mindspore 2.1.0__cp38-none-any.whl → 2.2.0__cp38-none-any.whl

mindspore/nn/layer/conv.py

@@ -20,6 +20,7 @@ import numpy as np
 
 from mindspore import context
 from mindspore.ops import operations as P
+import mindspore.common.dtype as mstype
 from mindspore.ops.primitive import _primexpr
 from mindspore.common.parameter import Parameter
 from mindspore.common.initializer import initializer, HeUniform, Uniform, _calculate_fan_in_and_fan_out
@@ -50,7 +51,8 @@ class _Conv(Cell):
                  weight_init,
                  bias_init,
                  data_format='NCHW',
-                 transposed=False):
+                 transposed=False,
+                 dtype=mstype.float32):
         """Initialize _Conv."""
         super(_Conv, self).__init__()
         self.in_channels = Validator.check_positive_int(in_channels, 'in_channels', self.cls_name)
@@ -97,7 +99,7 @@ class _Conv(Cell):
         if weight_init is None:
             weight_init = HeUniform(math.sqrt(5))
         self.weight_init = weight_init
-        self.weight = Parameter(initializer(self.weight_init, shape), name='weight')
+        self.weight = Parameter(initializer(self.weight_init, shape, dtype=dtype), name='weight')
 
         self.bias_init = bias_init
         if Validator.check_bool(has_bias, "has_bias", self.cls_name):
@@ -109,7 +111,7 @@ class _Conv(Cell):
                 else:
                     bias_init = 'zeros'
                 self.bias_init = bias_init
-            self.bias = Parameter(initializer(self.bias_init, [out_channels]), name='bias')
+            self.bias = Parameter(initializer(self.bias_init, [out_channels], dtype=dtype), name='bias')
         else:
             self.bias = None
 
@@ -139,31 +141,44 @@ class _Conv(Cell):
 
 class Conv2d(_Conv):
     r"""
-    Calculates the 2D convolution on the input tensor.
-    The input is typically of shape :math:`(N, C_{in}, H_{in}, W_{in})`,
-    where :math:`N` is batch size, :math:`C_{in}` is a number of channels,
-    :math:`H_{in}, W_{in}` are the height and width of the feature layer respectively.
-    For the tensor of each batch, its shape is :math:`(C_{in}, H_{in}, W_{in})`, the formula is defined as:
+    2D convolution layer.
+
+    Applies a 2D convolution over an input tensor which is typically of shape :math:`(N, C_{in}, H_{in}, W_{in})`,
+    where :math:`N` is batch size, :math:`C` is channel number, :math:`H` is feature height, :math:`W` is feature width.
+
+    The output is calculated based on formula:
 
     .. math::
 
         \text{out}(N_i, C_{\text{out}_j}) = \text{bias}(C_{\text{out}_j}) +
         \sum_{k = 0}^{C_{in} - 1} \text{ccor}({\text{weight}(C_{\text{out}_j}, k), \text{X}(N_i, k)})
 
-    where :math:`ccor` is the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
-    :math:`C_{in}` is the channel number of the input, :math:`out_{j}` corresponds to the :math:`j`-th channel of
-    the output and :math:`j` is in the range of :math:`[0, C_{out}-1]`. :math:`\text{weight}(C_{\text{out}_j}, k)`
-    is a convolution kernel slice with shape :math:`(\text{kernel_size[0]}, \text{kernel_size[1]})`,
-    where :math:`\text{kernel_size[0]}` and :math:`\text{kernel_size[1]}` are the height and width of the convolution
-    kernel respectively. :math:`\text{bias}` is the bias parameter and :math:`\text{X}` is the input tensor.
-    In this case, `data_format` of the input tensor is 'NCHW' and the shape of full convolution kernel is
-    :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size[0]}, \text{kernel_size[1]})`,
-    where `group` is the number of groups to split the input `x` in the channel dimension. If `data_format` of the
-    input tensor is 'NHWC', the shape of full convolution kernel will be
-    :math:`(C_{out}, \text{kernel_size[0]}, \text{kernel_size[1]}), C_{in} / \text{group}`.
-
-    For more details, please refers to the paper `Gradient Based Learning Applied to Document
-    Recognition <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_.
+    where :math:`bias` is the output channel bias, :math:`ccor` is
+    the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
+    , :math:`weight` is the convolution kernel value and :math:`X` represents the input feature map.
+
+    Here are the indices' meanings:
+    - :math:`i` corresponds to the batch number, ranging from 0 to N-1, where N is the batch size of the input.
+
+    - :math:`j` corresponds to the output channel, ranging from 0 to C_{out}-1, where C_{out} is the number of
+      output channels, which is also equal to the number of kernels.
+
+    - :math:`k` corresponds to the input channel, ranging from 0 to C_{in}-1, where C_{in} is the number of
+      input channels, which is also equal to the number of channels in the convolutional kernels.
+
+    Therefore, in the above formula, :math:`{bias}(C_{out_j})` represents the bias of the :math:`j`-th
+    output channel, :math:`{weight}(C_{out_j}, k)` represents the slice of the :math:`j`-th convolutional
+    kernel in the :math:`k`-th channel, and :math:`{X}(N_i, k)` represents the slice of the :math:`k`-th input
+    channel in the :math:`i`-th batch of the input feature map.
+
+    The shape of the convolutional kernel is given by :math:`(kernel\_size[0], kernel\_size[1])`,
+    where :math:`kernel\_size[0]` and :math:`kernel\_size[1]` are the height and width of the kernel, respectively.
+    If we consider the input and output channels as well as the `group` parameter, the complete kernel shape
+    will be :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size[0]}, \text{kernel_size[1]})`,
+    where `group` is the number of groups dividing `x`'s input channel when applying group convolution.
+
+    For more details about convolution layer, please refer to `Gradient Based Learning Applied to Document Recognition
+    <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_.
 
