diff --git a/python/paddle/fluid/layers/nn.py b/python/paddle/fluid/layers/nn.py
index a140894990bb16f66387203366102f16ccca08f3..bd5b11e1364b57453392eb2b28558559247298e6 100644
--- a/python/paddle/fluid/layers/nn.py
+++ b/python/paddle/fluid/layers/nn.py
@@ -120,7 +120,6 @@ __all__ = [
     'pad2d',
     'unique',
     'unique_with_counts',
-    'expand',
     'scale',
     'elementwise_add',
     'elementwise_div',
@@ -7970,128 +7969,6 @@ def flatten(x, axis=1, name=None):
     return out
 
 
-@deprecated(since='2.0.0', update_to="paddle.expand")
-def expand(x, expand_times, name=None):
-    """
-    :alias_main: paddle.expand
-        :alias: paddle.expand,paddle.tensor.expand,paddle.tensor.manipulation.expand
-        :old_api: paddle.fluid.layers.expand
-
-    This operation tiles ``x`` multiple times according to the parameter ``expand_times``.
-    The times number for each dimension of ``x`` is set by the parameter ``expand_times``.
-    The rank of ``x`` should be less than or equal to 6. Please note that size of ``expand_times`` must be the same
-    with X's rank. Following is a using case:
-
-
-    .. code-block:: text
-
-        Input(X) is a 3-D tensor with shape [2, 3, 1]:
-
-                [
-                   [[1], [2], [3]],
-                   [[4], [5], [6]]
-                ]
-
-        Attr(expand_times):  [1, 2, 2]
-
-        Output(Out) is a 3-D tensor with shape [2, 6, 2]:
-
-                [
-                    [[1, 1], [2, 2], [3, 3], [1, 1], [2, 2], [3, 3]],
-                    [[4, 4], [5, 5], [6, 6], [4, 4], [5, 5], [6, 6]]
-                ]
-
-    Args:
-        x (Variable): A ``Tensor`` or ``LoDTensor`` with dimension in [1, 6]. The data type is ``bool``, ``float32``, ``float64`` or ``int32`` .
-        expand_times (list|tuple|Variable): The data type is ``int32`` . If ``expand_times`` is a list or tuple, the elements of
-                it should be integers or Tensors with shape [1]. If ``expand_times`` is an Variable, it should be an 1-D Tensor.
-                Expand times number for each dimension of ``x`` .
-        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:
-        Variable: A ``Tensor`` or ``LoDTensor``. The data type is same as ``x``. After expanding, size of each dimension of output is equal to the size of the corresponding dimension of ``x`` multiplying the corresponding value given by ``expand_times`` .
-
-    Raises:
-        TypeError: The type of ``expand_times`` must be list, tuple or Variable.
-        ValueError: The elements of ``expand_times`` cannot be negative.
-
-    Examples:
-        .. code-block:: python
-
-            import paddle.fluid as fluid
-
-            # example 1:
-            data_1 = fluid.layers.fill_constant(shape=[2, 3, 1], dtype='int32', value=0)
-            expanded_1 = fluid.layers.expand(data_1, expand_times=[1, 2, 2])
-            # the shape of expanded_1 is [2, 6, 2].
-
-            # example 2:
-            data_2 = fluid.layers.fill_constant(shape=[12, 14], dtype="int32", value=3)
-            expand_times = fluid.layers.fill_constant(shape=[2], dtype="int32", value=4)
-            expanded_2 = fluid.layers.expand(data_2, expand_times=expand_times)
-            # the shape of expanded_2 is [48, 56].
-    """
-    if _non_static_mode():
-        attrs = ()
-        expand_times_tensor = None
-        if isinstance(expand_times, (list, tuple)):
-            expand_times = [
-                item.numpy().item(0) if isinstance(item, Variable) else item
-                for item in expand_times
-            ]
-            attrs += ('expand_times', expand_times)
-        elif isinstance(expand_times, Variable):
-            expand_times_tensor = expand_times
-            expand_times_tensor.stop_gradient = True
-
-        return _legacy_C_ops.expand(x, expand_times_tensor, *attrs)
-
-    inputs = {"X": [x]}
-    attrs = {}
-    check_variable_and_dtype(
-        x,
-        'x',
-        ['bool', 'float16', 'float32', 'float64', 'int32', 'int64'],
-        'expand',
-    )
-    check_type(expand_times, 'expand_times', (list, tuple, Variable), 'expand')
-    if convert_dtype(x.dtype) == 'bool' and x.stop_gradient == True:
-        raise ValueError(
-            "expand op bool date type must set the stop_gradient to be False"
-        )
-
-    helper = LayerHelper('expand', input=x, **locals())
-
-    def get_attr_expand_times(list_expand_times):
-        attrs_expand_times = []
-        for idx, times in enumerate(list_expand_times):
-            if isinstance(times, Variable):
-                attrs_expand_times.append(-1)
-            else:
-                attrs_expand_times.append(times)
-                assert (
-                    times > 0
-                ), "Each element given in expand_times must not be negative."
-        return attrs_expand_times
-
-    if isinstance(expand_times, Variable):
-        expand_times.stop_gradient = True
-        inputs['ExpandTimes'] = expand_times
-    elif isinstance(expand_times, (list, tuple)):
-        attrs['expand_times'] = get_attr_expand_times(expand_times)
-        if utils._contain_var(expand_times):
-            inputs['expand_times_tensor'] = utils._convert_to_tensor_list(
-                expand_times
-            )
-
-    dtype = helper.input_dtype(input_param_name='x')
-    out = helper.create_variable_for_type_inference(dtype)
-    helper.append_op(
-        type='expand', inputs=inputs, outputs={'Out': out}, attrs=attrs
-    )
-    return out
-
-
 from paddle.fluid.framework import convert_np_dtype_to_dtype_
 
