!7077 code format for nn.layer

Merge pull request !7077 from chenzhongming/zomi_master

mindspore/nn/layer/activation.py

@@ -16,13 +16,13 @@
 import numpy as np
 from mindspore.ops import operations as P
 from mindspore.ops import functional as F
+from mindspore.ops import _selected_ops
 from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer
 from mindspore.common.tensor import Tensor
 from mindspore._extends import cell_attr_register
-from mindspore.ops import _selected_ops
+from mindspore._checkparam import Validator as validator
 from ..cell import Cell
-from ..._checkparam import Validator as validator
 
 
 __all__ = ['Softmax',

mindspore/nn/layer/basic.py

@@ -20,7 +20,6 @@ import mindspore.common.dtype as mstype
 from mindspore.common.seed import get_seed
 from mindspore.common.tensor import Tensor
 from mindspore.common.initializer import initializer
-from mindspore._checkparam import check_int_positive, check_bool
 from mindspore.ops import operations as P
 from mindspore.ops import functional as F
 from mindspore.ops.functional import identity
@@ -28,12 +27,12 @@ from mindspore.ops.operations import _inner_ops as inner
 from mindspore.ops.primitive import constexpr
 from mindspore.common.parameter import Parameter
 from mindspore._extends import cell_attr_register
+from mindspore._checkparam import Rel, Validator as validator, check_int_positive, check_bool
 from mindspore.common.api import ms_function
 from mindspore import context
 from ..cell import Cell
 from .activation import get_activation
-from ..._checkparam import Validator as validator
-from ..._checkparam import Rel
+
 
 __all__ = ['Dropout', 'Flatten', 'Dense', 'ClipByNorm', 'Norm', 'OneHot', 'Pad', 'Unfold',
            'MatrixDiag', 'MatrixDiagPart', 'MatrixSetDiag']

mindspore/nn/layer/conv.py

@@ -13,7 +13,6 @@
 # limitations under the License.
 # ============================================================================
 """conv"""
-
 import numpy as np
 from mindspore import log as logger
 from mindspore import context
@@ -23,8 +22,7 @@ from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer, Initializer
 from mindspore.common.tensor import Tensor
 from mindspore._checkparam import ParamValidator as validator, Rel
-from mindspore._checkparam import Validator
-from mindspore._checkparam import check_bool, twice, check_int_positive
+from mindspore._checkparam import check_bool, twice, check_int_positive, Validator
 from mindspore._extends import cell_attr_register
 from ..cell import Cell
 

mindspore/nn/layer/embedding.py

@@ -18,12 +18,11 @@ from mindspore.common.tensor import Tensor
 from mindspore.ops import operations as P
 from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer
-from mindspore._checkparam import Validator
 from mindspore.communication.management import get_group_size
 from mindspore.context import ParallelMode
 from mindspore.parallel._utils import _get_parallel_mode
+from mindspore._checkparam import Rel, Validator as validator
 from ..cell import Cell
-from ..._checkparam import Validator as validator, Rel
 
 __all__ = ['Embedding', 'EmbeddingLookup']
 
@@ -171,7 +170,7 @@ class EmbeddingLookup(Cell):
             if not isinstance(manual_shapes, tuple):
                 raise TypeError("manual_shapes type must be tuple(int) cannot be {}!".format(type(manual_shapes)))
             for dim in manual_shapes:
-                Validator.check_integer('manul shape dim', dim, 0, Rel.GT, self.cls_name)
+                validator.check_integer('manul shape dim', dim, 0, Rel.GT, self.cls_name)
             self.gatherv2.add_prim_attr("manual_split", manual_shapes)
             self.embeddinglookup.add_prim_attr("manual_split", manual_shapes)
             self.gatherv2.shard(((get_group_size(), 1), (1, get_group_size())))

mindspore/nn/layer/image.py

@@ -20,8 +20,7 @@ from mindspore.common.tensor import Tensor
 from mindspore.ops import operations as P
 from mindspore.ops import functional as F
 from mindspore.ops.primitive import constexpr
-from mindspore._checkparam import Validator as validator
-from mindspore._checkparam import Rel
+from mindspore._checkparam import Rel, Validator as validator
 from .conv import Conv2d
 from .container import CellList
 from .pooling import AvgPool2d

mindspore/nn/layer/lstm.py

@@ -14,16 +14,14 @@
 # ============================================================================
 """lstm"""
 import math
-
 import numpy as np
-
-from mindspore._checkparam import Validator as validator
+from mindspore._checkparam import Rel, Validator as validator
 from mindspore.common.initializer import initializer
 from mindspore.common.parameter import Parameter
 from mindspore.common.tensor import Tensor
 from mindspore.nn.cell import Cell
 from mindspore.ops import operations as P
-from ..._checkparam import Rel
+
 