     Note:
         On Ascend platform, only group convolution in depthwise convolution scenarios is supported.
@@ -176,48 +191,62 @@ class Conv2d(_Conv):
             The data type is an integer or a tuple of two integers. An integer represents the height
             and width of the convolution kernel. A tuple of two integers represents the height
             and width of the convolution kernel respectively.
-        stride (Union[int, tuple[int]]): The movement stride of the 2D convolution kernel.
+        stride (Union[int, tuple[int]], optional): The movement stride of the 2D convolution kernel.
             The data type is an integer or a tuple of two or four integers. An integer represents the movement step size
             in both height and width directions. A tuple of two integers represents the movement step size in the height
             and width directions respectively. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"`` .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
-
-        padding (Union[int, tuple[int]]): The number of padding on the height and width directions of the input.
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input around its edges so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally, If the amount is even, it is
+              uniformly distributed around the input, if it is odd, the excess amount goes to the right/bottom side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible height and width. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              in the height and width directions is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
+
+        padding (Union[int, tuple[int]], optional): The number of padding
+            on the height and width directions of the input.
             The data type is an integer or a tuple of four integers. If `padding` is an integer,
             then the top, bottom, left, and right padding are all equal to `padding`.
             If `padding` is a tuple of 4 integers, then the top, bottom, left, and right padding
             is equal to `padding[0]`, `padding[1]`, `padding[2]`, and `padding[3]` respectively.
             The value should be greater than or equal to 0. Default: ``0`` .
-        dilation (Union[int, tuple[int]]): Dilation size of 2D convolution kernel.
-            The data type is an integer or a tuple of two or four integers. If :math:`k > 1`, the kernel is sampled
-            every `k` elements. The value of `k` on the height and width directions is in range of [1, H]
-            and [1, W] respectively. Default: ``1`` .
-        group (int): Splits filter into groups, `in_channels` and `out_channels` must be
+        dilation (Union(int, tuple[int]), optional): Specifies the dilation rate to use for dilated convolution.
+            It can be a single int or a tuple of 2 or 4 integers. A single int means the dilation size is the same
+            in both the height and width directions. A tuple of two ints represents the dilation size in
+            the height and width directions, respectively. For a tuple of four ints, the two ints correspond
+            to (N, C) dimension are treated as 1, and the two correspond to (H, W) dimensions is the
+            dilation size in the height and width directions respectively.
+            Assuming :math:`dilation=(d0, d1)`, the convolutional kernel samples the input with a
+            spacing of :math:`d0-1` elements in the height direction and :math:`d1-1` elements in the width direction.
+            The values in the height and width dimensions are in the ranges [1, H] and [1, W], respectively.
+            Default: ``1`` .
+        group (int, optional): Splits filter into groups, `in_channels` and `out_channels` must be
             divisible by `group`. If the group is equal to `in_channels` and `out_channels`,
             this 2D convolution layer also can be called 2D depthwise convolution layer. Default: ``1`` .
-        has_bias (bool): Whether the Conv2d layer has a bias parameter. Default: ``False`` .
-        weight_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of weight parameter.
+        has_bias (bool, optional): Whether the Conv2d layer has a bias parameter. Default: ``False`` .
+        weight_init (Union[Tensor, str, Initializer, numbers.Number], optional): Initialization method of
+            weight parameter.
             It can be a Tensor, a string, an Initializer or a numbers.Number. When a string is specified,
             values from ``'TruncatedNormal'`` , ``'Normal'`` , ``'Uniform'`` , ``'HeUniform'`` and ``'XavierUniform'``
             distributions as well as constant ``'One'`` and ``'Zero'`` distributions are possible. Alias
             ``'xavier_uniform'`` , ``'he_uniform'`` , ``'ones'`` and ``'zeros'`` are acceptable. Uppercase and
-            lowercase are both acceptable. Refer to the values of Initializer for more details. Default: ``None`` ,
-            weight will be initialized using HeUniform.
-        bias_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of bias parameter.
+            lowercase are both acceptable. Refer to the values of
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , weight will be initialized using ``'HeUniform'``.
+        bias_init (Union[Tensor, str, Initializer, numbers.Number], optional): Initialization method of bias parameter.
             Available initialization methods are the same as 'weight_init'. Refer to the values of
-            Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
-        data_format (str): The optional value for data format, is ``'NHWC'`` or ``'NCHW'`` .
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , bias will be initialized using ``'Uniform'`` .
+        data_format (str, optional): The optional value for data format, is ``'NHWC'`` or ``'NCHW'`` .
             Default: ``'NCHW'`` .
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
 
