m53297601
/
mindspore
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

enhancement of API of operations

Browse Source
pull/6169/head
simson 4 years ago
parent ec5ba10c82
commit d0aba89ffe
12 changed files with 149 additions and 146 deletions
Show Stats Download Patch File Download Diff File

20
mindspore/nn/graph_kernels/graph_kernels.py
Unescape View File

@ -310,7 +310,7 @@ class ReduceMean(GraphKernel):
Args:
keep_dims (bool): If True, keep these reduced dimensions and the length is 1.
If False, don't keep these dimensions. Default : False.
If False, don't keep these dimensions. Default: False.
Inputs:
- **input_x** (Tensor[Number]) - The input tensor.
@ -318,13 +318,13 @@ class ReduceMean(GraphKernel):
Only constant value is allowed.
Outputs:
Tensor, has the same dtype as the 'input_x'.
Tensor, has the same dtype as the `input_x`.
- If axis is (), and keep_dims is false,
- If axis is (), and keep_dims is False,
the output is a 0-D tensor representing the sum of all elements in the input tensor.
- If axis is int, set as 2, and keep_dims is false,
- If axis is int, set as 2, and keep_dims is False,
the shape of output is :math:`(x_1, x_3, ..., x_R)`.
- If axis is tuple(int), set as (2, 3), and keep_dims is false,
- If axis is tuple(int), set as (2, 3), and keep_dims is False,
the shape of output is :math:`(x_1, x_4, ..., x_R)`.
Examples:
@ -394,7 +394,7 @@ class SoftmaxCrossEntropyWithLogits(GraphKernel):
- **labels** (Tensor) - Ground truth labels, with shape :math:`(N, C)`.
Outputs:
Tuple of 2 Tensor, the loss shape is `(N,)`, and the dlogits with the same shape as `logits`.
Tuple of 2 Tensors, the loss shape is `(N,)`, and the dlogits with the same shape as `logits`.
Examples:
>>> logits = Tensor([[2, 4, 1, 4, 5], [2, 1, 2, 4, 3]], mindspore.float32)
@ -610,7 +610,7 @@ class LayerNormBetaGammaBackprop(GraphKernel):
- **input_gamma**(Tensor) - The fourth input of the forward function of LayerNorm.
Outputs:
Tuple of 2 Tensor, the backprop outputs.
Tuple of 2 Tensors, the backprop outputs.
- **pd_beta**(Tensor) - The first item of return value of this operator, will be used as
the second item of the LayerNorm's backprop function.
@ -1137,12 +1137,12 @@ class LambNextMV(GraphKernel):
- **inputsx3** (Tensor) - The thirteenth input tensor to be computed.
Outputs:
Tuple of 2 Tensor.
Tuple of 2 Tensors.
- **add3** (Tensor) - the shape is the same as the one after broadcasting, and the data type is
the one with high precision or high digits among the inputs.
the one with higher precision or higher digits among the inputs.
- **realdiv4** (Tensor) - the shape is the same as the one after broadcasting, and the data type is
the one with high precision or high digits among the inputs.
the one with higher precision or higher digits among the inputs.
Examples:
>>> lamb_next_mv = LambNextMV()

10
mindspore/nn/layer/math.py
Unescape View File

@ -46,13 +46,13 @@ class ReduceLogSumExp(Cell):
Only constant value is allowed.
Outputs:
Tensor, has the same dtype as the 'input_x'.
Tensor, has the same dtype as the `input_x`.
- If axis is (), and keep_dims is false,
- If axis is (), and keep_dims is False,
the output is a 0-D tensor representing the sum of all elements in the input tensor.
- If axis is int, set as 2, and keep_dims is false,
- If axis is int, set as 2, and keep_dims is False,
the shape of output is :math:`(x_1, x_3, ..., x_R)`.
- If axis is tuple(int), set as (2, 3), and keep_dims is false,
- If axis is tuple(int), set as (2, 3), and keep_dims is False,
the shape of output is :math:`(x_1, x_4, ..., x_R)`.
Examples:
@ -214,7 +214,7 @@ class LGamma(Cell):
- **input_x** (Tensor[Number]) - The input tensor. Only float16, float32 are supported.
Outputs:
Tensor, has the same shape and dtype as the 'input_x'.
Tensor, has the same shape and dtype as the `input_x`.
Examples:
>>> input_x = Tensor(np.array(2, 3, 4).astype(np.float32))

