and does not affect the final
             detection results. Default: False.
@@ -965,7 +966,7 @@ def multi_box_head(inputs,
        stride(int|list|tuple): The stride of conv2d. Default:1,
        name(str): Name of the prior box layer. Default: None.
        min_max_aspect_ratios_order(bool): If set True, the output prior box is
-            in order of [min, max, aspect_ratios], which is consistent with 
+            in order of [min, max, aspect_ratios], which is consistent with
             Caffe. Please note, this order affects the weights order of
             convolution layer followed by and does not affect the fininal
             detection results. Default: False.
@@ -1033,7 +1034,7 @@ def multi_box_head(inputs,
         min_sizes = []
         max_sizes = []
         step = int(math.floor(((max_ratio - min_ratio)) / (num_layer - 2)))
-        for ratio in range(min_ratio, max_ratio + 1, step):
+        for ratio in six.moves.range(min_ratio, max_ratio + 1, step):
             min_sizes.append(base_size * ratio / 100.)
             max_sizes.append(base_size * (ratio + step) / 100.)
         min_sizes = [base_size * .10] + min_sizes
diff --git a/python/paddle/fluid/layers/io.py b/python/paddle/fluid/layers/io.py
index 327ae309816344a0bcebfe70ffb59a00eab1d86f..f9b01203e2cb2c23c025f41c501fc19255f72199 100644
--- a/python/paddle/fluid/layers/io.py
+++ b/python/paddle/fluid/layers/io.py
@@ -13,6 +13,7 @@
 # limitations under the License.
 import contextlib
 import multiprocessing
+import six
 import threading
 
 from ..data_feeder import DataFeeder
@@ -69,7 +70,7 @@ def data(name,
     """
     helper = LayerHelper('data', **locals())
     shape = list(shape)
-    for i in range(len(shape)):
+    for i in six.moves.range(len(shape)):
         if shape[i] is None:
             shape[i] = -1
             append_batch_size = False
@@ -674,7 +675,7 @@ def py_reader(capacity,
 
         def __tensor_provider__():
             for slots in paddle_reader():
-                yield [slots[str(idx)] for idx in xrange(counter)]
+                yield [slots[str(idx)] for idx in six.moves.xrange(counter)]
 
         __set_tensor_provider__(__tensor_provider__)
 
@@ -1005,7 +1006,7 @@ class Preprocessor(object):
         source_lod_levels = self.underlying_reader.desc.lod_levels()
         self.source_var_names = [
             unique_name("preprocessor_source")
-            for _ in range(len(source_shapes))
+            for _ in six.moves.range(len(source_shapes))
         ]
         source_vars = []
         for var_name, shape, dtype, lod_level in zip(
diff --git a/python/paddle/fluid/nets.py b/python/paddle/fluid/nets.py
index 08480671d8a5c50bbec97930c451cbcdc241e1fe..46e4c701953c795fabc0f5cacf2cdaef0ea3268b 100644
--- a/python/paddle/fluid/nets.py
+++ b/python/paddle/fluid/nets.py
@@ -11,6 +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.
+import six
 from . import layers
 
 __all__ = [
@@ -210,7 +211,7 @@ def img_conv_group(input,
     conv_with_batchnorm = __extend_list__(conv_with_batchnorm)
     conv_batchnorm_drop_rate = __extend_list__(conv_batchnorm_drop_rate)
 
-    for i in range(len(conv_num_filter)):
+    for i in six.moves.range(len(conv_num_filter)):
         local_conv_act = conv_act
         if conv_with_batchnorm[i]:
             local_conv_act = None
diff --git a/python/paddle/fluid/parallel_executor.py b/python/paddle/fluid/parallel_executor.py
index eabe6bb901ad1b7afe71717eb3f5f5765f261201..97849672b223dd7ec93f276f1f796b3f06972e09 100644
--- a/python/paddle/fluid/parallel_executor.py
+++ b/python/paddle/fluid/parallel_executor.py
@@ -19,6 +19,7 @@ from . import framework
 from . import executor
 import warnings
 import sys
+import six
 import os
 