     Inputs:
         - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, H_{in}, W_{in})` \
@@ -230,17 +259,17 @@ class Conv2d(_Conv):
 
         .. math::
             \begin{array}{ll} \\
-                H_{out} ＝ \left \lceil{\frac{H_{in}}{\text{stride[0]}}} \right \rceil \\
-                W_{out} ＝ \left \lceil{\frac{W_{in}}{\text{stride[1]}}} \right \rceil \\
+                H_{out} = \left \lceil{\frac{H_{in}}{\text{stride[0]}}} \right \rceil \\
+                W_{out} = \left \lceil{\frac{W_{in}}{\text{stride[1]}}} \right \rceil \\
             \end{array}
 
         pad_mode is ``'valid'``:
 
         .. math::
             \begin{array}{ll} \\
-                H_{out} ＝ \left \lceil{\frac{H_{in} - \text{dilation[0]} \times (\text{kernel_size[0]} - 1) }
+                H_{out} = \left \lceil{\frac{H_{in} - \text{dilation[0]} \times (\text{kernel_size[0]} - 1) }
                 {\text{stride[0]}}} \right \rceil \\
-                W_{out} ＝ \left \lceil{\frac{W_{in} - \text{dilation[1]} \times (\text{kernel_size[1]} - 1) }
+                W_{out} = \left \lceil{\frac{W_{in} - \text{dilation[1]} \times (\text{kernel_size[1]} - 1) }
                 {\text{stride[1]}}} \right \rceil \\
             \end{array}
 
@@ -248,9 +277,9 @@ class Conv2d(_Conv):
 
         .. math::
             \begin{array}{ll} \\
-                H_{out} ＝ \left \lfloor{\frac{H_{in} + padding[0] + padding[1] - (\text{kernel_size[0]} - 1) \times
+                H_{out} = \left \lfloor{\frac{H_{in} + padding[0] + padding[1] - (\text{kernel_size[0]} - 1) \times
                 \text{dilation[0]} - 1 }{\text{stride[0]}} + 1} \right \rfloor \\
-                W_{out} ＝ \left \lfloor{\frac{W_{in} + padding[2] + padding[3] - (\text{kernel_size[1]} - 1) \times
+                W_{out} = \left \lfloor{\frac{W_{in} + padding[2] + padding[3] - (\text{kernel_size[1]} - 1) \times
                 \text{dilation[1]} - 1 }{\text{stride[1]}} + 1} \right \rfloor \\
             \end{array}
 
@@ -291,7 +320,8 @@ class Conv2d(_Conv):
                  has_bias=False,
                  weight_init=None,
                  bias_init=None,
-                 data_format='NCHW'):
+                 data_format='NCHW',
+                 dtype=mstype.float32):
         """Initialize Conv2d."""
         kernel_size = twice(kernel_size)
         stride = twice(stride)
@@ -314,7 +344,8 @@ class Conv2d(_Conv):
             has_bias,
             weight_init,
             bias_init,
-            data_format)
+            data_format,
+            dtype=dtype)
         self.conv2d = P.Conv2D(out_channel=self.out_channels,
                                kernel_size=self.kernel_size,
                                mode=1,
@@ -341,26 +372,45 @@ def _check_input_3d(input_shape, op_name):
 
 class Conv1d(_Conv):
     r"""
-    Calculates the 1D convolution on the input tensor. The input is typically of shape :math:`(N, C_{in}, L_{in})`,
-    where :math:`N` is batch size, :math:`C_{in}` is a number of channels and :math:`L_{in}` is a length of sequence.
-    For the tensor of each batch, its shape is :math:`(C_{in}, L_{in})`, and the formula is defined as:
+    1D convolution layer.
+
+    Applies a 1D convolution over an input tensor which is typically of shape :math:`(N, C_{in}, L_{in})`,
+    where :math:`N` is batch size, :math:`C` is channel number, :math:`L` is input sequence width.
+
+    The output is calculated based on formula:
 
     .. math::
 
         \text{out}(N_i, C_{\text{out}_j}) = \text{bias}(C_{\text{out}_j}) +
         \sum_{k = 0}^{C_{in} - 1} \text{ccor}({\text{weight}(C_{\text{out}_j}, k), \text{X}(N_i, k)})
 
-    where :math:`ccor` is the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
-    :math:`C_{in}` is the channel number of the input, :math:`out_{j}` corresponds to the :math:`j`-th channel of
-    the output and :math:`j` is in the range of :math:`[0, C_{out}-1]`. :math:`\text{weight}(C_{\text{out}_j}, k)`
-    is a convolution kernel slice with shape :math:`\text{kernel_size}`, where :math:`\text{kernel_size}`
-    is the width of the convolution kernel. :math:`\text{bias}` is the bias parameter,
-    and :math:`\text{X}` is the input tensor. The shape of full convolution kernel is
-    :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size})`,
-    where `group` is the number of groups to split the input `x` in the channel dimension.
+    where :math:`bias` is the output channel bias, :math:`ccor` is
+    the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
+    , :math:`weight` is the convolution kernel value and :math:`X` represents the input feature map.
+
+    Here are the indices' meanings:
+    - :math:`i` corresponds to the batch number, ranging from 0 to N-1, where N is the batch size of the input.
 