 
diff --git a/python/paddle/fluid/tests/unittests/dygraph_to_static/seq2seq_dygraph_model.py b/python/paddle/fluid/tests/unittests/dygraph_to_static/seq2seq_dygraph_model.py
index 4b52df98c22d590cc8e4b8a701df87e23e10df4c..c650870f75d08ad8d074fa57b858acf71f2b3170 100644
--- a/python/paddle/fluid/tests/unittests/dygraph_to_static/seq2seq_dygraph_model.py
+++ b/python/paddle/fluid/tests/unittests/dygraph_to_static/seq2seq_dygraph_model.py
@@ -186,9 +186,9 @@ class BaseModel(fluid.dygraph.Layer):
 
     def _expand_to_beam_size(self, x):
         x = fluid.layers.unsqueeze(x, [1])
-        expand_times = [1] * len(x.shape)
-        expand_times[1] = self.beam_size
-        x = fluid.layers.expand(x, expand_times)
+        expand_shape = [-1] * len(x.shape)
+        expand_shape[1] = self.beam_size * x.shape[1]
+        x = paddle.expand(x, expand_shape)
         return x
 
     def _real_state(self, state, new_state, step_mask):
@@ -386,19 +386,20 @@ class BaseModel(fluid.dygraph.Layer):
                 [[0.0] + [-self.kinf] * (self.beam_size - 1)], dtype="float32"
             )
         )
-        beam_state_log_probs = fluid.layers.expand(
-            beam_state_log_probs, [self.batch_size, 1]
+        beam_state_log_probs = paddle.expand(
+            beam_state_log_probs,
+            [self.batch_size * beam_state_log_probs.shape[0], -1],
         )
         dec_hidden, dec_cell = enc_hidden, enc_cell
         dec_hidden = [self._expand_to_beam_size(ele) for ele in dec_hidden]
         dec_cell = [self._expand_to_beam_size(ele) for ele in dec_cell]
 
