[CodeStyle][Xdoctest][6,13,81,84,142,144,146] Fix example code(paddle.Tensor.(angle,flatten),paddle.incubate.nn.(FusedFeedForward,FusedTransformerEncoderLayer),paddle.nn.(Transformer,TransformerDecoderLayer,TransformerEncoderLayer))

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

@@ -577,7 +577,7 @@ class FusedFeedForward(Layer):
  this property. For more information, please refer to :ref:`api_guide_Name`.
 
  Examples:
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> # doctest: +REQUIRES(env:GPU)
  >>> import paddle
@@ -588,7 +588,7 @@ class FusedFeedForward(Layer):
  >>> x = paddle.rand((1, 8, 8))
  >>> out = fused_feedforward_layer(x)
  >>> print(out.shape)
- [1, 8, 8]
+ paddle.Size([1, 8, 8])
  """
 
  name: str | None
@@ -789,7 +789,7 @@ class FusedTransformerEncoderLayer(Layer):
 
 
  Examples:
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> # doctest: +REQUIRES(env:GPU)
  >>> import paddle
@@ -803,7 +803,7 @@ class FusedTransformerEncoderLayer(Layer):
  >>> encoder_layer = FusedTransformerEncoderLayer(128, 2, 512)
  >>> enc_output = encoder_layer(enc_input, attn_mask)
  >>> print(enc_output.shape)
- [2, 4, 128]
+ paddle.Size([2, 4, 128])
 
  """
 

python/paddle/jit/dy2static/transformers/base.py

@@ -102,7 +102,8 @@ class ForLoopTuplePreTransformer(BaseTransformer):
 
  will be changed into :
 
- >>> UUID_iterator = _jst.Indexable(B) # make iterator-only to indexable list.
+ >>> # make iterator-only to indexable list.
+ >>> UUID_iterator = _jst.Indexable(B)
  >>> for UUID_target in UUID_iterator:
  >>> A = _jst.Unpack(UUID_target, structure)
  >>> C

python/paddle/jit/dy2static/utils.py

@@ -476,16 +476,18 @@ def wrap_as_closure(tree: gast.AST, closure_vars: list[str]) -> gast.AST:
 
  Before:
 
- >>> def fn(x):
- ... ...
+ >>> def fn(x): ...
 
  After:
 
  >>> def create_fn():
  ... closure_var_1 = None
- ... def fn(x):
- ... ...
+ ...
+ ... def fn(x): ...
+ ...
  ... return fn
+ ...
+ ...
  ... fn = create_fn()
  """
 

python/paddle/jit/sot/opcode_translator/executor/dispatcher.py

@@ -200,9 +200,7 @@ class Dispatcher:
 
  Examples:
 
- >>> def builtin_add(a: int, b: int) -> int:
- ... ...
- ...
+ >>> def builtin_add(a: int, b: int) -> int: ...
  >>> Dispatcher.register(builtin_add, ("int", "int"), lambda a, b: a + b)
  >>> handler = Dispatcher.dispatch(builtin_add, 1, 2)
  >>> handler(1, 2)
@@ -250,13 +248,10 @@ def register_decorator(cls, fn: Callable[..., Any]):
  fn: The function to be registered.
 
  Examples:
- >>> def builtin_add(a: int, b: int) -> int:
- ... ...
- ...
+ >>> def builtin_add(a: int, b: int) -> int: ...
  >>> @Dispatcher.register_decorator(builtin_add)
  ... def builtin_add_dispatcher(a: int, b: int) -> int:
  ... return a + b
- ...
  >>> handler = Dispatcher.dispatch(builtin_add, 1, 2)
  >>> handler(1, 2)
  3

python/paddle/jit/sot/opcode_translator/executor/tracker.py

@@ -158,7 +158,10 @@ class DanglingTracker(Tracker):
 
  Examples:
  >>> import operator
- >>> from sot.opcode_translator.executor.variables import BuiltinVariable, ConstantVariable
+ >>> from sot.opcode_translator.executor.variables import (
+ ... BuiltinVariable,
+ ... ConstantVariable,
+ ... )
  >>> a = ConstantVariable.wrap_literal(1, None)
  >>> b = ConstantVariable.wrap_literal(2, None)
  >>> c = BuiltinVariable(operator.add, None, DanglingTracker())(a, b)

python/paddle/nn/layer/transformer.py

@@ -608,7 +608,7 @@ class TransformerEncoderLayer(Layer):
 
  Examples:
 
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> import paddle
  >>> from paddle.nn import TransformerEncoderLayer
@@ -620,7 +620,7 @@ class TransformerEncoderLayer(Layer):
  >>> encoder_layer = TransformerEncoderLayer(128, 2, 512)
  >>> enc_output = encoder_layer(enc_input, attn_mask)
  >>> print(enc_output.shape)
- [2, 4, 128]
+ paddle.Size([2, 4, 128])
  """
 
  activation: Layer
@@ -972,7 +972,7 @@ class TransformerDecoderLayer(Layer):
 