52
mindspore/ops/operations/array_ops.py
Unescape View File

@ -181,8 +181,8 @@ class SameTypeShape(PrimitiveWithInfer):
Checks whether data type and shape of two tensors are the same.
Raises:
TypeError: If data type not the same.
ValueError: If shape of two tensors not the same.
TypeError: If the data types of two tensors are not the same.
ValueError: If the shapes of two tensors are not the same.
Inputs:
- **input_x** (Tensor) - The shape of tensor is :math:`(x_1, x_2, ..., x_R)`.
@ -362,7 +362,7 @@ class Reshape(PrimitiveWithInfer):
Reshapes input tensor with the same values based on a given shape tuple.
Raises:
ValueError: Given a shape tuple, if it has more than one -1; or if the product
ValueError: Given a shape tuple, if it has several -1; or if the product
of its elements is less than or equal to 0 or cannot be divided by the product
of the input tensor shape; or if it does not match the input's array size.
@ -671,10 +671,10 @@ class SparseGatherV2(GatherV2):
class Padding(PrimitiveWithInfer):
"""
Extend the last dimension of input tensor from 1 to pad_dim_size, fill with 0.
Extend the last dimension of input tensor from 1 to pad_dim_size, by filling with 0.
Args:
pad_dim_size (int): The extend value of last dimension of x, must be positive.
pad_dim_size (int): The value of the last dimension of x to be extended, which must be positive.
Inputs:
- **x** (Tensor) - The shape of tensor is :math:`(x_1, x_2, ..., x_R)`. The rank of x should be at least 2.
@ -921,7 +921,7 @@ class Fill(PrimitiveWithInfer):
class OnesLike(PrimitiveWithInfer):
"""
Creates a new tensor. All elements' value are 1.
Creates a new tensor. The values of all elements are 1.
Returns a tensor of ones with the same shape and type as the input.
@ -1025,7 +1025,7 @@ class TupleToArray(PrimitiveWithInfer):
class ScalarToArray(PrimitiveWithInfer):
"""
Converts scalar to `Tensor`.
Converts a scalar to a `Tensor`.
Inputs:
- **input_x** (Union[int, float]) - The input is a scalar. Only constant value is allowed.
@ -1054,7 +1054,7 @@ class ScalarToArray(PrimitiveWithInfer):
class ScalarToTensor(PrimitiveWithInfer):
"""
Converts scalar to `Tensor`, and convert data type to specified type.
Converts a scalar to a `Tensor`, and convert data type to specified type.
Inputs:
- **input_x** (Union[int, float]) - The input is a scalar. Only constant value is allowed.
@ -1653,11 +1653,11 @@ class ParallelConcat(PrimitiveWithInfer):
The input tensors are all required to have size 1 in the first dimension.
Inputs:
- **values** (tuple, list) - Tuple or list of input tensors. The data type and shape of these
tensors must be same.
- **values** (tuple, list) - A tuple or a list of input tensors. The data type and shape of these
tensors must be the same.
Outputs:
Tensor, data type same as `values`.
Tensor, data type is the same as `values`.
Examples:
>>> data1 = Tensor(np.array([[0, 1]]).astype(np.int32))
@ -1726,7 +1726,7 @@ class Pack(PrimitiveWithInfer):
If :math:`0 \le axis`, the shape of the output tensor is :math:`(x_1, x_2, ..., x_{axis}, N, x_{axis+1}, ..., x_R)`.
Args:
axis (int): Dimension along which to pack. Default: 0.
axis (int): Dimension to pack. Default: 0.
Negative values wrap around. The range is [-(R+1), R+1).
Inputs:
@ -1736,8 +1736,8 @@ class Pack(PrimitiveWithInfer):
Tensor. A packed Tensor with the same type as `input_x`.
Raises:
TypeError: If the data types of elements in input_x are not the same.
ValueError: If length of input_x is not greater than 1;
TypeError: If the data types of elements in `input_x` are not the same.
ValueError: If the length of `input_x` is not greater than 1;
or if axis is out of the range [-(R+1), R+1);
or if the shapes of elements in input_x are not the same.
@ -2267,7 +2267,7 @@ class Diag(PrimitiveWithInfer):
- **input_x** (Tensor) - The input tensor. The input shape should be less than 5d.
Outputs:
Tensor, has the same dtype as the 'input_x'.
Tensor, has the same dtype as the `input_x`.
Examples:
>>> input_x = Tensor([1, 2, 3, 4])
@ -2437,7 +2437,7 @@ class ResizeNearestNeighbor(PrimitiveWithInfer):
r"""
Resize the input tensor by using nearest neighbor algorithm.
Resize input tensor to given size by using nearest neighbor algorithm. The nearest
Resize the input tensor to a given size by using the nearest neighbor algorithm. The nearest
neighbor algorithm selects the value of the nearest point and does not consider the
values of neighboring points at all, yielding a piecewise-constant interpolant.
@ -2665,8 +2665,8 @@ class ScatterMax(_ScatterOp):
Inputs:
- **input_x** (Parameter) - The target parameter.
- **indices** (Tensor) - The index to do max operation whose data type should be mindspore.int32.
- **updates** (Tensor) - The tensor doing the maximum operation with `input_x`,
the data type is same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
- **updates** (Tensor) - The tensor that performs the maximum operation with `input_x`,
the data type is the same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
Outputs:
Parameter, the updated `input_x`.
@ -2739,8 +2739,8 @@ class ScatterAdd(_ScatterOp):
Inputs:
- **input_x** (Parameter) - The target parameter.
- **indices** (Tensor) - The index to do add operation whose data type should be mindspore.int32.
- **updates** (Tensor) - The tensor doing the add operation with `input_x`,
the data type is same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
- **updates** (Tensor) - The tensor that performs the add operation with `input_x`,
the data type is the same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
Outputs:
Parameter, the updated `input_x`.
@ -2841,8 +2841,8 @@ class ScatterDiv(_ScatterOp):
Inputs:
- **input_x** (Parameter) - The target parameter.
- **indices** (Tensor) - The index to do div operation whose data type should be mindspore.int32.
- **updates** (Tensor) - The tensor doing the div operation with `input_x`,
the data type is same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
- **updates** (Tensor) - The tensor that performs the div operation with `input_x`,
the data type is the same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
Outputs:
Parameter, the updated `input_x`.
@ -3510,12 +3510,12 @@ class ReverseSequence(PrimitiveWithInfer):
Reverses variable length slices.
Args:
seq_dim (int): The dimension along which reversal is performed. Required.
batch_dim (int): The input is sliced along this dimmension. Default: 0.
seq_dim (int): The dimension where reversal is performed. Required.
batch_dim (int): The input is sliced in this dimension. Default: 0.
Inputs:
- **x** (Tensor) - The input to reverse, support all number types including bool.
- **seq_lengths** (Tensor) - Must be 1-D vector with types: int32, int64.
- **x** (Tensor) - The input to reverse, supporting all number types including bool.
- **seq_lengths** (Tensor) - Must be a 1-D vector with int32 or int64 types.
Outputs:
Reversed tensor with the same shape and data type as input.

