diff --git a/python/paddle/fluid/tests/unittests/test_activation_op.py b/python/paddle/fluid/tests/unittests/test_activation_op.py
index 7d687dbd0f85f9235161dfa8528c849975aa5af3..124767a3364b078ea2c74795c03497f3dc24ba8c 100644
--- a/python/paddle/fluid/tests/unittests/test_activation_op.py
+++ b/python/paddle/fluid/tests/unittests/test_activation_op.py
@@ -70,25 +70,16 @@ class TestActivation(OpTest):
 
 
 class TestParameter(object):
-    def test_out(self):
-        with fluid.program_guard(fluid.Program()):
-            data = fluid.layers.data(name="X", shape=[1])
-            out = eval("paddle.%s(data, out=data)" % self.op_type)
-            place = fluid.CPUPlace()
-            exe = fluid.Executor(place)
-            result = exe.run(feed={"X": np.array([0.1])},
-                             fetch_list=[data, out])
-            self.assertEqual(result[0], result[1])
-
     def test_out_name(self):
         with fluid.program_guard(fluid.Program()):
+            np_x = np.array([0.1])
             data = fluid.layers.data(name="X", shape=[1])
-            out = eval("paddle.%s(data, name='Y', out=data)" % self.op_type)
+            out = eval("paddle.%s(data, name='Y')" % self.op_type)
             place = fluid.CPUPlace()
             exe = fluid.Executor(place)
-            result = exe.run(feed={"X": np.array([0.1])},
-                             fetch_list=[data, out])
-            self.assertEqual(result[0], result[1])
+            result, = exe.run(feed={"X": np_x}, fetch_list=[out])
+            expected = eval("np.%s(np_x)" % self.op_type)
+            self.assertEqual(result, expected)
 
     def test_dygraph(self):
         with fluid.dygraph.guard():
@@ -174,6 +165,17 @@ class TestAtan(TestActivation, TestParameter):
             return
         self.check_grad(['X'], 'Out')
 
+    def test_out_name(self):
+        with fluid.program_guard(fluid.Program()):
+            np_x = np.array([0.1])
+            data = fluid.layers.data(name="X", shape=[1])
+            out = paddle.atan(data, name='Y')
+            place = fluid.CPUPlace()
+            exe = fluid.Executor(place)
+            result, = exe.run(feed={"X": np_x}, fetch_list=[out])
+            expected = np.arctan(np_x)
+            self.assertEqual(result, expected)
+
     def test_dygraph(self):
         with fluid.dygraph.guard():
             np_x = np.array([0.1])
@@ -1034,21 +1036,18 @@ class TestPow_factor_tensor(TestActivation):
         factor_2 = fluid.layers.fill_constant([1], "float32", 3.0)
         out_1 = fluid.layers.pow(x, factor=factor_1)
         out_2 = fluid.layers.pow(x, factor=factor_2)
-        out_3 = paddle.pow(x, factor_1, out=res)
         out_4 = paddle.pow(x, factor_1, name='pow_res')
-        out_5 = paddle.pow(x, factor_1, out=res, name='pow_res')
         out_6 = paddle.pow(x, factor_2)
         self.assertEqual(('pow_res' in out_4.name), True)
 
         exe = fluid.Executor(place=fluid.CPUPlace())
-        res_1, res_2, res_3, res, res_6 = exe.run(
+        res_1, res_2, res, res_6 = exe.run(
             fluid.default_main_program(),
             feed={"x": input},
-            fetch_list=[out_1, out_2, out_3, res, out_6])
+            fetch_list=[out_1, out_2, res, out_6])
 
         assert np.array_equal(res_1, np.power(input, 2))
         assert np.array_equal(res_2, np.power(input, 3))
-        assert np.array_equal(res_3, res)
         assert np.array_equal(res_6, np.power(input, 3))
 
     def test_error(self):
diff --git a/python/paddle/fluid/tests/unittests/test_addcmul.py b/python/paddle/fluid/tests/unittests/test_addcmul.py
index 981df8cbdcdea6f72686a39ae56c2d2021ebc00d..6657ebe77ad57be224876ab83642338ef3f4bdb1 100644
--- a/python/paddle/fluid/tests/unittests/test_addcmul.py
+++ b/python/paddle/fluid/tests/unittests/test_addcmul.py
@@ -118,17 +118,6 @@ class TestAddcmul(unittest.TestCase):
             out = paddle.addcmul(input, tensor1, tensor2)
             self.assertEqual(out.shape, input.shape)
 