-    For more details, please refers to the paper `Gradient Based Learning Applied to Document
-    Recognition <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_.
+    - :math:`j` corresponds to the output channel, ranging from 0 to C_{out}-1, where C_{out} is the number of
+      output channels, which is also equal to the number of kernels.
+
+    - :math:`k` corresponds to the input channel, ranging from 0 to C_{in}-1, where C_{in} is the number of
+      input channels, which is also equal to the number of channels in the convolutional kernels.
+
+    Therefore, in the above formula, :math:`{bias}(C_{out_j})` represents the bias of the :math:`j`-th
+    output channel, :math:`{weight}(C_{out_j}, k)` represents the slice of the :math:`j`-th convolutional
+    kernel in the :math:`k`-th channel, and :math:`{X}(N_i, k)` represents the slice of the :math:`k`-th input
+    channel in the :math:`i`-th batch of the input feature map.
+
+    The shape of the convolutional kernel is given by :math:`(kernel\_size)`,
+    where :math:`kernel\_size` is the width of the kernel.
+    If we consider the input and output channels as well as the `group` parameter, the complete kernel shape
+    will be :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size})`,
+    where `group` is the number of groups dividing `x`'s input channel when applying group convolution.
+
+    For more details about convolution layer, please refer to `Gradient Based Learning Applied to Document Recognition
+    <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_
+    and `ConvNets <http://cs231n.github.io/convolutional-networks/>`_ .
 
     Note:
         On Ascend platform, only group convolution in depthwise convolution scenarios is supported.
@@ -370,39 +420,52 @@ class Conv1d(_Conv):
         in_channels (int): The channel number of the input tensor of the Conv1d layer.
         out_channels (int): The channel number of the output tensor of the Conv1d layer.
         kernel_size (int): Specifies the width of the 1D convolution kernel.
-        stride (int): The movement stride of the 1D convolution kernel. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"`` .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
-
-        padding (int): The number of padding on both sides of input.
-            The value should be greater than or equal to 0. Default: ``0`` .
-        dilation (int): Dilation size of 1D convolution kernel. If :math:`k > 1`, the kernel is sampled
-            every `k` elements. The value of `k` is in range of [1, L]. Default: ``1`` .
-        group (int): Splits filter into groups, `in_channels` and `out_channels` must be
+        stride (int, optional): The movement stride of the 1D convolution kernel. Default: ``1`` .
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input at the begin and end so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally. If the amount is even, it is
+              uniformly distributed around the input, if it is odd, the excess padding is goes to the right side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible length. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              at the begin and end is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
+
+        padding (Union(int, tuple[int], list[int]), optional):  Specifies the amount of padding to apply on
+            both side of `input` when `pad_mode` is set to ``"pad"``. The
+            paddings of left and right are the same, equal to padding or padding[0] when padding is a tuple of
+            1 integer. Default: ``0`` .
+        dilation (Union(int, tuple[int]), optional): Specifies the dilation rate to use for dilated convolution.
+            It can be a single int or a tuple of 1 integer.
+            Assuming :math:`dilation=(d0,)`, the convolutional kernel samples the input with a
+            spacing of :math:`d0-1` elements in the width direction.
+            The value should be in the ranges [1, L].
+            Default: ``1`` .
+        group (int, optional): Splits filter into groups, `in_channels` and `out_channels` must be
             divisible by `group`. Default: ``1`` .
-        has_bias (bool): Whether the Conv1d layer has a bias parameter. Default: ``False`` .
-        weight_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of weight parameter.
+        has_bias (bool, optional): Whether the Conv1d layer has a bias parameter. Default: ``False`` .
+        weight_init (Union[Tensor, str, Initializer, numbers.Number], optional):
+            Initialization method of weight parameter.
             It can be a Tensor, a string, an Initializer or a numbers.Number. When a string is specified,
             values from ``'TruncatedNormal'`` , ``'Normal'`` , ``'Uniform'`` , ``'HeUniform'`` and ``'XavierUniform'``
             distributions as well as constant 'One' and 'Zero' distributions are possible. Alias ``'xavier_uniform'`` ,
             ``'he_uniform'`` , ``'ones'`` and ``'zeros'`` are acceptable. Uppercase and lowercase are both acceptable.
-            Refer to the values of Initializer for more details. Default: ``None`` ,
-            weight will be initialized using HeUniform.
-        bias_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of bias parameter.
+            Refer to the values of
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , weight will be initialized using ``'HeUniform'``.
+        bias_init (Union[Tensor, str, Initializer, numbers.Number], optional): Initialization method of bias parameter.
             Available initialization methods are the same as 'weight_init'. Refer to the values of
-            Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , bias will be initialized using ``'Uniform'``.
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
 
     Inputs:
-        - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, L_{in})`.
+        - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, L_{in})` .
 
