[cherry-pick2.4]fix numpy issue in codeblock examples (#47664)

* #46765

* #47042

* Remove redundant numpy import (#47483)

* #47555

* resolve conflict

* resolve conflict

* resolve conflict

* resolve conflict

* resolve conflict

* for_codestyle

* fix sample code paddle.linalg.multi_dot

Co-authored-by: Kevin吴嘉文 <[email protected]>

Author: Ligoml

python/paddle/distributed/fleet/fleet.py

@@ -1120,7 +1120,6 @@ def amp_init(
         Examples:
             .. code-block:: python
 
-                import numpy as np
                 import paddle
                 import paddle.nn.functional as F
                 paddle.enable_static()

python/paddle/incubate/nn/functional/fused_transformer.py

@@ -945,7 +945,6 @@ def fused_multi_transformer(
             # required: gpu
             import paddle
             import paddle.incubate.nn.functional as F
-            import numpy as np
 
             # input: [batch_size, seq_len, embed_dim]
             x = paddle.rand(shape=(2, 4, 128), dtype="float32")

python/paddle/incubate/optimizer/lookahead.py

@@ -166,8 +166,7 @@ def step(self):
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+                inp = paddle.rand([1,10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)
@@ -280,8 +279,8 @@ def minimize(
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+
+                inp = paddle.rand([1, 10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)

python/paddle/incubate/optimizer/modelaverage.py

@@ -342,8 +342,7 @@ def minimize(
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+                inp = paddle.rand([1, 10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)
@@ -378,8 +377,7 @@ def step(self):
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+                inp = paddle.rand([1, 10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)
@@ -425,8 +423,7 @@ def apply(self, executor=None, need_restore=True):
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+                inp = paddle.rand([1, 10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)
@@ -500,8 +497,7 @@ def restore(self, executor=None):
             .. code-block:: python
 
                 import paddle
-                import numpy as np
-                inp = paddle.to_tensor(np.random.random([1, 10]).astype('float32'))
+                inp = paddle.rand([1, 10], dtype="float32")
                 linear = paddle.nn.Linear(10, 1)
                 out = linear(inp)
                 loss = paddle.mean(out)

python/paddle/nn/functional/activation.py

@@ -1313,10 +1313,12 @@ def softshrink(x, threshold=0.5, name=None):
 
             import paddle
             import paddle.nn.functional as F
-            import numpy as np
 
-            x = paddle.to_tensor(np.array([-0.9, -0.2, 0.1, 0.8]))
-            out = F.softshrink(x) # [-0.4, 0, 0, 0.3]
+            x = paddle.to_tensor([-0.9, -0.2, 0.1, 0.8])
+            out = F.softshrink(x)
+            print(out)
+            # Tensor(shape=[4], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [-0.39999998,  0.        ,  0.        ,  0.30000001])
     """
     if threshold < 0:
         raise ValueError(
@@ -1365,10 +1367,12 @@ def softsign(x, name=None):
 
             import paddle
             import paddle.nn.functional as F
-            import numpy as np
 
-            x = paddle.to_tensor(np.array([-0.4, -0.2, 0.1, 0.3]))
-            out = F.softsign(x) # [-0.285714, -0.166667, 0.0909091, 0.230769]
+            x = paddle.to_tensor([-0.4, -0.2, 0.1, 0.3])
+            out = F.softsign(x)
+            print(out)
+            # Tensor(shape=[4], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [-0.28571430, -0.16666666,  0.09090909,  0.23076925])
     """
     if in_dygraph_mode():
         return _C_ops.softsign(x)
@@ -1405,10 +1409,12 @@ def swish(x, name=None):
 
             import paddle
             import paddle.nn.functional as F
-            import numpy as np
 