-    def test_addcmul_has_out(self):
-        program = Program()
-        with program_guard(program):
-            input = fluid.data(name='in', shape=[4, 100], dtype='float32')
-            tensor1 = fluid.data(name='t1', shape=[100], dtype='float32')
-            tensor2 = fluid.data(name='t2', shape=[100], dtype='float32')
-            out = fluid.data(name='out', shape=[4, 100], dtype='float32')
-
-            out = paddle.addcmul(input, tensor1, tensor2, out=out)
-            self.assertEqual(out.shape, input.shape)
-
 
 class InvalidInputTest(unittest.TestCase):
     def test_error(self):
diff --git a/python/paddle/fluid/tests/unittests/test_clamp.py b/python/paddle/fluid/tests/unittests/test_clamp.py
index ce18321ca9f5f5c66e13221830b22f0bba74d6fc..d8d7fe01f8de8686724ea8ebc00491269f2cc0bd 100644
--- a/python/paddle/fluid/tests/unittests/test_clamp.py
+++ b/python/paddle/fluid/tests/unittests/test_clamp.py
@@ -20,6 +20,18 @@ import unittest
 
 
 class TestClampAPI(unittest.TestCase):
+    def test_dygraph_clamp(self):
+        in1 = np.array([[1.2, 3.5], [4.5, 6.4]]).astype('float32')
+        with fluid.dygraph.guard():
+            x1 = fluid.dygraph.to_variable(in1)
+            out1 = tensor.clamp(x1, min=3.5, max=5.0)
+            out2 = tensor.clamp(x1, min=2.5)
+            self.assertTrue(
+                np.allclose(
+                    out1.numpy(), in1.clip(
+                        min=3.5, max=5.0)))
+            self.assertTrue(np.allclose(out2.numpy(), in1.clip(min=2.5)))
+
     def test_clamp(self):
         data_shape = [1, 9, 9, 4]
         data = np.random.random(data_shape).astype('float32')
diff --git a/python/paddle/fluid/tests/unittests/test_elementwise_add_op.py b/python/paddle/fluid/tests/unittests/test_elementwise_add_op.py
index 8339e598930e65976747907138256a60525cf46e..6eeb355a6ba3a9c20156ebfd1389d50e92a5a0f5 100644
--- a/python/paddle/fluid/tests/unittests/test_elementwise_add_op.py
+++ b/python/paddle/fluid/tests/unittests/test_elementwise_add_op.py
@@ -389,44 +389,6 @@ class TestElementwiseAddOpError(unittest.TestCase):
 
 
 class TestAddOp(unittest.TestCase):
-    def test_out(self):
-        with fluid.program_guard(fluid.Program()):
-            x = fluid.data(name="x", shape=[3], dtype="float32")
-            y = fluid.data(name='y', shape=[3], dtype='float32')
-
-            res = fluid.data(name="output", shape=[3], dtype="float32")
-            y_1 = paddle.add(x, y, out=res)
-
-            place = fluid.CPUPlace()
-            exe = fluid.Executor(place)
-            data1 = np.array([2, 3, 4], dtype='float32')
-            data2 = np.array([1, 5, 2], dtype='float32')
-            np_res, np_y_1 = exe.run(feed={'x': data1,
-                                           'y': data2},
-                                     fetch_list=[res, y_1])
-
-            self.assertEqual((np_res == np_y_1).all(), True)
-
-    def test_out_gpu(self):
-        if not fluid.core.is_compiled_with_cuda():
-            return
-        with fluid.program_guard(fluid.Program()):
-            x = fluid.data(name="x", shape=[3], dtype="float32")
-            y = fluid.data(name='y', shape=[3], dtype='float32')
-
-            res = fluid.data(name="output", shape=[3], dtype="float32")
-            y_1 = paddle.add(x, y, out=res)
-
-            place = fluid.CUDAPlace(0)
-            exe = fluid.Executor(place)
-            data1 = np.array([2, 3, 4], dtype='float32')
-            data2 = np.array([1, 5, 2], dtype='float32')
-            np_res, np_y_1 = exe.run(feed={'x': data1,
-                                           'y': data2},
-                                     fetch_list=[res, y_1])
-
-            self.assertEqual((np_res == np_y_1).all(), True)
-
     def test_name(self):
         with fluid.program_guard(fluid.Program()):
             x = fluid.data(name="x", shape=[2, 3], dtype="float32")
diff --git a/python/paddle/fluid/tests/unittests/test_elementwise_div_op.py b/python/paddle/fluid/tests/unittests/test_elementwise_div_op.py
index 91e0dc2b73849c0329397ee47a95bc2228d16063..de0fc591b664728387ccb988f3611fe034989627 100644
--- a/python/paddle/fluid/tests/unittests/test_elementwise_div_op.py
+++ b/python/paddle/fluid/tests/unittests/test_elementwise_div_op.py
@@ -241,44 +241,6 @@ class TestElementwiseDivBroadcast(unittest.TestCase):
 