     Outputs:
         Tensor of shape :math:`(N, C_{out}, L_{out})`.
@@ -456,7 +519,8 @@ class Conv1d(_Conv):
                  group=1,
                  has_bias=False,
                  weight_init=None,
-                 bias_init=None):
+                 bias_init=None,
+                 dtype=mstype.float32):
         """Initialize Conv1d."""
         Validator.check_value_type("kernel_size", kernel_size, [int], self.cls_name)
         Validator.check_value_type("stride", stride, [int], self.cls_name)
@@ -495,7 +559,8 @@ class Conv1d(_Conv):
             group,
             has_bias,
             weight_init,
-            bias_init)
+            bias_init,
+            dtype=dtype)
         self.padding = (0, 0, padding, padding)
         Validator.check_string(pad_mode, ['valid', 'same', 'pad'], 'pad_mode', self.cls_name)
         self.conv2d = P.Conv2D(out_channel=self.out_channels,
@@ -531,31 +596,48 @@ def _check_input_5dims(input_shape, op_name):
 
 class Conv3d(_Conv):
     r"""
-    Calculates the 3D convolution on the input tensor. The input is typically of shape
-    :math:`(N, C_{in}, D_{in}, H_{in}, W_{in})`,
-    where :math:`N` is batch size, :math:`C_{in}` is a number of channels,
-    :math:`D_{in}, H_{in}, W_{in}` are the depth, height and width of the feature layer respectively.
-    For the tensor of each batch, its shape is :math:`(C_{in}, D_{in}, H_{in}, W_{in})`, the formula is defined as:
+    3D convolution layer.
+
+    Applies a 3D convolution over an input tensor which is typically of shape
+    :math:`(N, C_{in}, D_{in}, H_{in}, W_{in})`, where :math:`N` is batch size, :math:`C` is channel number,
+    :math:`D` is feature depth, :math:`H` is feature height, :math:`W` is feature width.
+
+    The output is calculated based on formula:
 
     .. math::
 
         \text{out}(N_i, C_{\text{out}_j}) = \text{bias}(C_{\text{out}_j}) +
         \sum_{k = 0}^{C_{in} - 1} \text{ccor}({\text{weight}(C_{\text{out}_j}, k), \text{X}(N_i, k)})
 
-    where :math:`ccor` is the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
-    :math:`C_{in}` is the channel number of the input, :math:`out_{j}` corresponds to the :math:`j`-th channel of
-    the output and :math:`j` is in the range of :math:`[0, C_{out}-1]`. :math:`\text{weight}(C_{\text{out}_j}, k)`
-    is a convolution kernel slice with shape
-    :math:`(\text{kernel_size[0]}, \text{kernel_size[1]}, \text{kernel_size[2]})`,
-    where :math:`\text{kernel_size[0]}`, :math:`\text{kernel_size[1]}` and :math:`\text{kernel_size[2]}` are
-    the depth, height and width of the convolution kernel respectively. :math:`\text{bias}` is the bias parameter
-    and :math:`\text{X}` is the input tensor.
-    The shape of full convolution kernel is
-    :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size[0]}, \text{kernel_size[1]}, \text{kernel_size[2]})`,
-    where `group` is the number of groups to split the input `x` in the channel dimension.
-
-    For more details, please refers to the paper `Gradient Based Learning Applied to Document
-    Recognition <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_.
+    where :math:`bias` is the output channel bias, :math:`ccor` is
+    the `cross-correlation <https://en.wikipedia.org/wiki/Cross-correlation>`_,
+    , :math:`weight` is the convolution kernel value and :math:`X` represents the input feature map.
+
+    Here are the indices' meanings:
+    - :math:`i` corresponds to the batch number, ranging from 0 to N-1, where N is the batch size of the input.
+
+    - :math:`j` corresponds to the output channel, ranging from 0 to C_{out}-1, where C_{out} is the number of
+      output channels, which is also equal to the number of kernels.
+
+    - :math:`k` corresponds to the input channel, ranging from 0 to C_{in}-1, where C_{in} is the number of
+      input channels, which is also equal to the number of channels in the convolutional kernels.
+
+    Therefore, in the above formula, :math:`{bias}(C_{out_j})` represents the bias of the :math:`j`-th
+    output channel, :math:`{weight}(C_{out_j}, k)` represents the slice of the :math:`j`-th convolutional
+    kernel in the :math:`k`-th channel, and :math:`{X}(N_i, k)` represents the slice of the :math:`k`-th input
+    channel in the :math:`i`-th batch of the input feature map.
+
+    The shape of the convolutional kernel is given by
+    :math:`(\text{kernel_size[0]}, \text{kernel_size[1]}, \text{kernel_size[2]})`
+    where :math:`kernel\_size[0]` , :math:`kernel\_size[1]` and :math:`kernel\_size[2]` are the depth,
+    height and width of the kernel, respectively.
+    If we consider the input and output channels as well as the `group` parameter, the complete kernel shape
+    will be :math:`(C_{out}, C_{in} / \text{group}, \text{kernel_size[0]},
+    \text{kernel_size[1]}, \text{kernel_size[2]})`,
+    where `group` is the number of groups dividing `x`'s input channel when applying group convolution.
+
+    For more details about convolution layer, please refer to `Gradient Based Learning Applied to Document Recognition
+    <http://vision.stanford.edu/cs598_spring07/papers/Lecun98.pdf>`_.
 