-        batch_pos = fluid.layers.expand(
+        batch_pos = paddle.expand(
             fluid.layers.unsqueeze(
                 to_variable(np.arange(0, self.batch_size, 1, dtype="int64")),
                 [1],
             ),
-            [1, self.beam_size],
+            [-1, self.beam_size],
         )
         predicted_ids = []
         parent_ids = []
@@ -442,9 +443,9 @@ class BaseModel(fluid.dygraph.Layer):
             )
 
             step_log_probs = fluid.layers.elementwise_mul(
-                fluid.layers.expand(
+                paddle.expand(
                     fluid.layers.unsqueeze(beam_finished, [2]),
-                    [1, 1, self.tar_vocab_size],
+                    [-1, -1, self.tar_vocab_size],
                 ),
                 noend_mask_tensor,
                 axis=-1,
@@ -650,9 +651,9 @@ class AttentionModel(fluid.dygraph.Layer):
 
     def tile_beam_merge_with_batch(self, x):
         x = fluid.layers.unsqueeze(x, [1])  # [batch_size, 1, ...]
-        expand_times = [1] * len(x.shape)
-        expand_times[1] = self.beam_size
-        x = fluid.layers.expand(x, expand_times)  # [batch_size, beam_size, ...]
+        expand_shape = [-1] * len(x.shape)
+        expand_shape[1] = self.beam_size * x.shape[1]
+        x = paddle.expand(x, expand_shape)  # [batch_size, beam_size, ...]
         x = paddle.transpose(
             x, list(range(2, len(x.shape))) + [0, 1]
         )  # [..., batch_size, beam_size]
@@ -670,9 +671,9 @@ class AttentionModel(fluid.dygraph.Layer):
 
     def _expand_to_beam_size(self, x):
         x = fluid.layers.unsqueeze(x, [1])
-        expand_times = [1] * len(x.shape)
-        expand_times[1] = self.beam_size
-        x = fluid.layers.expand(x, expand_times)
+        expand_shape = [-1] * len(x.shape)
+        expand_shape[1] = self.beam_size * x.shape[1]
+        x = paddle.expand(x, expand_shape)
         return x
 
     def _real_state(self, state, new_state, step_mask):
diff --git a/python/paddle/fluid/tests/unittests/dygraph_to_static/test_sentiment.py b/python/paddle/fluid/tests/unittests/dygraph_to_static/test_sentiment.py
index 13593e1c02c082a4055b384b907628fe3c3c3cab..ed7a3b8cc2f6e09491bd016c958882cee26a2e14 100644
--- a/python/paddle/fluid/tests/unittests/dygraph_to_static/test_sentiment.py
+++ b/python/paddle/fluid/tests/unittests/dygraph_to_static/test_sentiment.py
@@ -95,7 +95,7 @@ class CNN(fluid.dygraph.Layer):
         o_np_mask = (paddle.reshape(inputs, [-1, 1]) != self.dict_dim).astype(
             dtype='float32'
         )
-        mask_emb = fluid.layers.expand(o_np_mask, [1, self.hid_dim])
+        mask_emb = paddle.expand(o_np_mask, [-1, self.hid_dim])
         emb = emb * mask_emb
         emb = paddle.reshape(
             emb, shape=[-1, self.channels, self.seq_len, self.hid_dim]
@@ -141,7 +141,7 @@ class BOW(fluid.dygraph.Layer):
         o_np_mask = (paddle.reshape(inputs, [-1, 1]) != self.dict_dim).astype(
             dtype='float32'
         )
-        mask_emb = fluid.layers.expand(o_np_mask, [1, self.hid_dim])
+        mask_emb = paddle.expand(o_np_mask, [-1, self.hid_dim])
         emb = emb * mask_emb
         emb = paddle.reshape(emb, shape=[-1, self.seq_len, self.hid_dim])
         bow_1 = fluid.layers.reduce_sum(emb, dim=1)
@@ -189,7 +189,7 @@ class GRU(fluid.dygraph.Layer):
         o_np_mask = (paddle.reshape(inputs, [-1, 1]) != self.dict_dim).astype(
             'float32'
         )
-        mask_emb = fluid.layers.expand(o_np_mask, [1, self.hid_dim])
+        mask_emb = paddle.expand(o_np_mask, [-1, self.hid_dim])
         emb = emb * mask_emb
         emb = paddle.reshape(emb, shape=[self.batch_size, -1, self.hid_dim])
         fc_1 = self._fc1(emb)
@@ -243,7 +243,8 @@ class BiGRU(fluid.dygraph.Layer):
         o_np_mask = (paddle.reshape(inputs, [-1, 1]) != self.dict_dim).astype(
             'float32'
         )
-        mask_emb = fluid.layers.expand(o_np_mask, [1, self.hid_dim])
+        mask_emb = paddle.expand(o_np_mask, [-1, self.hid_dim])
+
         emb = emb * mask_emb
         emb = paddle.reshape(emb, shape=[self.batch_size, -1, self.hid_dim])
         fc_1 = self._fc1(emb)
diff --git a/python/paddle/fluid/tests/unittests/dygraph_to_static/transformer_dygraph_model.py b/python/paddle/fluid/tests/unittests/dygraph_to_static/transformer_dygraph_model.py
index ee11e045d9aada4a0bd48c34e15bb1dcd8bec200..18c94262c913b436e6e94b10345b4f380be0d098 100644
--- a/python/paddle/fluid/tests/unittests/dygraph_to_static/transformer_dygraph_model.py
+++ b/python/paddle/fluid/tests/unittests/dygraph_to_static/transformer_dygraph_model.py
@@ -701,9 +701,9 @@ class Transformer(Layer):
             tensor = paddle.reshape(
                 tensor, [tensor.shape[0], 1] + list(tensor.shape[1:])
             )
-            tile_dims = [1] * len(tensor.shape)
+            tile_dims = [-1] * len(tensor.shape)
             tile_dims[1] = beam_size
-            return layers.expand(tensor, tile_dims)
+            return paddle.expand(tensor, tile_dims)
 