 
 class TestDivOp(unittest.TestCase):
-    def test_out(self):
-        with fluid.program_guard(fluid.Program()):
-            x = fluid.data(name="x", shape=[3], dtype="float32")
-            y = fluid.data(name='y', shape=[3], dtype='float32')
-
-            res = fluid.data(name="output", shape=[3], dtype="float32")
-            y_1 = paddle.div(x, y, out=res)
-
-            place = fluid.CPUPlace()
-            exe = fluid.Executor(place)
-            data1 = np.array([2, 3, 4], dtype='float32')
-            data2 = np.array([1, 5, 2], dtype='float32')
-            np_res, np_y_1 = exe.run(feed={'x': data1,
-                                           'y': data2},
-                                     fetch_list=[res, y_1])
-
-            self.assertEqual((np_res == np_y_1).all(), True)
-
-    def test_out_gpu(self):
-        if not fluid.core.is_compiled_with_cuda():
-            return
-        with fluid.program_guard(fluid.Program()):
-            x = fluid.data(name="x", shape=[3], dtype="float32")
-            y = fluid.data(name='y', shape=[3], dtype='float32')
-
-            res = fluid.data(name="output", shape=[3], dtype="float32")
-            y_1 = paddle.div(x, y, out=res)
-
-            place = fluid.CUDAPlace(0)
-            exe = fluid.Executor(place)
-            data1 = np.array([2, 3, 4], dtype='float32')
-            data2 = np.array([1, 5, 2], dtype='float32')
-            np_res, np_y_1 = exe.run(feed={'x': data1,
-                                           'y': data2},
-                                     fetch_list=[res, y_1])
-
-            self.assertEqual((np_res == np_y_1).all(), True)
-
     def test_name(self):
         with fluid.program_guard(fluid.Program()):
             x = fluid.data(name="x", shape=[2, 3], dtype="float32")
diff --git a/python/paddle/fluid/tests/unittests/test_inverse_op.py b/python/paddle/fluid/tests/unittests/test_inverse_op.py
index 3aecbe6e00b3b06e0e4f764314691103c561467a..ca6bf685a9b90550dc2c31b94b014e9a2774a2cb 100644
--- a/python/paddle/fluid/tests/unittests/test_inverse_op.py
+++ b/python/paddle/fluid/tests/unittests/test_inverse_op.py
@@ -86,14 +86,10 @@ class TestInverseAPI(unittest.TestCase):
         if core.is_compiled_with_cuda():
             self.places.append(fluid.CUDAPlace(0))
 
-    def check_static_result(self, place, with_out=False):
+    def check_static_result(self, place):
         with fluid.program_guard(fluid.Program(), fluid.Program()):
             input = fluid.data(name="input", shape=[4, 4], dtype="float64")
-            if with_out:
-                out = fluid.data(name="output", shape=[4, 4], dtype="float64")
-            else:
-                out = None
-            result = paddle.inverse(input=input, out=out)
+            result = paddle.inverse(input=input)
 
             input_np = np.random.random([4, 4]).astype("float64")
             result_np = np.linalg.inv(input_np)
diff --git a/python/paddle/fluid/tests/unittests/test_kron_op.py b/python/paddle/fluid/tests/unittests/test_kron_op.py
index 1047f1bf1e5774a284dbe2e77fbf3fe89c2d7af2..68ad35489ce35aab8a207229733e9bf0142754cc 100644
--- a/python/paddle/fluid/tests/unittests/test_kron_op.py
+++ b/python/paddle/fluid/tests/unittests/test_kron_op.py
@@ -93,8 +93,7 @@ class TestKronLayer(unittest.TestCase):
             with fluid.program_guard(main, start):
                 a_var = fluid.data("a", [-1, -1], dtype="float64")
                 b_var = fluid.data("b", [-1, -1], dtype="float64")
-                out_var = fluid.layers.create_tensor("float64", "c")
-                paddle.kron(a_var, b_var, out=out_var)
+                out_var = paddle.kron(a_var, b_var)
 
         place = fluid.CPUPlace()
         exe = fluid.Executor(place)
diff --git a/python/paddle/fluid/tests/unittests/test_logsumexp.py b/python/paddle/fluid/tests/unittests/test_logsumexp.py
index 791f0307cd8b646508980c7676684f3a99ddb710..508b4a7b72da8affbc7ddf590b8142a41d1f3191 100644
--- a/python/paddle/fluid/tests/unittests/test_logsumexp.py
+++ b/python/paddle/fluid/tests/unittests/test_logsumexp.py
@@ -71,15 +71,6 @@ class TestLogSumExpOp(unittest.TestCase):
                         x, keepdim=True).numpy(),
                     np.log(np.sum(np.exp(np_x), keepdims=True))))
 