 __all__ = ['ParallelExecutor', 'ExecutionStrategy', 'BuildStrategy']
@@ -95,7 +96,7 @@ class ParallelExecutor(object):
         self._places = []
         self._act_places = []
         if use_cuda:
-            for i in range(core.get_cuda_device_count()):
+            for i in six.moves.range(core.get_cuda_device_count()):
                 p = core.Place()
                 self._act_places.append(core.CUDAPlace(i))
                 p.set_place(self._act_places[-1])
@@ -103,7 +104,7 @@ class ParallelExecutor(object):
         else:
             cpu_num = int(
                 os.environ.get('CPU_NUM', multiprocessing.cpu_count()))
-            for i in range(cpu_num):
+            for i in six.moves.range(cpu_num):
                 p = core.Place()
                 self._act_places.append(core.CPUPlace())
                 p.set_place(self._act_places[-1])
diff --git a/python/paddle/fluid/tests/demo/pyreader.py b/python/paddle/fluid/tests/demo/pyreader.py
index 82065401935036ca346fa395c033f0f57100f01b..737644a25f412a534498f2ab90e034dcac6f1890 100644
--- a/python/paddle/fluid/tests/demo/pyreader.py
+++ b/python/paddle/fluid/tests/demo/pyreader.py
@@ -13,6 +13,7 @@
 # limitations under the License.
 
 import numpy
+import six
 
 import paddle
 import paddle.dataset.mnist as mnist
@@ -31,7 +32,7 @@ def network(is_train):
 
     hidden = img
 
-    for i in xrange(2):
+    for i in six.moves.xrange(2):
         hidden = fluid.layers.fc(input=hidden, size=100, act='tanh')
         hidden = fluid.layers.dropout(
             hidden, dropout_prob=0.5, is_test=not is_train)
@@ -74,7 +75,7 @@ def main():
 
     test_reader.decorate_paddle_reader(paddle.batch(mnist.test(), 512))
 
-    for epoch_id in xrange(10):
+    for epoch_id in six.moves.xrange(10):
         train_reader.start()
         try:
             while True:
diff --git a/python/paddle/fluid/tests/unittests/dist_transformer.py b/python/paddle/fluid/tests/unittests/dist_transformer.py
index ee8020a73546cb9037e9dc4be589c62bb1b6b937..6bd4ecbbe169360c1a1b8dc7e23364b6cf79a61e 100644
--- a/python/paddle/fluid/tests/unittests/dist_transformer.py
+++ b/python/paddle/fluid/tests/unittests/dist_transformer.py
@@ -22,6 +22,7 @@ import paddle.fluid as fluid
 from paddle.fluid import core
 import os
 import sys
+import six
 import transformer_model
 import paddle.dataset.wmt16 as wmt16
 
@@ -222,7 +223,7 @@ class DistTransformer2x2(object):
 
         first_loss, = exe.run(fetch_list=[avg_cost.name])
         print(first_loss)
-        for i in xrange(5):
+        for i in six.moves.xrange(5):
             _ = exe.run(fetch_list=[avg_cost.name])
         last_loss, = exe.run(fetch_list=[avg_cost.name])
         print(last_loss)
diff --git a/python/paddle/fluid/tests/unittests/test_split_ids_op.py b/python/paddle/fluid/tests/unittests/test_split_ids_op.py
index ca7861309839d183e18c168403881a0b1b5bf309..20bba3ac338cacff65061c04c2ee0f2fc6ad6de0 100644
--- a/python/paddle/fluid/tests/unittests/test_split_ids_op.py
+++ b/python/paddle/fluid/tests/unittests/test_split_ids_op.py
@@ -14,6 +14,7 @@
 
 import unittest
 import numpy as np
+import six
 from op_test import OpTest
 import paddle.fluid.core as core
 from paddle.fluid.op import Operator
@@ -59,7 +60,7 @@ class TestSpliteIds(unittest.TestCase):
         x_tensor = x.get_tensor()
         x_tensor.set(np_array, place)
 
-        outs_name = ["out%d" % i for i in xrange(3)]
+        outs_name = ["out%d" % i for i in six.moves.xrange(3)]
         outs = [
             scope.var(var_name).get_selected_rows() for var_name in outs_name
         ]