diff --git a/caffe2fluid/convert.py b/caffe2fluid/convert.py
index b0252e3c03db3626696a3672971f0704461417e7..5e9ecac73697dffca4f0956c4d03e730435bec11 100755
--- a/caffe2fluid/convert.py
+++ b/caffe2fluid/convert.py
@@ -42,7 +42,8 @@ def convert(def_path, caffemodel_path, data_output_path, code_output_path,
         if code_output_path:
             print_stderr('Saving source...')
             with open(code_output_path, 'wb') as src_out:
-                src_out.write(transformer.transform_source())
+#                 print(type(transformer.transform_source()))
+                src_out.write(str.encode(transformer.transform_source()))
         print_stderr('set env variable before using converted model '\
                 'if used custom_layers:')
         custom_pk_path = os.path.dirname(os.path.abspath(__file__))
diff --git a/caffe2fluid/examples/imagenet/compare.py b/caffe2fluid/examples/imagenet/compare.py
index c995e6df17a4be068984cece06a9b3a33f6ea4f4..18766df5f43e8028da103332b844a2cfe035cb9c 100644
--- a/caffe2fluid/examples/imagenet/compare.py
+++ b/caffe2fluid/examples/imagenet/compare.py
@@ -6,6 +6,7 @@
 
 import sys
 import os
+import functools
 
 
 def walk_dir(rootdir):
@@ -28,11 +29,11 @@ def calc_diff(f1, f2):
     d1 = d1.flatten()
     d2 = d2.flatten()
 
-    d1_num = reduce(lambda x, y: x * y, d1.shape)
-    d2_num = reduce(lambda x, y: x * y, d2.shape)
+    d1_num = functools.reduce(lambda x, y: x * y, d1.shape)
+    d2_num = functools.reduce(lambda x, y: x * y, d2.shape)
     if d1_num != d2_num:
-        print d1.shape
-        print d2.shape
+        print(d1.shape)
+        print(d2.shape)
         assert (d1_num == d2_num), "their shape is not consistent"
 
     try:
@@ -54,7 +55,7 @@ def compare(path1, path2, no_exception):
         print('[max_df:%.4e, sq_df:%.4e] when compare %s <=> %s' %
               (max_df, sq_df, os.path.basename(f1), os.path.basename(f2)))
         if no_exception is False:
-            assert (max_df < 1e-5), \
+            assert (max_df < 1e-4), \
                     'max_df is too large with value[%.6e]' % (max_df)
             assert (sq_df < 1e-10), \
                     'sq_df is too large with value[%.6e]' % (sq_df)
diff --git a/caffe2fluid/examples/imagenet/infer.py b/caffe2fluid/examples/imagenet/infer.py
index 9de51e1af9685478c3a30b7692e6472bf2ce17fd..3962d96ba1043ca8b8e4121f294c1fb0c10e928c 100644
--- a/caffe2fluid/examples/imagenet/infer.py
+++ b/caffe2fluid/examples/imagenet/infer.py
@@ -57,7 +57,7 @@ def build_model(net_file, net_name):
 
     fluid = import_fluid()
     inputs_dict = MyNet.input_shapes()
-    input_name = inputs_dict.keys()[0]
+    input_name = list(inputs_dict.keys())[0]
     input_shape = inputs_dict[input_name]
     images = fluid.layers.data(
         name=input_name, shape=input_shape, dtype='float32')
@@ -222,8 +222,8 @@ def infer(model_path, imgfile, net_file=None, net_name=None, debug=True):
         feed_shapes = ret['feed_shapes']
         net = ret['net']
 
-    input_name = feed_names[0]
-    input_shape = feed_shapes[0]
+    input_name = list(feed_names)[0]
+    input_shape = list(feed_shapes)[0]
 
     np_images = load_data(imgfile, input_shape)
     results = exe.run(program=program,
@@ -249,7 +249,7 @@ def caffe_infer(prototxt, caffemodel, datafile):
     import caffe
 
     net = caffe.Net(prototxt, caffemodel, caffe.TEST)
-    input_layer = net.blobs.keys()[0]
+    input_layer = list(net.blobs.keys())[0]
     print('got name of input layer is:%s' % (input_layer))
     input_shape = list(net.blobs[input_layer].data.shape[1:])
 