-            np_x = np.random.uniform(0.1, 1, [2, 3, 4]).astype(np.float32)
-            x = fluid.dygraph.to_variable(np_x)
-            helper = LayerHelper("test_logsumexp")
-            out = helper.create_variable(
-                type=x.type, name='out', dtype=x.dtype, persistable=False)
-            paddle.logsumexp(x, out=out)
-            self.assertTrue(
-                np.allclose(out.numpy(), np.log(np.sum(np.exp(np_x)))))
-
 
 if __name__ == '__main__':
     unittest.main()
diff --git a/python/paddle/fluid/tests/unittests/test_matmul_op.py b/python/paddle/fluid/tests/unittests/test_matmul_op.py
index 018b5c62862f075c4ce4243bf3c008bddcf416be..3eb822bfed89b80bccc08fe0d96b6c4b2f9b4ec4 100644
--- a/python/paddle/fluid/tests/unittests/test_matmul_op.py
+++ b/python/paddle/fluid/tests/unittests/test_matmul_op.py
@@ -245,23 +245,6 @@ for dim in [4]:
 
 class API_TestMm(unittest.TestCase):
     def test_out(self):
-        with fluid.program_guard(fluid.Program()):
-            x = fluid.data(name="x", shape=[3, 2], dtype="float64")
-            y = fluid.data(name='y', shape=[2, 3], dtype='float64')
-            res = fluid.data(name="output", shape=[3, 3], dtype="float64")
-            y_1 = paddle.mm(x, y, out=res)
-            exe = fluid.Executor(fluid.CPUPlace())
-            data1 = np.random.rand(3, 2)
-            data2 = np.random.rand(2, 3)
-            np_res, expected_result = exe.run(feed={'x': data1,
-                                                    'y': data2},
-                                              fetch_list=[res, y_1])
-        self.assertTrue(
-            np.allclose(
-                np.array(np_res), np.array(expected_result), atol=1e-5),
-            "two value is\
-            {}\n{}, check diff!".format(np_res, expected_result))
-
         with fluid.program_guard(fluid.Program()):
             x = fluid.data(name="x", shape=[2], dtype="float64")
             y = fluid.data(name='y', shape=[2], dtype='float64')
@@ -280,18 +263,6 @@ class API_TestMm(unittest.TestCase):
             "two value is\
             {}\n{}, check diff!".format(np_res, expected_result))
 
-    def test_dygraph_with_out(self):
-        device = fluid.CPUPlace()
-        with fluid.dygraph.guard(device):
-            input_array1 = np.random.rand(3, 4).astype("float64")
-            input_array2 = np.random.rand(4, 3).astype("float64")
-            out_array = np.random.rand(3, 3).astype("float64")
-            data1 = fluid.dygraph.to_variable(input_array1)
-            data2 = fluid.dygraph.to_variable(input_array2)
-            paddle_out_holder = fluid.dygraph.to_variable(out_array)
-            out = paddle.mm(data1, data2, out=paddle_out_holder)
-        self.assertTrue(np.allclose(paddle_out_holder.numpy(), out.numpy()))
-
     def test_dygraph_without_out(self):
         device = fluid.CPUPlace()
         with fluid.dygraph.guard(device):
diff --git a/python/paddle/tensor/math.py b/python/paddle/tensor/math.py
index 72cf76c5c725bc6664eb91f7aa8e18074a287169..878fdbfc1f5761317fb5f8a32bbee5f5ef7f5bc0 100644
--- a/python/paddle/tensor/math.py
+++ b/python/paddle/tensor/math.py
@@ -1,4 +1,4 @@
-#   Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+# 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.
@@ -140,7 +140,7 @@ def generate_op_noattr(op_type):
     """
     op_proto = OpProtoHolder.instance().get_op_proto(op_type)
 
-    def func(x, name=None, out=None):
+    def func(x, name=None):
         if in_dygraph_mode():
             op = getattr(core.ops, op_type)
             return op(x)
@@ -149,14 +149,7 @@ def generate_op_noattr(op_type):
                                  op_type)
         helper = LayerHelper(op_type, **locals())
 
-        if name and out:
-            warnings.warn(
-                "Both name and out parameters have been set in fluid.tensor.math.%s(), only out will take effect to specify the result storage. "
-                "You can discard either one to solve this warning." % op_type,
-                category=UserWarning,
-                stacklevel=2)
-        if not out:
-            out = helper.create_variable_for_type_inference(dtype=x.dtype)
+        out = helper.create_variable_for_type_inference(dtype=x.dtype)
         helper.append_op(type=op_type, inputs={"X": x}, outputs={"Out": out})
         return out
 
@@ -193,7 +186,7 @@ Examples:
     return func
 
 @templatedoc()
-def pow(input, exponent, out=None, name=None):
+def pow(input, exponent, name=None):
     """
 	:alias_main: paddle.pow
 	:alias: paddle.pow,paddle.tensor.pow,paddle.tensor.math.pow
@@ -205,8 +198,6 @@ def pow(input, exponent, out=None, name=None):
     Args:
         input(Variable): A ``Tensor`` or ``LoDTensor`` . The data type is ``float32`` or ``float64``.
         exponent(float32|Variable): A scalar with type ``float32`` or a ``Tensor`` with shape [1] and type ``float32``.
-        out (Variable, optional):  The Variable that stores results of the operation.
-            If out is None, a new Variable will be created to store the results.
         name(str, optional): The default value is None. Normally there is no need for user to set this property.
             For more information, please refer to :ref:`api_guide_Name` .
 