-            x = paddle.to_tensor(np.array([-2., 0., 1.]))
-            out = F.swish(x) # [-0.238406, 0., 0.731059]
+            x = paddle.to_tensor([-2., 0., 1.])
+            out = F.swish(x)
+            print(out)
+            # Tensor(shape=[3], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [-0.23840584,  0.        ,  0.73105854])
     """
     if in_dygraph_mode():
         return _C_ops.swish(x, 1.0)
@@ -1487,10 +1493,12 @@ def tanhshrink(x, name=None):
 
             import paddle
             import paddle.nn.functional as F
-            import numpy as np
 
-            x = paddle.to_tensor(np.array([-0.4, -0.2, 0.1, 0.3]))
-            out = F.tanhshrink(x) # [-0.020051, -0.00262468, 0.000332005, 0.00868739]
+            x = paddle.to_tensor([-0.4, -0.2, 0.1, 0.3])
+            out = F.tanhshrink(x)
+            print(out)
+            # Tensor(shape=[4], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [-0.02005106, -0.00262468,  0.00033200,  0.00868741])
     """
     if in_dygraph_mode():
         return _C_ops.tanh_shrink(x)
@@ -1536,10 +1544,12 @@ def thresholded_relu(x, threshold=1.0, name=None):
 
             import paddle
             import paddle.nn.functional as F
-            import numpy as np
 
-            x = paddle.to_tensor(np.array([2., 0., 1.]))
-            out = F.thresholded_relu(x) # [2., 0., 0.]
+            x = paddle.to_tensor([2., 0., 1.])
+            out = F.thresholded_relu(x)
+            print(out)
+            # Tensor(shape=[3], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [2., 0., 0.])
     """
 
     if in_dygraph_mode():

python/paddle/nn/functional/common.py

@@ -1965,18 +1965,16 @@ def label_smooth(label, prior_dist=None, epsilon=0.1, name=None):
         .. code-block:: python
 
             import paddle
-            import numpy as np
-
-            x_data = np.array([[[0, 1, 0],
-                                [ 1,  0, 1]]]).astype("float32")
-            print(x_data.shape)
             paddle.disable_static()
-            x = paddle.to_tensor(x_data, stop_gradient=False)
+
+            x = paddle.to_tensor([[[0, 1, 0],
+                                [ 1,  0, 1]]], dtype="float32", stop_gradient=False)
+
             output = paddle.nn.functional.label_smooth(x)
             print(output)
-
-            #[[[0.03333334 0.93333334 0.03333334]
-            #  [0.93333334 0.03333334 0.93333334]]]
+            # Tensor(shape=[1, 2, 3], dtype=float32, place=Place(gpu:0), stop_gradient=False,
+            #        [[[0.03333334, 0.93333334, 0.03333334],
+            #          [0.93333334, 0.03333334, 0.93333334]]])
     """
     if epsilon > 1.0 or epsilon < 0.0:
         raise ValueError("The value of epsilon must be between 0 and 1.")

python/paddle/nn/functional/conv.py

@@ -396,26 +396,22 @@ def conv1d(
 
           import paddle
           import paddle.nn.functional as F
-          import numpy as np
-          x = np.array([[[4, 8, 1, 9],
-            [7, 2, 0, 9],
-            [6, 9, 2, 6]]]).astype(np.float32)
-          w=np.array(
-          [[[9, 3, 4],
-            [0, 0, 7],
-            [2, 5, 6]],
-           [[0, 3, 4],
-            [2, 9, 7],
-            [5, 6, 8]]]).astype(np.float32)
-
-          x_var = paddle.to_tensor(x)
-          w_var = paddle.to_tensor(w)
-          y_var = F.conv1d(x_var, w_var)
-          y_np = y_var.numpy()
-          print(y_np)
 
-          # [[[133. 238.]
-          #   [160. 211.]]]
+          x = paddle.to_tensor([[[4, 8, 1, 9],
+                                 [7, 2, 0, 9],
+                                 [6, 9, 2, 6]]], dtype="float32")
+          w = paddle.to_tensor([[[9, 3, 4],
+                                 [0, 0, 7],
+                                 [2, 5, 6]],
+                                [[0, 3, 4],
+                                 [2, 9, 7],
+                                 [5, 6, 8]]], dtype="float32")
+
+          y = F.conv1d(x, w)
+          print(y)
+          # Tensor(shape=[1, 2, 2], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+          #        [[[133., 238.],
+          #          [160., 211.]]])
     """
     cudnn_version = get_cudnn_version()
     if cudnn_version is not None:
@@ -949,24 +945,20 @@ def conv1d_transpose(
     Examples:
         .. code-block:: python
 
