diff --git a/paddle/fluid/API.spec b/paddle/fluid/API.spec
index 534acffa7ffdc07bfcbfb328b540221c2b988de5..dc162e248a8b3701c157f79155f4cee232144d9b 100644
--- a/paddle/fluid/API.spec
+++ b/paddle/fluid/API.spec
@@ -54,8 +54,6 @@ paddle.fluid.DistributeTranspiler.get_pserver_programs ArgSpec(args=['self', 'en
 paddle.fluid.DistributeTranspiler.get_startup_program ArgSpec(args=['self', 'endpoint', 'pserver_program', 'startup_program'], varargs=None, keywords=None, defaults=(None, None))
 paddle.fluid.DistributeTranspiler.get_trainer_program ArgSpec(args=['self', 'wait_port'], varargs=None, keywords=None, defaults=(True,))
 paddle.fluid.DistributeTranspiler.transpile ArgSpec(args=['self', 'trainer_id', 'program', 'pservers', 'trainers', 'sync_mode', 'startup_program'], varargs=None, keywords=None, defaults=(None, '127.0.0.1:6174', 1, True, None))
-paddle.fluid.InferenceTranspiler.__init__ 
-paddle.fluid.InferenceTranspiler.transpile ArgSpec(args=['self', 'program', 'place', 'scope'], varargs=None, keywords=None, defaults=(None,))
 paddle.fluid.memory_optimize ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level'], varargs=None, keywords=None, defaults=(None, False, 0))
 paddle.fluid.release_memory ArgSpec(args=['input_program', 'skip_opt_set'], varargs=None, keywords=None, defaults=(None,))
 paddle.fluid.DistributeTranspilerConfig.__init__ 
@@ -338,8 +336,6 @@ paddle.fluid.transpiler.DistributeTranspiler.get_pserver_programs ArgSpec(args=[
 paddle.fluid.transpiler.DistributeTranspiler.get_startup_program ArgSpec(args=['self', 'endpoint', 'pserver_program', 'startup_program'], varargs=None, keywords=None, defaults=(None, None))
 paddle.fluid.transpiler.DistributeTranspiler.get_trainer_program ArgSpec(args=['self', 'wait_port'], varargs=None, keywords=None, defaults=(True,))
 paddle.fluid.transpiler.DistributeTranspiler.transpile ArgSpec(args=['self', 'trainer_id', 'program', 'pservers', 'trainers', 'sync_mode', 'startup_program'], varargs=None, keywords=None, defaults=(None, '127.0.0.1:6174', 1, True, None))
-paddle.fluid.transpiler.InferenceTranspiler.__init__ 
-paddle.fluid.transpiler.InferenceTranspiler.transpile ArgSpec(args=['self', 'program', 'place', 'scope'], varargs=None, keywords=None, defaults=(None,))
 paddle.fluid.transpiler.memory_optimize ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level'], varargs=None, keywords=None, defaults=(None, False, 0))
 paddle.fluid.transpiler.release_memory ArgSpec(args=['input_program', 'skip_opt_set'], varargs=None, keywords=None, defaults=(None,))
 paddle.fluid.transpiler.HashName.__init__ ArgSpec(args=['self', 'pserver_endpoints'], varargs=None, keywords=None, defaults=None)
diff --git a/python/paddle/fluid/tests/book/test_image_classification.py b/python/paddle/fluid/tests/book/test_image_classification.py
index 9fe361425c128590da910128beaccb3336f8ba57..cba486cf5996a0b48da83bf83ab47dbf1153e2f4 100644
--- a/python/paddle/fluid/tests/book/test_image_classification.py
+++ b/python/paddle/fluid/tests/book/test_image_classification.py
@@ -223,7 +223,7 @@ def infer(use_cuda, save_dirname=None):
 
         # Use inference_transpiler to speedup
         inference_transpiler_program = inference_program.clone()
-        t = fluid.InferenceTranspiler()
+        t = fluid.transpiler.InferenceTranspiler()
         t.transpile(inference_transpiler_program, place)
 
         # Construct feed as a dictionary of {feed_target_name: feed_target_data}
diff --git a/python/paddle/fluid/transpiler/__init__.py b/python/paddle/fluid/transpiler/__init__.py
index 8429e2fd7c5141f064c66d8f406889bca1510fe2..28c7ae5341b20f0f79da8cf682d279fc4cc3fa19 100644
--- a/python/paddle/fluid/transpiler/__init__.py
+++ b/python/paddle/fluid/transpiler/__init__.py
@@ -20,6 +20,6 @@ from .memory_optimization_transpiler import memory_optimize, release_memory
 from .ps_dispatcher import HashName, RoundRobin
 
 __all__ = [
-    "DistributeTranspiler", "InferenceTranspiler", "memory_optimize",
-    "release_memory", "HashName", "RoundRobin", "DistributeTranspilerConfig"
+    "DistributeTranspiler", "memory_optimize", "release_memory", "HashName",
+    "RoundRobin", "DistributeTranspilerConfig"
 ]