@@ -223,16 +214,17 @@ def pow(input, exponent, out=None, name=None):
             x = fluid.data(name="x", shape=[32,32], dtype="float32")
 
             # example 1: argument exponent is float
-            res = fluid.data(name="output", shape=[32,32], dtype="float32")
-            y_1 = paddle.pow(x, 2.0, out=res)
+            y_1 = paddle.pow(x, 2.0)
             # y_1 is x^{2.0}
 
             # example 2: argument exponent is Variable
             exponent_tensor = fluid.layers.fill_constant([1], "float32", 3.0)
-            res = fluid.data(name="output", shape=[32,32], dtype="float32")
-            y_2 = paddle.pow(x, exponent_tensor, out=res)
+            y_2 = paddle.pow(x, exponent_tensor)
             # y_2 is x^{3.0}
     """
+    if in_dygraph_mode():
+        return core.ops.pow(input, "exponent", exponent)
+
     helper = LayerHelper('pow', **locals())
     inputs = {'X': input}
     attrs = {}
@@ -242,22 +234,11 @@ def pow(input, exponent, out=None, name=None):
     else:
         attrs['factor'] = exponent
 
-    if out is None:
-        out = helper.create_variable_for_type_inference(dtype=input.dtype)
-    else:
-        check_dtype(
-            out.dtype, out.name,
-            convert_dtype(input.dtype), 'pow',
-            '(The out data type in pow must be the same with input data type.)')
-        if name:
-            warnings.warn(
-                "The output Variable name of the paddle.tensor.pow operation can only be given by parameter out or name. \
-                When parameter out and name are set at the same time, out has a higher priority than name. \
-                Finally, the output Variable name is same as the out name %s"
-                                                                              %
-                out.name,
-                category=UserWarning,
-                stacklevel=2)
+    out = helper.create_variable_for_type_inference(dtype=input.dtype)
+    check_dtype(
+        out.dtype, out.name,
+        convert_dtype(input.dtype), 'pow',
+        '(The out data type in pow must be the same with input data type.)')
 
     helper.append_op(
         type='pow', inputs=inputs, outputs={'Out': out}, attrs=attrs)
@@ -307,13 +288,11 @@ def _elementwise_op(helper):
     axis = helper.kwargs.get('axis', -1)
     use_mkldnn = helper.kwargs.get('use_mkldnn', False)
     name = helper.kwargs.get('name', None)
-    out = helper.kwargs.get('out', None)
-    if out is None:
-        if name is None:
-            out = helper.create_variable_for_type_inference(dtype=x.dtype)
-        else:
-            out = helper.create_variable(
-                name=name, dtype=x.dtype, persistable=False)
+    if name is None:
+        out = helper.create_variable_for_type_inference(dtype=x.dtype)
+    else:
+        out = helper.create_variable(
+            name=name, dtype=x.dtype, persistable=False)
 
     helper.append_op(
         type=op_type,
@@ -325,7 +304,7 @@ def _elementwise_op(helper):
     return helper.append_activation(out)
 
 
-def add(x, y, alpha=1, out=None, name=None):
+def add(x, y, alpha=1, name=None):
     """
 Examples:
 
@@ -417,9 +396,7 @@ Examples:
 
         x = fluid.data(name="x", shape=[3], dtype="float32")
         y = fluid.data(name='y', shape=[3], dtype='float32')
-
-        output = fluid.data(name="output", shape=[3], dtype="float32")
-        z = paddle.add(x, y, out=output)
+        z = paddle.add(x, y)
 
         place = fluid.CPUPlace()
         exe = fluid.Executor(place)
@@ -456,18 +433,10 @@ Examples:
         return _elementwise_op_in_dygraph(
             x, y, axis=axis, act=act, op_name=op_type)
 
-    original_op_type = 'add'
-    if name and out:
-        warnings.warn(
-            "Both name and out parameters have been set in paddle.tensor.%s, only out will take effect to specify the result storage. "
-            "You can discard either one to solve this warning." %
-            original_op_type,
-            category=UserWarning,
-            stacklevel=2)
     return _elementwise_op(LayerHelper(op_type, **locals()))
 
 
-def div(x, y, out=None, name=None):
+def div(x, y, name=None):
     """
 Examples:
 
@@ -537,8 +506,7 @@ Examples:
 
         x = fluid.data(name="x", shape=[2,3,4,5], dtype='float32')
         y = fluid.data(name="y", shape=[5], dtype='float32')
-        output = fluid.data(name="output", shape=[2,3,4,5], dtype="float32")
-        z = paddle.div(x, y, out=output)
+        z = paddle.div(x, y)
         # z = x / y
 
         place = fluid.CPUPlace()
@@ -573,14 +541,6 @@ Examples:
         return _elementwise_op_in_dygraph(
             x, y, axis=axis, act=act, op_name=op_type)
 