@@ -266,7 +266,7 @@ def caffe_infer(prototxt, caffemodel, datafile):
     for k, v in net.blobs.items():
         k = k.replace('/', '_')
         names.append(k)
-        results.append(v.data.copy())
+        results.append(v.data[0].copy())
 
     dump_path = 'results.caffe'
     dump_results(results, names, dump_path)
diff --git a/caffe2fluid/examples/imagenet/tools/diff.sh b/caffe2fluid/examples/imagenet/tools/diff.sh
index 25e5d3b6c1bc301fbc505ce45103ddf091fd86f7..45250d2aedee3b155bbf656dc44c07e8bfcb0e3c 100755
--- a/caffe2fluid/examples/imagenet/tools/diff.sh
+++ b/caffe2fluid/examples/imagenet/tools/diff.sh
@@ -29,14 +29,14 @@ fi
 
 mkdir -p $results_root
 
-prototxt="models.caffe/$model_name/${model_name}.prototxt"
-caffemodel="models.caffe/${model_name}/${model_name}.caffemodel"
+prototxt="$2/${model_name}.prototxt"
+caffemodel="$2/${model_name}.caffemodel"
 
 #1, dump layers' results from paddle
 paddle_results="$results_root/${model_name}.paddle"
 rm -rf $paddle_results
 rm -rf "results.paddle"
-bash ./tools/run.sh $model_name ./models.caffe/$model_name ./models/$model_name
+bash ./tools/run.sh $model_name $2 $3
 if [[ $? -ne 0 ]] || [[ ! -e "results.paddle" ]];then
     echo "not found paddle's results, maybe failed to convert"
     exit 1
@@ -47,7 +47,7 @@ mv results.paddle $paddle_results
 caffe_results="$results_root/${model_name}.caffe"
 rm -rf $caffe_results
 rm -rf "results.caffe"
-PYTHON=`which cfpython`
+PYTHON=`which python`
 if [[ -z $PYTHON ]];then
     PYTHON=`which python`
 fi
diff --git a/caffe2fluid/examples/imagenet/tools/run.sh b/caffe2fluid/examples/imagenet/tools/run.sh
index 7eb23f4c1257da259f62af6ce152fb3a5fda3b43..fc9fdc2f037ca0c38515ca39e55bff1ebeb65cb8 100755
--- a/caffe2fluid/examples/imagenet/tools/run.sh
+++ b/caffe2fluid/examples/imagenet/tools/run.sh
@@ -10,6 +10,8 @@
 #
 
 #set -x
+
+
 if [[ $# -lt 3 ]];then
     echo "usage:"
     echo "  bash $0 [model_name] [cf_model_path] [pd_model_path] [only_convert]"
@@ -21,7 +23,6 @@ else
     pd_model_path=$3
     only_convert=$4
 fi
-
 proto_file=$cf_model_path/${model_name}.prototxt
 caffemodel_file=$cf_model_path/${model_name}.caffemodel
 weight_file=$pd_model_path/${model_name}.npy
@@ -41,7 +42,7 @@ if [[ ! -e $pd_model_path ]];then
     mkdir $pd_model_path
 fi
 
-PYTHON=`which cfpython`
+PYTHON=`which python`
 if [[ -z $PYTHON ]];then
     PYTHON=`which python`
 fi
@@ -60,7 +61,7 @@ else
 fi
 
 if [[ -z $only_convert ]];then
-    PYTHON=`which pdpython`
+    PYTHON=`which python`
     if [[ -z $PYTHON ]];then
         PYTHON=`which python`
     fi
diff --git a/caffe2fluid/kaffe/custom_layers/__init__.py b/caffe2fluid/kaffe/custom_layers/__init__.py
index 8505aee05f90f999347b687349fdfd7c7caf1a0f..3e292e6d4f06b813dd665a5016cd9372e30f3e32 100644
--- a/caffe2fluid/kaffe/custom_layers/__init__.py
+++ b/caffe2fluid/kaffe/custom_layers/__init__.py
@@ -4,19 +4,20 @@
 from .register import get_registered_layers
 #custom layer import begins
 