     Note:
         On Ascend platform, only group convolution in depthwise convolution scenarios is supported.
@@ -565,49 +647,65 @@ class Conv3d(_Conv):
         in_channels (int): The channel number of the input tensor of the Conv3d layer.
         out_channels (int): The channel number of the output tensor of the Conv3d layer.
         kernel_size (Union[int, tuple[int]]): Specifies the depth, height and width of the 3D convolution kernel.
-            The data type is an integer or a tuple of three integers. An integer represents the depth, height
-            and width of the convolution kernel. A tuple of three integers represents the depth, height
-            and width of the convolution kernel respectively.
-        stride (Union[int, tuple[int]]): The movement stride of the 3D convolution kernel.
+            It can be a single int or a tuple of 3 integers. A single int means the value is for depth, height
+            and the width. A tuple of 3 ints means the first value is
+            for depth and the rest is for the height and width.
+        stride (Union[int, tuple[int]], optional): The movement stride of the 3D convolution kernel.
             The data type is an integer or a tuple of three integers. An integer represents the movement step size
             in depth, height and width directions. A tuple of three integers represents the movement step size
             in the depth, height and width directions respectively. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"`` .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
-
-        padding (Union(int, tuple[int])): The number of padding on the depth, height and width directions of the input.
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input around its depth/height/width dimension so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally.  If the amount is even,
+              it isuniformly distributed around the input, if it is odd, the excess amount goes
+              to the front/right/bottom side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible depth, height and width. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              in the depth, height and width dimension is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
+
+        padding (Union(int, tuple[int]), optional): The number of padding on the depth,
+            height and width directions of the input.
             The data type is an integer or a tuple of six integers. If `padding` is an integer,
             then the head, tail, top, bottom, left, and right padding are all equal to `padding`.
             If `padding` is a tuple of six integers, then the head, tail, top, bottom, left, and right padding
             is equal to `padding[0]`, `padding[1]`, `padding[2]`, `padding[3]`, `padding[4]` and `padding[5]`
             respectively. The value should be greater than or equal to 0. Default: ``0`` .
-        dilation (Union[int, tuple[int]]): Dilation size of 3D convolution kernel.
-            The data type is an integer or a tuple of three integers. If :math:`k > 1`, the kernel is sampled
-            every `k` elements. The value of `k` on the depth, height and width directions is in range of
-            [1, D], [1, H] and [1, W] respectively. Default: ``1`` .
-        group (int): Splits filter into groups, `in_channels` and `out_channels` must be
+        dilation (Union[int, tuple[int]], optional): Specifies the dilation rate to use for dilated convolution.
+            It can be a single int or a tuple of 3 integers. A single int means the dilation size is the same
+            in the depth, height and width directions. A tuple of 3 ints represents the dilation size in
+            the depth, height and width directions, respectively.
+            Assuming :math:`dilation=(d0, d1, d2)`, the convolutional kernel samples the input with a
+            spacing of :math:`d0-1` elements in the depth direction, :math:`d1-1` elements in the height direction,
+            :math:`d2-1` elements in the width direction respectively.
+            The values in the depth, height and width dimensions are in
+            the ranges [1, D], [1, H] and [1, W], respectively.
+            Default: ``1`` .
+        group (int, optional): Splits filter into groups, `in_channels` and `out_channels` must be
             divisible by `group`. Default: ``1`` .
-        has_bias (bool): Whether the Conv3d layer has a bias parameter. Default: ``False`` .
-        weight_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of weight parameter.
+        has_bias (bool, optional): Whether the Conv3d layer has a bias parameter. Default: ``False`` .
+        weight_init (Union[Tensor, str, Initializer, numbers.Number], optional):
+            Initialization method of weight parameter.
             It can be a Tensor, a string, an Initializer or a numbers.Number. When a string is specified,
             values from ``'TruncatedNormal'`` , ``'Normal'`` , ``'Uniform'`` , ``'HeUniform'`` and ``'XavierUniform'``
             distributions as well as constant ``'One'`` and ``'Zero'`` distributions are possible. Alias
             ``'xavier_uniform'`` , ``'he_uniform'`` , ``'ones'`` and ``'zeros'`` are acceptable. Uppercase and
-            lowercase are both acceptable. Refer to the values of Initializer for more details. Default: ``None`` ,
-            weight will be initialized using HeUniform.
-        bias_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of bias parameter.
+            lowercase are both acceptable. Refer to the values of
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , weight will be initialized using ``'HeUniform'``.
+        bias_init (Union[Tensor, str, Initializer, numbers.Number], optional): Initialization method of bias parameter.
             Available initialization methods are the same as 'weight_init'. Refer to the values of
-            Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
-        data_format (str): The optional value for data format. Currently only support ``'NCDHW'``.
+            `Initializer <https://www.mindspore.cn/docs/en/r2.2/api_python/mindspore.common.initializer.html>`_,
+            for more details. Default: ``None`` , bias will be initialized using ``'Uniform'`` .
+        data_format (str, optional): The optional value for data format. Currently only support ``'NCDHW'`` .
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
+
 