         def merge_batch_beams(tensor):
             var_dim_in_state = 2  # count in beam dim
@@ -757,8 +757,9 @@ class Transformer(Layer):
         def mask_probs(probs, finished, noend_mask_tensor):
             finished = layers.cast(finished, dtype=probs.dtype)
             probs = layers.elementwise_mul(
-                layers.expand(
-                    layers.unsqueeze(finished, [2]), [1, 1, self.trg_vocab_size]
+                paddle.expand(
+                    layers.unsqueeze(finished, [2]),
+                    [-1, -1, self.trg_vocab_size],
                 ),
                 noend_mask_tensor,
                 axis=-1,
@@ -785,11 +786,11 @@ class Transformer(Layer):
         noend_array = [-inf] * self.trg_vocab_size
         noend_array[eos_id] = 0
         noend_mask_tensor = to_variable(np.array(noend_array, dtype="float32"))
-        batch_pos = layers.expand(
+        batch_pos = paddle.expand(
             layers.unsqueeze(
                 to_variable(np.arange(0, batch_size, 1, dtype="int64")), [1]
             ),
-            [1, beam_size],
+            [-1, beam_size],
         )
         predict_ids = []
         parent_ids = []
diff --git a/python/paddle/fluid/tests/unittests/ipu/test_expand_op_ipu.py b/python/paddle/fluid/tests/unittests/ipu/test_expand_op_ipu.py
index 784a6a41a41f563d650ba559252739f3c3af54cb..d0cf2d5be951041cdde86a6e37727f116b621a97 100644
--- a/python/paddle/fluid/tests/unittests/ipu/test_expand_op_ipu.py
+++ b/python/paddle/fluid/tests/unittests/ipu/test_expand_op_ipu.py
@@ -46,7 +46,7 @@ class TestBase(IPUOpTest):
         x = paddle.static.data(
             name=self.feed_list[0], shape=self.feed_shape[0], dtype="float32"
         )
-        out = paddle.fluid.layers.expand(x, **self.attrs)
+        out = paddle.expand(x, **self.attrs)
         self.fetch_list = [out.name]
 
     def run_model(self, exec_mode):
@@ -82,9 +82,7 @@ class TestCase1(TestBase):
         expand_times = paddle.fluid.layers.fill_constant(
             shape=[len(self.feed_shape[0])], dtype="int32", value=2
         )
-        out = paddle.fluid.layers.expand(
-            x, expand_times=expand_times, **self.attrs
-        )
+        out = paddle.expand(x, expand_times, **self.attrs)
         self.fetch_list = [out.name]
 
 
diff --git a/python/paddle/fluid/tests/unittests/npu/test_expand_op_npu.py b/python/paddle/fluid/tests/unittests/npu/test_expand_op_npu.py
index 0ac8cfc5d3c13de6a720070dab34e869251e4d43..5b4bef5361b4cfb8bbda58d2e55477a93e6a523c 100644
--- a/python/paddle/fluid/tests/unittests/npu/test_expand_op_npu.py
+++ b/python/paddle/fluid/tests/unittests/npu/test_expand_op_npu.py
@@ -97,7 +97,7 @@ class TestExpandNet(unittest.TestCase):
                 name="label", shape=[32, 1], dtype='int64'
             )
 