-import axpy
-import flatten
-import argmax
-import reshape
-import roipooling
-import priorbox
-import permute
-import detection_out
-import normalize
-import select
-import crop
-import power
-import reduction
+from . import axpy
+from . import flatten
+from . import argmax
+from . import argmax
+from . import reshape
+from . import roipooling
+from . import priorbox
+from . import permute
+from . import detection_out
+from . import normalize
+from . import select
+from . import crop
+from . import power
+from . import reduction
 
 #custom layer import ends
 
diff --git a/caffe2fluid/kaffe/graph.py b/caffe2fluid/kaffe/graph.py
index baea3cc1dc9431d07d0d3ca7191a429d1ef0f398..b325b93edf6a78e99a75e49bc8e0a47146759a61 100644
--- a/caffe2fluid/kaffe/graph.py
+++ b/caffe2fluid/kaffe/graph.py
@@ -152,14 +152,14 @@ class Graph(object):
                 data_shape = '--'
                 out_shape = node.output_shape or '--'
                 s.append('{:<20} {:<30} {:>20} {:>20}'.format(
-                    node.kind, node.name, data_shape, tuple(out_shape)))
+                    node.kind, node.name, data_shape, str(tuple(out_shape))))
             else:
                 for d in node.data:
                     #data_shape = node.data[0].shape if node.data else '--'
                     data_shape = d.shape
                     out_shape = node.output_shape or '--'
                     s.append('{:<20} {:<30} {:>20} {:>20}'.format(
-                        node.kind, node.name, data_shape, tuple(out_shape)))
+                        node.kind, node.name, str(data_shape), str(tuple(out_shape))))
         return '\n'.join(s)
 
 
@@ -368,4 +368,4 @@ class NodeMapper(NodeDispatch):
         return mapped_node
 
     def commit(self, mapped_chains):
-        raise NotImplementedError('Must be implemented by subclass.')
+        raise NotImplementedError('Must be implemented by subclass.')
\ No newline at end of file
diff --git a/caffe2fluid/kaffe/layers.py b/caffe2fluid/kaffe/layers.py
index 0d0aa1adc3c5a5e21f74d188d1784a94cf4acf9e..9205663c019a28db710eae5ed6330e3da2b548f4 100644
--- a/caffe2fluid/kaffe/layers.py
+++ b/caffe2fluid/kaffe/layers.py
@@ -1,8 +1,9 @@
 import re
 import numbers
 from collections import namedtuple
-
-import custom_layers
+import sys
+sys.path.append('~/paddlepaddle/X2Paddle_v2.0/caffe2fluid/kaffe/custom_layers')
+from . import custom_layers
 from .shapes import *
 
 LAYER_DESCRIPTORS = {
diff --git a/caffe2fluid/kaffe/net_template.py b/caffe2fluid/kaffe/net_template.py
index 86a6628c72c99359fb29290403de62401fc074c6..cb601bbc18fb0f81889a2b77d71a709cdeb858db 100644
--- a/caffe2fluid/kaffe/net_template.py
+++ b/caffe2fluid/kaffe/net_template.py
@@ -42,7 +42,7 @@ class MyNet(object):
     def convert(cls, npy_model, fluid_path, outputs=None):
         fluid = import_fluid()
         shapes = cls.input_shapes()
-        input_name = shapes.keys()[0]
+        input_name = list(shapes.keys())[0]    #need to review
         feed_data = {}
         for name, shape in shapes.items():
             data_layer = fluid.layers.data(
@@ -157,5 +157,5 @@ def generate_main_code(net_name):
 if __name__ == "__main__":
     """ just for testing
     """
-    print generate_net_code('Attribute', "{'data': [3, 277, 277]}")
-    print generate_main_code('Attribute')
+    print(generate_net_code('Attribute', "{'data': [3, 277, 277]}"))
+    print(generate_main_code('Attribute'))
diff --git a/caffe2fluid/kaffe/paddle/network.py b/caffe2fluid/kaffe/paddle/network.py
index 883b24c67458305ffd11efa0eb0f7f51d1512683..0286bceab13a264adce6af2aa467a0ba51c61cbe 100644
--- a/caffe2fluid/kaffe/paddle/network.py
+++ b/caffe2fluid/kaffe/paddle/network.py
@@ -2,6 +2,8 @@ import sys
 import os
 import math
 import numpy as np
+from past.builtins import basestring    
+
 