-    original_op_type = 'div'
-    if name and out:
-        warnings.warn(
-            "Both name and out parameters have been set in paddle.tensor.%s, only out will take effect to specify the result storage. "
-            "You can discard either one to solve this warning." %
-            original_op_type,
-            category=UserWarning,
-            stacklevel=2)
     return _elementwise_op(LayerHelper(op_type, **locals()))
 
 
@@ -634,9 +594,6 @@ for func in [
 
         additional_args_lines = [
             "alpha (int|float, optional): The alpha factor of the input. Default is 1. If alpha is not 1, the equation becomes Out = X + alpha * Y.",
-            "out (Variable, optinal): The Variable that stores results of the operation. Default is None. If out is None, \
-            a new Variable will be created to store the results."
-                                                                 ,
             "name (string, optional): Name of the output. \
             Default is None. It's used to print debug info for developers. Details: \
             :ref:`api_guide_Name` "
@@ -858,7 +815,7 @@ def elementwise_sum(inputs, name=None):
     return out
 
 
-def mm(input, mat2, out=None, name=None):
+def mm(input, mat2, name=None):
     """
 	:alias_main: paddle.mm
 	:alias: paddle.mm,paddle.tensor.mm,paddle.tensor.math.mm
@@ -876,9 +833,6 @@ def mm(input, mat2, out=None, name=None):
     Args:
         x (Variable): The input variable which is a Tensor or LoDTensor.
         mat2 (Variable): The input variable which is a Tensor or LoDTensor.
-        out(Variable, optional): Optional output which can be any created
-            Variable that meets the requirements to store the result of operation.
-            if out is None, a new Varibale will be create to store the result.
         name(str, optional): The default value is None. Normally there is no need for
             user to set this property. For more information, please refer to :ref:`api_guide_Name`
 
@@ -914,8 +868,7 @@ def mm(input, mat2, out=None, name=None):
             out = paddle.mm(x, mat2) # out shape is [2, 2]
     """
     if in_dygraph_mode():
-        if out is None:
-            out = _varbase_creator(dtype=input.dtype)
+        out = _varbase_creator(dtype=input.dtype)
         core.ops.matmul(input, mat2, out)
         return out
 
@@ -955,8 +908,7 @@ def mm(input, mat2, out=None, name=None):
     __check_input(input, mat2)
 
     helper = LayerHelper('mm', **locals())
-    if out is None:
-        out = helper.create_variable_for_type_inference(dtype=input.dtype)
+    out = helper.create_variable_for_type_inference(dtype=input.dtype)
     helper.append_op(
         type='matmul', inputs={'X': input,
                                'Y': mat2}, outputs={'Out': out})
@@ -1032,7 +984,7 @@ def addmm(input, x, y, alpha=1.0, beta=1.0, name=None):
     return out
 
 
-def logsumexp(x, dim=None, keepdim=False, out=None, name=None):
+def logsumexp(x, dim=None, keepdim=False, name=None):
     """
 	:alias_main: paddle.logsumexp
 	:alias: paddle.logsumexp,paddle.tensor.logsumexp,paddle.tensor.math.logsumexp
@@ -1052,7 +1004,6 @@ def logsumexp(x, dim=None, keepdim=False, out=None, name=None):
        keep_dim (bool, optional): Whether to reserve the reduced dimension in the output Tensor.
          The result tensor will have one fewer dimension than the :attr:`input` unless :attr:`keep_dim`
          is true, default value is False.
-       out (Variable), optional):  Enable user to explicitly specify an output variable to save result.
        name (str, optional): The default value is None.  Normally there is no need for user to
          set this property.  For more information, please refer to :ref:`api_guide_Name`
 
@@ -1094,16 +1045,10 @@ def logsumexp(x, dim=None, keepdim=False, out=None, name=None):
     exp_out = layers.exp(x)
     sum_out = layers.reduce_sum(exp_out, dim, keepdim)
 
-    if out is not None:
-        check_variable_and_dtype(out, 'out', [x.dtype], op_type)
-        helper = LayerHelper(op_type, **locals())
-        helper.append_op(type="log", inputs={"X": sum_out}, outputs={"Out": out})
-        return out
-
     return layers.log(sum_out, name)
 
 
-def inverse(input, out=None, name=None):
+def inverse(input, name=None):
     """
 	:alias_main: paddle.inverse
 	:alias: paddle.inverse,paddle.tensor.inverse,paddle.tensor.math.inverse
@@ -1117,9 +1062,6 @@ def inverse(input, out=None, name=None):
             dimensions should be equal. When the number of dimensions is
             greater than 2, it is treated as batches of square matrix. The data
             type can be float32 and float64.
-        out (Variable, optional): Optional output which can be any created
-            Variable that meets the requirements to store the result of operation.
-            If out is None, a new Varibale will be create to store the result.
         name (str, optional): The default value is None. Normally there is no need for
             user to set this property. For more information,
             please refer to :ref:`api_guide_Name`
@@ -1165,20 +1107,16 @@ def inverse(input, out=None, name=None):
                 "of dimensions is no less than 2. But reviced: %d, "
                 "input's shape: %s." % (len(input.shape), input.shape))
 