-            res = paddle.fluid.layers.expand(a, [1, 32])
+            res = paddle.expand(a, [-1, 32])
             loss = res.sum()
             sgd = fluid.optimizer.SGD(learning_rate=0.01)
             sgd.minimize(loss)
diff --git a/python/paddle/fluid/tests/unittests/test_dynamic_rnn_stop_gradient.py b/python/paddle/fluid/tests/unittests/test_dynamic_rnn_stop_gradient.py
index f3f971b5778d8fbefe16891c33caeaf94cae4844..d20ad3d0c0ec673bcfa39e52e791bb4a82fe0cb6 100644
--- a/python/paddle/fluid/tests/unittests/test_dynamic_rnn_stop_gradient.py
+++ b/python/paddle/fluid/tests/unittests/test_dynamic_rnn_stop_gradient.py
@@ -43,9 +43,9 @@ def build_and_run_program(place, batch_size, beam_size, stop_gradient=False):
         for _ in range(20):
             bs = layers.cast(bs, 'int64')
         bs.stop_gradient = stop_gradient
-        batch_pos = layers.expand(
+        batch_pos = paddle.expand(
             layers.unsqueeze(paddle.arange(0, bs, 1, dtype=bs.dtype), [1]),
-            [1, beam_size],
+            [-1, beam_size],
         )
         topk_coordinates = paddle.stack([batch_pos, indices], axis=2)
         topk_coordinates.stop_gradient = stop_gradient
diff --git a/python/paddle/fluid/tests/unittests/test_eager_deletion_while_op.py b/python/paddle/fluid/tests/unittests/test_eager_deletion_while_op.py
index 37ee4897e77f27b2b17df8767c2e9533fe99adc5..8c359a39195e3722427a464d8e0f52ffd6430c95 100644
--- a/python/paddle/fluid/tests/unittests/test_eager_deletion_while_op.py
+++ b/python/paddle/fluid/tests/unittests/test_eager_deletion_while_op.py
@@ -128,7 +128,7 @@ class TestEagerDeletionWhileOpBase(unittest.TestCase):
         sum_result = layers.array_read(array=mem_array, i=j)
         sum_result.persistable = True
         tmp = layers.unsqueeze(sum_result, axes=[0])
-        tmp = layers.expand(tmp, expand_times=[10, 1])
+        tmp = paddle.expand(tmp, [10, -1])
         fc = layers.fc(tmp, size=256)
         loss = paddle.mean(sum_result)
 
diff --git a/python/paddle/fluid/tests/unittests/test_expand_op.py b/python/paddle/fluid/tests/unittests/test_expand_op.py
index fd3dac2472e7c8089656d17effd36765431fade4..9fb4c7c804e0178dff50341a8513264aacd1141d 100644
--- a/python/paddle/fluid/tests/unittests/test_expand_op.py
+++ b/python/paddle/fluid/tests/unittests/test_expand_op.py
@@ -16,8 +16,6 @@ import unittest
 import numpy as np
 from op_test import OpTest
 import paddle.fluid as fluid
-from paddle.fluid import Program, program_guard
-import paddle
 
 
 # Situation 1: expand_times is a list(without tensor)
@@ -201,62 +199,5 @@ class TestExpandOpInt64_t(OpTest):
         self.check_output()
 