 __all__ = ['LSTM', 'LSTMCell']
 
@@ -32,7 +30,7 @@ class LSTM(Cell):
     r"""
     LSTM (Long Short-Term Memory) layer.
 
-    Applies a LSTM to the input.
+    Apply LSTM layer to the input.
 
     There are two pipelines connecting two consecutive cells in a LSTM model; one is cell state pipeline
     and the other is hidden state pipeline. Denote two consecutive time nodes as :math:`t-1` and :math:`t`.
@@ -88,25 +86,11 @@ class LSTM(Cell):
           (num_directions * `num_layers`, batch_size, `hidden_size`).
 
     Examples:
-        >>> class LstmNet(nn.Cell):
-        >>>     def __init__(self, input_size, hidden_size, num_layers, has_bias, batch_first, bidirectional):
-        >>>         super(LstmNet, self).__init__()
-        >>>         self.lstm = nn.LSTM(input_size=input_size,
-        >>>                             hidden_size=hidden_size,
-        >>>                             num_layers=num_layers,
-        >>>                             has_bias=has_bias,
-        >>>                             batch_first=batch_first,
-        >>>                             bidirectional=bidirectional,
-        >>>                             dropout=0.0)
-        >>>
-        >>>     def construct(self, inp, h0, c0):
-        >>>         return self.lstm(inp, (h0, c0))
-        >>>
-        >>> net = LstmNet(10, 12, 2, has_bias=True, batch_first=True, bidirectional=False)
+        >>> net = nn.LSTM(10, 12, 2, has_bias=True, batch_first=True, bidirectional=False)
         >>> input = Tensor(np.ones([3, 5, 10]).astype(np.float32))
         >>> h0 = Tensor(np.ones([1 * 2, 3, 12]).astype(np.float32))
         >>> c0 = Tensor(np.ones([1 * 2, 3, 12]).astype(np.float32))
-        >>> output, (hn, cn) = net(input, h0, c0)
+        >>> output, (hn, cn) = net(input, (h0, c0))
     """
 
     def __init__(self,
@@ -159,7 +143,7 @@ class LSTMCell(Cell):
     r"""
     LSTM (Long Short-Term Memory) layer.
 
-    Applies a LSTM layer to the input.
+    Apply LSTM layer to the input.
 
     There are two pipelines connecting two consecutive cells in a LSTM model; one is cell state pipeline
     and the other is hidden state pipeline. Denote two consecutive time nodes as :math:`t-1` and :math:`t`.
@@ -224,20 +208,7 @@ class LSTMCell(Cell):
         - **state** - reserved
 
     Examples:
-        >>> class LstmNet(nn.Cell):
-        >>>     def __init__(self, input_size, hidden_size, has_bias, batch_first, bidirectional):
-        >>>         super(LstmNet, self).__init__()
-        >>>         self.lstm = nn.LSTMCell(input_size=input_size,
-        >>>                             hidden_size=hidden_size,
-        >>>                             has_bias=has_bias,
-        >>>                             batch_first=batch_first,
-        >>>                             bidirectional=bidirectional,
-        >>>                             dropout=0.0)
-        >>>
-        >>>     def construct(self, inp, h, c, w):
-        >>>         return self.lstm(inp, h, c, w)
-        >>>
-        >>> net = LstmNet(10, 12, has_bias=True, batch_first=True, bidirectional=False)
+        >>> net = nn.LSTMCell(10, 12, has_bias=True, batch_first=True, bidirectional=False)
         >>> input = Tensor(np.ones([3, 5, 10]).astype(np.float32))
         >>> h = Tensor(np.ones([1, 3, 12]).astype(np.float32))
         >>> c = Tensor(np.ones([1, 3, 12]).astype(np.float32))

mindspore/nn/layer/math.py

@@ -21,8 +21,7 @@ from mindspore.common.tensor import Tensor
 from mindspore.ops.primitive import constexpr
 from ..cell import Cell
 from ...common import dtype as mstype
-from ..._checkparam import Validator as validator
-from ..._checkparam import Rel
+from ..._checkparam import Rel, Validator as validator
 