-        if out is not None:
-            check_variable_and_dtype(out, 'out', input.dtype, 'inverse')
-
     _check_input(input)
 
     helper = LayerHelper('inverse', **locals())
-    if out is None:
-        out = helper.create_variable_for_type_inference(dtype=input.dtype)
+    out = helper.create_variable_for_type_inference(dtype=input.dtype)
     helper.append_op(
         type='inverse', inputs={'Input': [input] }, outputs={'Output': [out]})
     return out
 
 
-def max(input, dim=None, keep_dim=False, out=None, name=None):
+def max(input, dim=None, keep_dim=False, name=None):
     """
 	:alias_main: paddle.max
 	:alias: paddle.max,paddle.tensor.max,paddle.tensor.math.max
@@ -1197,9 +1135,6 @@ def max(input, dim=None, keep_dim=False, out=None, name=None):
             output Tensor. The result tensor will have one fewer dimension
             than the :attr:`input` unless :attr:`keep_dim` is true, default
             value is False.
-        out(Variable, optional): Optional output which can be any created
-            Variable that meets the requirements to store the result of operation.
-            if out is None, a new Varibale will be create to store the result.
         name(str, optional): The default value is None.  Normally there is no need for
             user to set this property.  For more information, please refer to :ref:`api_guide_Name`
 
@@ -1231,8 +1166,7 @@ def max(input, dim=None, keep_dim=False, out=None, name=None):
     """
 
     helper = LayerHelper('max', **locals())
-    if out is None:
-        out = helper.create_variable_for_type_inference(
+    out = helper.create_variable_for_type_inference(
             dtype=helper.input_dtype())
     if dim is not None and not isinstance(dim, list):
         dim = [dim]
@@ -1258,7 +1192,7 @@ def max(input, dim=None, keep_dim=False, out=None, name=None):
     return out
 
 
-def min(input, dim=None, keep_dim=False, out=None, name=None):
+def min(input, dim=None, keep_dim=False, name=None):
     """
 	:alias_main: paddle.min
 	:alias: paddle.min,paddle.tensor.min,paddle.tensor.math.min
@@ -1277,10 +1211,7 @@ def min(input, dim=None, keep_dim=False, out=None, name=None):
             output Tensor. The result tensor will have one fewer dimension
             than the :attr:`input` unless :attr:`keep_dim` is true, default
             value is False.
-        out(Variable, optional): Optional output which can be any created
-            Variable that meets the requirements to store the result of operation.
-            if out is None, a new Varibale will be create to store the result.
-        name(str, optional): The default value is None.  Normally there is no need for
+        name(str, optional): The default value is None.  Normally there is no need for 
             user to set this property.  For more information, please refer to :ref:`api_guide_Name`
 
     Returns:
@@ -1310,9 +1241,8 @@ def min(input, dim=None, keep_dim=False, out=None, name=None):
     """
 
     helper = LayerHelper('min', **locals())
-    if out is None:
-        out = helper.create_variable_for_type_inference(
-            dtype=helper.input_dtype())
+    out = helper.create_variable_for_type_inference(
+        dtype=helper.input_dtype())
     if dim is not None and not isinstance(dim, list):
         dim = [dim]
 
@@ -1380,7 +1310,7 @@ def log1p(x, name=None):
     return out
 
 
-def addcmul(input, tensor1, tensor2, value=1.0, out=None, name=None):
+def addcmul(input, tensor1, tensor2, value=1.0, name=None):
     """
 	:alias_main: paddle.addcmul
 	:alias: paddle.addcmul,paddle.tensor.addcmul,paddle.tensor.math.addcmul
@@ -1396,10 +1326,6 @@ def addcmul(input, tensor1, tensor2, value=1.0, out=None, name=None):
         tensor1(Variable): The tensor to be multiplied. A Tensor with type float32, float64, int32, int64.
         tensor2(Variable): The tensor to be multiplied. A Tensor with type float32, float64, int32, int64.
         value(int|float): The multiplier for tensor1*tensor2. For float32 and float64 type input, value must be float, otherwise an integer.
-        out(Variable, Optional): The variable that specifies the output of the
-            operator, which can be Variable that has been created in the
-            program. The default value is None, and a new Variable will be
-            created to save the output. Default: None.
         name(str, Optional): For details, please refer to :ref:`api_guide_Name`.
                         Generally, no setting is required. Default: None.
     Returns:
@@ -1422,14 +1348,11 @@ def addcmul(input, tensor1, tensor2, value=1.0, out=None, name=None):
     if convert_dtype(input.dtype) in ['int32', 'int64']:
         check_type(value, 'value', int, 'addcmul')
 