 
-class TestExpandError(unittest.TestCase):
-    def test_errors(self):
-        with program_guard(Program(), Program()):
-            x1 = fluid.create_lod_tensor(
-                np.array([[-1]]), [[1]], fluid.CPUPlace()
-            )
-            expand_times = [2, 2]
-            self.assertRaises(TypeError, fluid.layers.expand, x1, expand_times)
-            x2 = fluid.layers.data(name='x2', shape=[4], dtype="uint8")
-            self.assertRaises(TypeError, fluid.layers.expand, x2, expand_times)
-            x3 = fluid.layers.data(name='x3', shape=[4], dtype="bool")
-            x3.stop_gradient = True
-            self.assertRaises(ValueError, fluid.layers.expand, x3, expand_times)
-
-
-# Test python API
-class TestExpandAPI(unittest.TestCase):
-    def test_api(self):
-        input = np.random.random([12, 14]).astype("float32")
-        x = fluid.layers.data(
-            name='x', shape=[12, 14], append_batch_size=False, dtype="float32"
-        )
-
-        positive_2 = fluid.layers.fill_constant([1], "int32", 2)
-        expand_times = fluid.layers.data(
-            name="expand_times", shape=[2], append_batch_size=False
-        )
-
-        out_1 = fluid.layers.expand(x, expand_times=[2, 3])
-        out_2 = fluid.layers.expand(x, expand_times=[positive_2, 3])
-        out_3 = fluid.layers.expand(x, expand_times=expand_times)
-
-        g0 = fluid.backward.calc_gradient(out_2, x)
-
-        exe = fluid.Executor(place=fluid.CPUPlace())
-        res_1, res_2, res_3 = exe.run(
-            fluid.default_main_program(),
-            feed={"x": input, "expand_times": np.array([1, 3]).astype("int32")},
-            fetch_list=[out_1, out_2, out_3],
-        )
-        assert np.array_equal(res_1, np.tile(input, (2, 3)))
-        assert np.array_equal(res_2, np.tile(input, (2, 3)))
-        assert np.array_equal(res_3, np.tile(input, (1, 3)))
-
-
-class TestExpandDygraphAPI(unittest.TestCase):
-    def test_expand_times_is_tensor(self):
-        with paddle.fluid.dygraph.guard():
-            a = paddle.rand([2, 5])
-            b = paddle.fluid.layers.expand(a, expand_times=[2, 3])
-            c = paddle.fluid.layers.expand(
-                a, expand_times=paddle.to_tensor([2, 3], dtype='int32')
-            )
-            np.testing.assert_array_equal(b.numpy(), np.tile(a.numpy(), [2, 3]))
-            np.testing.assert_array_equal(c.numpy(), np.tile(a.numpy(), [2, 3]))
-
-
 if __name__ == "__main__":
     unittest.main()
diff --git a/python/paddle/fluid/tests/unittests/test_imperative_basic.py b/python/paddle/fluid/tests/unittests/test_imperative_basic.py
index 076c38773d5832dc848da90e552676465fe24f9b..5f543c04842da21c9959caf72c91f5178b0a3413 100644
--- a/python/paddle/fluid/tests/unittests/test_imperative_basic.py
+++ b/python/paddle/fluid/tests/unittests/test_imperative_basic.py
@@ -738,13 +738,13 @@ class TestImperative(unittest.TestCase):
                 name='inp2', shape=[3, 3], dtype=np.float32
             )
 
-            a = fluid.layers.expand(
+            a = paddle.expand(
                 paddle.reshape(fluid.layers.reduce_sum(inp_data1), [1, 1]),
-                [4, 1],
+                [4, -1],
             )
-            b = fluid.layers.expand(
+            b = paddle.expand(
                 paddle.reshape(fluid.layers.reduce_sum(inp_data2), [1, 1]),
-                [4, 1],
+                [4, -1],
             )
             cond = fluid.layers.less_than(x=a, y=b)
 