 
 __all__ = ['ReduceLogSumExp', 'Range', 'LinSpace', 'LGamma', 'MatMul']

mindspore/nn/layer/normalization.py

@@ -19,11 +19,10 @@ from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer
 from mindspore.ops.primitive import constexpr
 import mindspore.context as context
-from mindspore._checkparam import check_bool, check_typename
+from mindspore._checkparam import check_bool, check_typename, check_int_positive
 from mindspore._extends import cell_attr_register
 from mindspore.communication.management import get_group_size, get_rank
 from mindspore.communication import management
-from mindspore._checkparam import check_int_positive
 from mindspore.ops import _selected_ops
 from ..cell import Cell
 

mindspore/nn/layer/pooling.py

@@ -15,11 +15,10 @@
 """pooling"""
 from mindspore.ops import operations as P
 from mindspore.ops import functional as F
-from mindspore._checkparam import Validator as validator
+from mindspore._checkparam import Rel, Validator as validator
 from mindspore.ops.primitive import constexpr
-from ... import context
+import mindspore.context as context
 from ..cell import Cell
-from ..._checkparam import Rel
 
 __all__ = ['AvgPool2d', 'MaxPool2d', 'AvgPool1d']
 

mindspore/nn/layer/quant.py

@@ -16,7 +16,6 @@
 
 from functools import partial
 import numpy as np
-
 from mindspore import nn
 import mindspore.common.dtype as mstype
 from mindspore.ops import operations as P
@@ -24,15 +23,11 @@ from mindspore.ops import functional as F
 from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer
 from mindspore.common.tensor import Tensor
-from mindspore._checkparam import check_int_positive, check_bool, twice
-from mindspore._checkparam import Rel
+from mindspore._checkparam import Rel, check_int_positive, check_bool, twice, ParamValidator as validator
 import mindspore.context as context
-
 from .normalization import BatchNorm2d, BatchNorm1d
 from .activation import get_activation, ReLU, LeakyReLU
 from ..cell import Cell
-from . import conv, basic
-from ..._checkparam import ParamValidator as validator
 from ...ops.operations import _quant_ops as Q
 
 __all__ = [
@@ -127,17 +122,17 @@ class Conv2dBnAct(Cell):
                  after_fake=True):
         super(Conv2dBnAct, self).__init__()
 
-        self.conv = conv.Conv2d(in_channels,
-                                out_channels,
-                                kernel_size=kernel_size,
-                                stride=stride,
-                                pad_mode=pad_mode,
-                                padding=padding,
-                                dilation=dilation,
-                                group=group,
-                                has_bias=has_bias,
-                                weight_init=weight_init,
-                                bias_init=bias_init)
+        self.conv = nn.Conv2d(in_channels,
+                              out_channels,
+                              kernel_size=kernel_size,
+                              stride=stride,
+                              pad_mode=pad_mode,
+                              padding=padding,
+                              dilation=dilation,
+                              group=group,
+                              has_bias=has_bias,
+                              weight_init=weight_init,
+                              bias_init=bias_init)
         self.has_bn = validator.check_bool("has_bn", has_bn)
         self.has_act = activation is not None
         self.after_fake = after_fake
@@ -200,7 +195,7 @@ class DenseBnAct(Cell):
                  activation=None,
                  after_fake=True):
         super(DenseBnAct, self).__init__()
-        self.dense = basic.Dense(
+        self.dense = nn.Dense(
             in_channels,
             out_channels,
             weight_init,
@@ -1349,11 +1344,6 @@ class QuantBlock(Cell):
 
     Outputs:
         Tensor of shape :math:`(N, out\_channels)`.
-
-    Examples:
-        >>> net = nn.Dense(3, 4)
-        >>> input = Tensor(np.random.randint(0, 255, [2, 3]), mindspore.float32)
-        >>> net(input)
     """
 
     def __init__(self,