-
-
           import paddle
           import paddle.nn.functional as F
-          import numpy as np
-          
+
           # shape: (1, 2, 4)
-          x=np.array([[[4, 0, 9, 7],
-                       [8, 0, 9, 2,]]]).astype(np.float32)
+          x = paddle.to_tensor([[[4, 0, 9, 7],
+                                [8, 0, 9, 2,]]], dtype="float32")
           # shape: (2, 1, 2)
-          w=np.array([[[7, 0]],
-                      [[4, 2]]]).astype(np.float32)
-          x_var = paddle.to_tensor(x)
-          w_var = paddle.to_tensor(w)
-          y_var = F.conv1d_transpose(x_var, w_var)
-          print(y_var)
-          
-          # [[[60. 16. 99. 75.  4.]]]
+          w = paddle.to_tensor([[[7, 0]],
+                                [[4, 2]]], dtype="float32")
+
+          y = F.conv1d_transpose(x, w)
+          print(y)
+          # Tensor(shape=[1, 1, 5], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+          #        [[[60., 16., 99., 75., 4. ]]])
     """
     cudnn_version = get_cudnn_version()
     if cudnn_version is not None:

python/paddle/nn/functional/loss.py

@@ -252,16 +252,14 @@ def fluid_softmax_with_cross_entropy(
         .. code-block:: python
 
             import paddle
-            import numpy as np
 
-            data = np.random.rand(128).astype("float32")
-            label = np.random.rand(1).astype("int64")
-            data = paddle.to_tensor(data)
-            label = paddle.to_tensor(label)
-            linear = paddle.nn.Linear(128, 100)
-            x = linear(data)
-            out = paddle.nn.functional.softmax_with_cross_entropy(logits=x, label=label)
+            logits = paddle.to_tensor([0.4, 0.6, 0.9])
+            label = paddle.randint(high=2, shape=[1], dtype="int64")
+
+            out = paddle.nn.functional.softmax_with_cross_entropy(logits=logits, label=label)
             print(out)
+            # Tensor(shape=[1], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [1.15328646])
     """
     if _non_static_mode():
         if core.is_compiled_with_npu():
@@ -1778,7 +1776,6 @@ def ctc_loss(
 
             # declarative mode
             import paddle.nn.functional as F
-            import numpy as np
             import paddle
 
             # length of the longest logit sequence
@@ -1790,8 +1787,7 @@ def ctc_loss(
             # class num
             class_num = 3
 
-            np.random.seed(1)
-            log_probs = np.array([[[4.17021990e-01, 7.20324516e-01, 1.14374816e-04],
+            log_probs = paddle.to_tensor([[[4.17021990e-01, 7.20324516e-01, 1.14374816e-04],
                                     [3.02332580e-01, 1.46755889e-01, 9.23385918e-02]],
 
                                     [[1.86260208e-01, 3.45560730e-01, 3.96767467e-01],
@@ -1804,30 +1800,30 @@ def ctc_loss(
                                     [9.68261600e-01, 3.13424170e-01, 6.92322612e-01]],
 