diff --git a/python/paddle/fluid/tests/unittests/test_imperative_ocr_attention_model.py b/python/paddle/fluid/tests/unittests/test_imperative_ocr_attention_model.py
index 0eb20a7dca158bcebb0a2563349980292fc01a3f..5ab7ef78a3622252e3ca4dfe635fd9372a4ac73a 100644
--- a/python/paddle/fluid/tests/unittests/test_imperative_ocr_attention_model.py
+++ b/python/paddle/fluid/tests/unittests/test_imperative_ocr_attention_model.py
@@ -306,8 +306,9 @@ class SimpleAttention(fluid.dygraph.Layer):
         decoder_state_proj_reshape = paddle.reshape(
             decoder_state_fc, [-1, 1, decoder_state_fc.shape[1]]
         )
-        decoder_state_expand = fluid.layers.expand(
-            decoder_state_proj_reshape, [1, encoder_proj.shape[1], 1]
+        decoder_state_expand = paddle.expand(
+            decoder_state_proj_reshape,
+            [-1, encoder_proj.shape[1], -1],
         )
         concated = fluid.layers.elementwise_add(
             encoder_proj, decoder_state_expand
diff --git a/python/paddle/fluid/tests/unittests/test_imperative_star_gan_with_gradient_penalty.py b/python/paddle/fluid/tests/unittests/test_imperative_star_gan_with_gradient_penalty.py
index 2c8d408316b6d484dad7238558bf5afef3688111..d623a277cf00627d0b830104d3114212b3e7bb9b 100644
--- a/python/paddle/fluid/tests/unittests/test_imperative_star_gan_with_gradient_penalty.py
+++ b/python/paddle/fluid/tests/unittests/test_imperative_star_gan_with_gradient_penalty.py
@@ -310,9 +310,7 @@ class Generator(fluid.dygraph.Layer):
     def forward(self, input, label_trg):
         shape = input.shape
         label_trg_e = paddle.reshape(label_trg, [-1, label_trg.shape[1], 1, 1])
-        label_trg_e = fluid.layers.expand(
-            x=label_trg_e, expand_times=[1, 1, shape[2], shape[3]]
-        )
+        label_trg_e = paddle.expand(label_trg_e, [-1, -1, shape[2], shape[3]])
 
         input1 = fluid.layers.concat([input, label_trg_e], 1)
 
diff --git a/python/paddle/fluid/tests/unittests/test_layers.py b/python/paddle/fluid/tests/unittests/test_layers.py
index ad1226e12d4a3c4f6b15efde309917ef9680b507..f191a948c9d2c34eda251397b11e2d6454d66452 100644
--- a/python/paddle/fluid/tests/unittests/test_layers.py
+++ b/python/paddle/fluid/tests/unittests/test_layers.py
@@ -3554,14 +3554,6 @@ class TestBook(LayerTest):
             out = layers.cross_entropy(x, label, False, 4)
             return out
 
-    def make_expand(self):
-        with program_guard(
-            fluid.default_main_program(), fluid.default_startup_program()
-        ):
-            x = self._get_data(name="input", shape=[10], dtype='int32')
-            out = layers.expand(x, [1, 2])
-            return out
-
     def make_uniform_random_batch_size_like(self):
         with program_guard(
             fluid.default_main_program(), fluid.default_startup_program()
diff --git a/python/paddle/fluid/tests/unittests/test_nn_grad.py b/python/paddle/fluid/tests/unittests/test_nn_grad.py
index 274aa25142e40a6073abd90d076fb4bbf63599a7..c7473895b08fb27091096648274fd1e20334ec97 100644
--- a/python/paddle/fluid/tests/unittests/test_nn_grad.py
+++ b/python/paddle/fluid/tests/unittests/test_nn_grad.py
@@ -114,31 +114,6 @@ class TestReduceSumWithDimDoubleGradCheck(unittest.TestCase):
 
 
 class TestReshapeDoubleGradCheck(unittest.TestCase):
-    @prog_scope()
-    def func(self, place):
-        x_shape = [3, 12]
-        expand_times = [4, 9]
-        eps = 0.005
-        dtype = np.float64
-
-        x = layers.data('x', x_shape, False, dtype)
-        x.persistable = True
-        out = layers.expand(x, expand_times)
-        x_arr = np.random.uniform(-1, 1, x_shape).astype(dtype)
-
-        gradient_checker.double_grad_check(
-            [x], out, x_init=x_arr, place=place, eps=eps
-        )
-
-    def test_grad(self):
-        places = [fluid.CPUPlace()]
-        if core.is_compiled_with_cuda():
-            places.append(fluid.CUDAPlace(0))
-        for p in places:
-            self.func(p)
-
-
-class TestExpandDoubleGradCheck(unittest.TestCase):
     @prog_scope()
     def func(self, place):
         x_shape = [3, 12]