 
 def import_fluid():
@@ -108,7 +110,7 @@ class Network(object):
                 continue
 
             layer = self.layers[op_name]
-            for param_name, data in data_dict[op_name].iteritems():
+            for param_name, data in data_dict[op_name].items():    #need to review
                 try:
                     name = '%s_%s' % (op_name, param_name)
                     v = fluid.global_scope().find_var(name)
diff --git a/caffe2fluid/kaffe/paddle/transformer.py b/caffe2fluid/kaffe/paddle/transformer.py
index b07393f1b0b649748bbd08b66c0546282c59d0d0..77ef7aa5132c9a6cc0f379fa6dc9078b53fd8b03 100644
--- a/caffe2fluid/kaffe/paddle/transformer.py
+++ b/caffe2fluid/kaffe/paddle/transformer.py
@@ -1,5 +1,5 @@
 import numpy as np
-
+from past.builtins import basestring
 from ..errors import KaffeError, print_stderr
 from ..graph import GraphBuilder, NodeMapper
 from ..layers import NodeKind
@@ -34,6 +34,8 @@ class PaddleNode(object):
         '''Emits the Python source for this node.'''
         # Format positional arguments
         args = map(self.format, self.args)
+        args = list(args)
+
         # Format any keyword arguments
         if self.kwargs:
             args += [self.pair(k, v) for k, v in self.kwargs]
@@ -335,7 +337,9 @@ class Transformer(object):
         ]
 
         self.graph = graph.transformed(transformers)
-
+        
+        
+            
         #for the purpose of recording name mapping because of fused nodes
         trace = SubNodeFuser.traced_names()
         chg2real = {}
@@ -372,11 +376,14 @@ class Transformer(object):
                 # Convert parameters to dictionaries
                 ParameterNamer(),
             ]
+            
             self.graph = self.graph.transformed(transformers)
+            
             self.params = {
                 node.name: node.data
                 for node in self.graph.nodes if node.data
             }
+            
             self.params['caffe2fluid_name_trace'] = self.graph.get_name_trace()
 
         return self.params
diff --git a/caffe2fluid/kaffe/transformers.py b/caffe2fluid/kaffe/transformers.py
index f436ec8cbfb49394ef2880dbf0cb46aeff7a6ca0..1c3f8f1277724eb1417be71c378b452e3cf5d9e2 100644
--- a/caffe2fluid/kaffe/transformers.py
+++ b/caffe2fluid/kaffe/transformers.py
@@ -38,7 +38,7 @@ class DataInjector(object):
         caffe = get_caffe_resolver().caffe
         net = caffe.Net(self.def_path, self.data_path, caffe.TEST)
         data = lambda blob: blob.data
-        self.params = [(k, map(data, v)) for k, v in net.params.items()]
+        self.params = [(k, list(map(data, v))) for k, v in net.params.items()]
 
     def load_using_pb(self):
         data = get_caffe_resolver().NetParameter()
@@ -155,6 +155,9 @@ class DataReshaper(object):
                 continue
 
             transpose_order = self.map(node.kind)
+#             node_data = list(node.data)
+#             print(node)
+#             print(node_data)
             weights = node.data[0]
             if node.kind == NodeKind.InnerProduct:
                 # The FC layer connected to the spatial layer needs to be