                                     [[8.76389146e-01, 8.94606650e-01, 8.50442126e-02],
-                                    [3.90547849e-02, 1.69830427e-01, 8.78142476e-01]]]).astype("float32")
-            labels = np.array([[1, 2, 2],
-                            [1, 2, 2]]).astype("int32")
-            input_lengths = np.array([5, 5]).astype("int64")
-            label_lengths = np.array([3, 3]).astype("int64")
-
-            log_probs = paddle.to_tensor(log_probs)
-            labels = paddle.to_tensor(labels)
-            input_lengths = paddle.to_tensor(input_lengths)
-            label_lengths = paddle.to_tensor(label_lengths)
+                                    [3.90547849e-02, 1.69830427e-01, 8.78142476e-01]]],
+                                    dtype="float32")
+            labels = paddle.to_tensor([[1, 2, 2],
+                                    [1, 2, 2]], dtype="int32")
+            input_lengths = paddle.to_tensor([5, 5], dtype="int64")
+            label_lengths = paddle.to_tensor([3, 3], dtype="int64")
 
             loss = F.ctc_loss(log_probs, labels,
                 input_lengths,
                 label_lengths,
                 blank=0,
                 reduction='none')
-            print(loss)  #[3.9179852 2.9076521]
+            print(loss)
+            # Tensor(shape=[2], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [3.91798496, 2.90765190])
 
             loss = F.ctc_loss(log_probs, labels,
                 input_lengths,
                 label_lengths,
                 blank=0,
                 reduction='mean')
-            print(loss)  #[1.1376063]
+            print(loss)
+            # Tensor(shape=[1], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [1.13760614])
 
     """
 
@@ -2265,16 +2261,14 @@ def softmax_with_cross_entropy(
         .. code-block:: python
 
             import paddle
-            import numpy as np
 
-            data = np.random.rand(128).astype("float32")
-            label = np.random.rand(1).astype("int64")
-            data = paddle.to_tensor(data)
-            label = paddle.to_tensor(label)
-            linear = paddle.nn.Linear(128, 100)
-            x = linear(data)
-            out = paddle.nn.functional.softmax_with_cross_entropy(logits=x, label=label)
+            logits = paddle.to_tensor([0.4, 0.6, 0.9], dtype="float32")
+            label = paddle.to_tensor([1], dtype="int64")
+
+            out = paddle.nn.functional.softmax_with_cross_entropy(logits=logits, label=label)
             print(out)
+            # Tensor(shape=[1], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [1.15328646])
     """
     return fluid_softmax_with_cross_entropy(
         logits,
@@ -3869,18 +3863,26 @@ def soft_margin_loss(input, label, reduction='mean', name=None):
         .. code-block:: python
 
             import paddle
-            import numpy as np
 
             input = paddle.to_tensor([[0.5, 0.6, 0.7],[0.3, 0.5, 0.2]], 'float32')
             label = paddle.to_tensor([[1.0, -1.0, 1.0],[-1.0, 1.0, 1.0]], 'float32')
             output = paddle.nn.functional.soft_margin_loss(input, label)
+            print(output)
+            # Tensor(shape=[1], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [0.64022040])
+
+            input = paddle.uniform(shape=(5, 5), dtype="float32", min=0.1, max=0.8)
+            label = paddle.randint(0, 2, shape=(5, 5), dtype="int64")
+            label[label==0]=-1
 
-            input_np = np.random.uniform(0.1, 0.8, size=(5, 5)).astype(np.float64)
-            label_np = np.random.randint(0, 2, size=(5, 5)).astype(np.int64)
-            label_np[label_np==0]=-1
-            input = paddle.to_tensor(input_np)
-            label = paddle.to_tensor(label_np)
             output = paddle.nn.functional.soft_margin_loss(input, label, reduction='none')
+            print(output)
+            # Tensor(shape=[5, 5], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+            #        [[1.09917796, 0.52613139, 0.56263304, 0.82736146, 0.38776723],
+            #         [1.07179427, 1.11924267, 0.49877715, 1.10026348, 0.46184641],
+            #         [0.84367639, 0.74795729, 0.44629076, 0.55123353, 0.77659678],
+            #         [0.39465919, 0.76651484, 0.54485321, 0.76609844, 0.77166790],
+            #         [0.51283568, 0.84757161, 0.78913331, 1.05268764, 0.45318675]])
     """
     if reduction not in ['sum', 'mean', 'none']:
         raise ValueError(