diff --git a/python/paddle/fluid/__init__.py b/python/paddle/fluid/__init__.py
index 37d368946770978700abe49eef6825e1d96839f0..c8a435748dc5b51bf9e57b5b597e1422f0380e8e 100644
--- a/python/paddle/fluid/__init__.py
+++ b/python/paddle/fluid/__init__.py
@@ -60,6 +60,7 @@ __all__ = framework.__all__ + executor.__all__ + concurrency.__all__ +\
     'io',
     'initializer',
     'layers',
+    'transpiler'
     'nets',
     'optimizer',
     'learning_rate_decay',
diff --git a/python/paddle/fluid/trainer.py b/python/paddle/fluid/trainer.py
index a9fa2359e019d0301110dfc3573e2d2534c6a47a..1cbecd69e59882212d623b7fcaf741f0370a7a15 100644
--- a/python/paddle/fluid/trainer.py
+++ b/python/paddle/fluid/trainer.py
@@ -19,6 +19,7 @@ import executor
 import data_feeder
 import contextlib
 import io
+import transpiler
 
 # optimizer is same as the parameter of Trainer.__init__. Rename it to opt_module
 import optimizer as opt_module
diff --git a/python/setup.py.in b/python/setup.py.in
index a811b509a90b8b0d84451f54462a0308c062d022..c42601d335f01491156dc3591341c1a3213aecfe 100644
--- a/python/setup.py.in
+++ b/python/setup.py.in
@@ -68,7 +68,8 @@ packages=['paddle',
           'paddle.fluid',
           'paddle.fluid.proto',
           'paddle.fluid.proto.profiler',
-          'paddle.fluid.layers']
+          'paddle.fluid.layers',
+          'paddle.fluid.transpiler']
 
 if '${WITH_FLUID_ONLY}'== 'OFF':
     packages+=['paddle.proto',