-    if out is not None:
-        layers.assign(layers.elementwise_add(input, layers.elementwise_mul(tensor1, tensor2) * value), out)
-    else:
-        out = layers.elementwise_add(input, layers.elementwise_mul(tensor1, tensor2) * value)
+    out = layers.elementwise_add(input, layers.elementwise_mul(tensor1, tensor2) * value)
     return out
 
 
-def clamp(input, min=None, max=None, output=None, name=None):
+def clamp(input, min=None, max=None, name=None):
     """
 	:alias_main: paddle.clamp
 	:alias: paddle.clamp,paddle.tensor.clamp,paddle.tensor.math.clamp
@@ -1450,8 +1373,6 @@ def clamp(input, min=None, max=None, output=None, name=None):
             with shape [1] and type ``int32``, ``float32``, ``float64``.
         max (float32|Variable): The upper bound with type ``float32`` or a ``Tensor``
             with shape [1] and type ``int32``, ``float32``, ``float64``.
-        output (Variable, optional): A tensor or LoDTensor. If :attr:`output` is None,
-            a new tensor will be created as :attr:`output`. Default: None.
         name (str, optional): The default value is None. Normally there is no
             need for user to set this property. For more information, please
             refer to :ref:`api_guide_Name`.
@@ -1482,6 +1403,11 @@ def clamp(input, min=None, max=None, output=None, name=None):
 
     assert min is not None or max is not None, "either min or max should be defined."
 
+    if in_dygraph_mode():
+        min = sys.float_info.min if min is None else min
+        max = sys.float_info.max if max is None else max
+        return core.ops.clip(input, "min", min, "max", max)
+
     if min is not None:
         check_type(min, 'min', (float, Variable), 'clamp')
         if isinstance(min, Variable):
@@ -1509,14 +1435,14 @@ def clamp(input, min=None, max=None, output=None, name=None):
         attrs['max'] = max
 
     helper = LayerHelper('clamp', **locals())
-    if output is None:
-        output = helper.create_variable_for_type_inference(
+    output = helper.create_variable_for_type_inference(
             dtype=helper.input_dtype())
     helper.append_op(
         type='clip', inputs=inputs, outputs={'Out': [output]}, attrs=attrs)
 
     return output
 
+
 def trace(x, offset=0, axis1=0, axis2=1, name=None):
     """
 	:alias_main: paddle.trace
@@ -1611,7 +1537,7 @@ def trace(x, offset=0, axis1=0, axis2=1, name=None):
     return out
 
 @templatedoc(op_type="kron")
-def kron(x, y, out=None, name=None):
+def kron(x, y, name=None):
     """
 	:alias_main: paddle.kron
 	:alias: paddle.kron,paddle.tensor.kron,paddle.tensor.math.kron
@@ -1624,10 +1550,6 @@ ${comment}
         y (Variable): the second operand of kron op, data type: float16,
             float32, float64, int32 or int64. Its data type should be the same
             with x.
-        out (Variable, optional): Optional output which can be any created
-            Variable that meets the requirements to store the result of
-            operation. If out is None, a new Varibale will be create to store
-            the result. Defaults to None.
         name(str, optional): The default value is None.  Normally there is no
             need for user to set this property.  For more information, please
             refer to :ref:`api_guide_Name`.
@@ -1668,9 +1590,6 @@ ${comment}
     check_variable_and_dtype(x, 'x', ['float16', 'float32', 'float64', 'int32', 'int64'], 'kron')
     check_variable_and_dtype(y, 'y', ['float16', 'float32', 'float64', 'int32', 'int64'], 'kron')
 
-    if out is None:
-        out = helper.create_variable_for_type_inference(dtype=x.dtype)
-    else:
-        check_variable_and_dtype(out, 'out', ['float16', 'float32', 'float64', 'int32', 'int64'], 'kron')
+    out = helper.create_variable_for_type_inference(dtype=x.dtype)
     helper.append_op(type="kron", inputs={"X": x, "Y": y}, outputs={"Out": out})
     return out
diff --git a/tools/check_api_approvals.sh b/tools/check_api_approvals.sh
index 7d939b93aadc767d6f3a05c73041efeac58c9927..b1f3f84b36ee295529661cf74b13e71d620254c9 100644
--- a/tools/check_api_approvals.sh
+++ b/tools/check_api_approvals.sh
@@ -85,7 +85,7 @@ api_spec_diff=`python ${PADDLE_ROOT}/tools/check_api_source_without_core_ops.py
 if [ "$api_spec_diff" != "" ]; then
     echo_line="You must have one RD (zhiqiu (Recommend) or phlrain) approval for the api change for the opreator-related api without 'core.ops'.\n"
     echo_line="${echo_line}For more details, please click [https://github.com/PaddlePaddle/Paddle/wiki/paddle_api_development_manual.md]\n"
-    echo_line="${echo_line}Related APIs: ${api_spec_diff}"
+    echo_line="${echo_line}Related APIs: ${api_spec_diff}\n"
     check_approval 1 6888866 43953930
 fi