     Inputs:
         - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, D_{in}, H_{in}, W_{in})`.
@@ -687,7 +785,8 @@ class Conv3d(_Conv):
                  has_bias=False,
                  weight_init=None,
                  bias_init=None,
-                 data_format='NCDHW'):
+                 data_format='NCDHW',
+                 dtype=mstype.float32):
         """Initialize Conv3d."""
         if not (in_channels % group == 0 and out_channels % group == 0):
             raise ValueError("The argument 'group' should be divisible by 'in_channels' " \
@@ -711,8 +810,9 @@ class Conv3d(_Conv):
             has_bias,
             weight_init,
             bias_init,
-            data_format)
-        out_channels = self.out_channels // group
+            data_format,
+            dtype=dtype)
+        out_channels = self.out_channels
         self.conv3d = P.Conv3D(out_channel=out_channels,
                                kernel_size=self.kernel_size,
                                mode=1,
@@ -720,33 +820,17 @@ class Conv3d(_Conv):
                                pad=self.padding,
                                stride=self.stride,
                                dilation=self.dilation,
-                               group=1,
+                               group=group,
                                data_format=self.data_format)
         self.bias_add = P.BiasAdd(data_format=self.data_format)
         self.shape = P.Shape()
-        self.concat = P.Concat(1)
-        self.split_0 = P.Split(0, self.group)
-        self.split_1 = P.Split(1, self.group)
 
     def construct(self, x):
         x_shape = self.shape(x)
         _check_input_5dims(x_shape, self.cls_name)
-        if self.group == 1:
-            out = self.conv3d(x, self.weight)
-            if self.has_bias:
-                out = self.bias_add(out, self.bias)
-        else:
-            features = self.split_1(x)
-            weights = self.split_0(self.weight)
-            outputs = ()
-            for i in range(self.group):
-                output = self.conv3d(features[i], weights[i])
-                outputs = outputs + (output,)
-            out = self.concat(outputs)
-            if self.bias is not None:
-                new_shape = [1 for _ in range(out.ndim)]
-                new_shape[1] = self.out_channels
-                out = out + self.bias.reshape(new_shape)
+        out = self.conv3d(x, self.weight)
+        if self.has_bias:
+            out = self.bias_add(out, self.bias)
         return out
 
 
@@ -776,17 +860,21 @@ class Conv3dTranspose(_Conv):
             The data type is an integer or a tuple of three integers. An integer represents the movement step size
             in depth, height and width directions. A tuple of three integers represents the movement step size
             in the depth, height and width directions respectively. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"``  .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input around its depth/height/width dimension so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally.  If the amount is even,
+              it isuniformly distributed around the input, if it is odd, the excess amount goes
+              to the front/right/bottom side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible depth, height and width. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              in the depth, height and width dimension is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
 
         padding (Union(int, tuple[int])): The number of padding on the depth, height and width directions of the input.
             The data type is an integer or a tuple of six integers. If `padding` is an integer,
@@ -820,10 +908,11 @@ class Conv3dTranspose(_Conv):
             Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
         data_format (str): The optional value for data format. Currently only support ``'NCDHW'`` .
             Default: ``'NCDHW'`` .
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
 
     Inputs:
         - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, D_{in}, H_{in}, W_{in})`.
-          Currently input data type only support float16 and float32.
+          Currently input data dtype only support float16 and float32.
 
     Outputs:
         Tensor, the shape is :math:`(N, C_{out}, D_{out}, H_{out}, W_{out})`.
@@ -902,7 +991,8 @@ class Conv3dTranspose(_Conv):
                  has_bias=False,
                  weight_init=None,
                  bias_init=None,
-                 data_format='NCDHW'):
+                 data_format='NCDHW',
+                 dtype=mstype.float32):
         """Initialize Conv3dTranspose."""
         if not (in_channels % group == 0 and out_channels % group == 0):
             raise ValueError("The argument 'group' should be divisible by 'in_channels' " \
@@ -929,7 +1019,8 @@ class Conv3dTranspose(_Conv):
             weight_init,
             bias_init,
             data_format,
-            transposed=True)
+            transposed=True,
+            dtype=dtype)
         self.conv3d_transpose = P.Conv3DTranspose(in_channel=self.in_channels,
                                                   out_channel=self.out_channels,
                                                   kernel_size=self.kernel_size,
@@ -996,17 +1087,20 @@ class Conv2dTranspose(_Conv):
             The data type is an integer or a tuple of two integers. An integer represents the movement step size
             in both height and width directions. A tuple of two integers represents the movement step size in the height
             and width directions respectively. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"`` .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input around its edges so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally, If the amount is even, it is
+              uniformly distributed around the input, if it is odd, the excess amount goes to the right/bottom side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible height and width. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              in the height and width directions is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
 