  Examples:
 
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> import paddle
  >>> from paddle.nn import TransformerDecoderLayer
@@ -986,12 +986,11 @@ class TransformerDecoderLayer(Layer):
  >>> # cross attention mask: [batch_size, n_head, tgt_len, src_len]
  >>> cross_attn_mask = paddle.rand((2, 2, 4, 6))
  >>> decoder_layer = TransformerDecoderLayer(128, 2, 512)
- >>> output = decoder_layer(dec_input,
- ... enc_output,
- ... self_attn_mask,
- ... cross_attn_mask)
+ >>> output = decoder_layer(
+ ... dec_input, enc_output, self_attn_mask, cross_attn_mask
+ ... )
  >>> print(output.shape)
- [2, 4, 128]
+ paddle.Size([2, 4, 128])
  """
 
  normalize_before: bool
@@ -1498,7 +1497,7 @@ class Transformer(Layer):
 
  Examples:
 
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> import paddle
  >>> from paddle.nn import Transformer
@@ -1514,13 +1513,15 @@ class Transformer(Layer):
  >>> # memory_mask: [batch_size, n_head, tgt_len, src_len]
  >>> cross_attn_mask = paddle.rand((2, 2, 6, 4))
  >>> transformer = Transformer(128, 2, 4, 4, 512)
- >>> output = transformer(enc_input,
- ... dec_input,
- ... enc_self_attn_mask,
- ... dec_self_attn_mask,
- ... cross_attn_mask)
+ >>> output = transformer(
+ ... enc_input,
+ ... dec_input,
+ ... enc_self_attn_mask,
+ ... dec_self_attn_mask,
+ ... cross_attn_mask,
+ ... )
  >>> print(output.shape)
- [2, 6, 128]
+ paddle.Size([2, 6, 128])
  """
 
  encoder: Layer

python/paddle/tensor/manipulation.py

@@ -2035,18 +2035,23 @@ def flatten(
 
  Examples:
 
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> import paddle
 
- >>> image_shape=(2, 3, 4, 4)
+ >>> image_shape = (2, 3, 4, 4)
 
- >>> x = paddle.arange(end=image_shape[0] * image_shape[1] * image_shape[2] * image_shape[3])
+ >>> x = paddle.arange(
+ ... end=image_shape[0]
+ ... * image_shape[1]
+ ... * image_shape[2]
+ ... * image_shape[3]
+ ... )
  >>> img = paddle.reshape(x, image_shape)
 
  >>> out = paddle.flatten(img, start_axis=1, stop_axis=2)
  >>> print(out.shape)
- [2, 12, 4]
+ paddle.Size([2, 12, 4])
 
  >>> # out shares data with img in dygraph mode
  >>> img[0, 0, 0, 0] = -1

python/paddle/tensor/math.py

@@ -5880,19 +5880,27 @@ def angle(x: Tensor, name: str | None = None) -> Tensor:
  Tensor: An N-D Tensor of real data type with the same precision as that of x's data type.
 
  Examples:
- .. code-block:: python
+ .. code-block:: pycon
 
  >>> import paddle
 
- >>> x = paddle.to_tensor([-2, -1, 0, 1]).unsqueeze(-1).astype('float32')
+ >>> x = (
+ ... paddle.to_tensor([-2, -1, 0, 1])
+ ... .unsqueeze(-1)
+ ... .astype('float32')
+ ... )
  >>> y = paddle.to_tensor([-2, -1, 0, 1]).astype('float32')
  >>> z = x + 1j * y
  >>> z
  Tensor(shape=[4, 4], dtype=complex64, place=Place(cpu), stop_gradient=True,
- [[(-2-2j), (-2-1j), (-2+0j), (-2+1j)],
- [(-1-2j), (-1-1j), (-1+0j), (-1+1j)],
- [-2j , -1j , 0j , 1j ],
- [ (1-2j), (1-1j), (1+0j), (1+1j)]])
+ [[(-2.00000000-2.00000000j), (-2.00000000-1.00000000j),
+ (-2.00000000+0.00000000j), (-2.00000000+1.00000000j)],
+ [(-1.00000000-2.00000000j), (-1.00000000-1.00000000j),
+ (-1.00000000+0.00000000j), (-1.00000000+1.00000000j)],
+ [(0.00000000-2.00000000j) , (0.00000000-1.00000000j) ,
+ (0.00000000+0.00000000j), (0.00000000+1.00000000j)],
+ [ (1.00000000-2.00000000j), (1.00000000-1.00000000j),
+ (1.00000000+0.00000000j), (1.00000000+1.00000000j)]])
 
  >>> theta = paddle.angle(z)
  >>> theta