         padding (Union[int, tuple[int]]): The number of padding on the height and width directions of the input.
             The data type is an integer or a tuple of four integers. If `padding` is an integer,
@@ -1037,6 +1131,7 @@ class Conv2dTranspose(_Conv):
         bias_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of bias parameter.
             Available initialization methods are the same as 'weight_init'. Refer to the values of
             Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
 
     Inputs:
         - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, H_{in}, W_{in})`.
@@ -1109,7 +1204,8 @@ class Conv2dTranspose(_Conv):
                  group=1,
                  has_bias=False,
                  weight_init=None,
-                 bias_init=None):
+                 bias_init=None,
+                 dtype=mstype.float32):
         """Initialize Conv2dTranspose."""
         kernel_size = twice(kernel_size)
         stride = twice(stride)
@@ -1135,7 +1231,8 @@ class Conv2dTranspose(_Conv):
             has_bias,
             weight_init,
             bias_init,
-            transposed=True)
+            transposed=True,
+            dtype=dtype)
 
         self.in_channels = in_channels
         self.out_channels = out_channels
@@ -1218,17 +1315,20 @@ class Conv1dTranspose(_Conv):
         out_channels (int): The channel number of the output tensor of the Conv1dTranspose layer.
         kernel_size (int): Specifies the width of the 1D convolution kernel.
         stride (int): The movement stride of the 1D convolution kernel. Default: ``1`` .
-        pad_mode (str): Specifies padding mode. The optional values are
-            ``"same"`` , ``"valid"`` , ``"pad"`` . Default: ``"same"`` .
-
-            - ``"same"``: The width of the output is the same as the value of the input divided by `stride`.
-              If this mode is set, the value of `padding` must be 0.
-
-            - ``"valid"``: Returns a valid calculated output without padding. Excess pixels that do not satisfy the
-              calculation will be discarded. If this mode is set, the value of `padding` must be 0.
-
-            - ``"pad"``: Pads the input. Padding `padding` size of zero on both sides of the input.
-              If this mode is set, the value of `padding` must be greater than or equal to 0.
+        pad_mode (str, optional): Specifies the padding mode with a padding value of 0. It can be set to:
+            ``"same"`` , ``"valid"`` or ``"pad"`` . Default: ``"same"`` .
+
+            - ``"same"``: Pad the input at the begin and end so that the shape of input and output
+              are the same when `stride` is set to ``1``.
+              The amount of padding to is calculated by the operator internally. If the amount is even, it is
+              uniformly distributed around the input, if it is odd, the excess padding is goes to the right side.
+              If this mode is set, `padding` must be 0.
+            - ``"valid"``: No padding is applied to the input, and the output returns the maximum
+              possible length. Extra pixels that could not complete a full stride will
+              be discarded. If this mode is set, `padding` must be 0.
+            - ``"pad"``: Pad the input with a specified amount. In this mode, the amount of padding
+              at the begin and end is determined by the `padding` parameter.
+              If this mode is set, `padding` must be greater than or equal to 0.
 
         padding (int): The number of padding on both sides of input.
             The value should be greater than or equal to 0. Default: ``0`` .
@@ -1247,6 +1347,7 @@ class Conv1dTranspose(_Conv):
         bias_init (Union[Tensor, str, Initializer, numbers.Number]): Initialization method of bias parameter.
             Available initialization methods are the same as 'weight_init'. Refer to the values of
             Initializer for more details. Default: ``None`` , bias will be initialized using Uniform.
+        dtype (:class:`mindspore.dtype`): Dtype of Parameters. Default: ``mstype.float32`` .
 
     Inputs:
         - **x** (Tensor) - Tensor of shape :math:`(N, C_{in}, L_{in})`.
@@ -1294,7 +1395,8 @@ class Conv1dTranspose(_Conv):
                  group=1,
                  has_bias=False,
                  weight_init=None,
-                 bias_init=None):
+                 bias_init=None,
+                 dtype=mstype.float32):
         """Initialize Conv1dTranspose."""
         Validator.check_value_type("kernel_size", kernel_size, [int], self.cls_name)
         Validator.check_value_type("stride", stride, [int], self.cls_name)
@@ -1331,7 +1433,8 @@ class Conv1dTranspose(_Conv):
             has_bias,
             weight_init,
             bias_init,
-            transposed=True)
+            transposed=True,
+            dtype=dtype)
         self.padding = (0, 0, padding, padding)
         self.in_channels = in_channels
         self.out_channels = out_channels