6
mindspore/ops/operations/comm_ops.py
Unescape View File

@ -26,7 +26,7 @@ class ReduceOp:
"""
Operation options for reduce tensors.
There are four kinds of operation options, "SUM","MAX","MIN","PROD".
There are four kinds of operation options, "SUM", "MAX", "MIN", and "PROD".
- SUM: Take the sum.
- MAX: Take the maximum.
@ -232,12 +232,12 @@ class ReduceScatter(PrimitiveWithInfer):
Args:
op (str): Specifies an operation used for element-wise reductions,
like sum, max, avg. Default: ReduceOp.SUM.
like SUM, MAX, AVG. Default: ReduceOp.SUM.
group (str): The communication group to work on. Default: "hccl_world_group".
Raises:
TypeError: If any of operation and group is not a string.
ValueError: If the first dimension of input can not be divided by rank size.
ValueError: If the first dimension of the input cannot be divided by the rank size.
Examples:
>>> from mindspore import Tensor

2
mindspore/ops/operations/control_ops.py
Unescape View File

@ -145,7 +145,7 @@ class Merge(PrimitiveWithInfer):
One and only one of the inputs should be selected as the output
Inputs:
- **inputs** (Union(Tuple, List)) - The data to be merged. All tuple elements should have same data type.
- **inputs** (Union(Tuple, List)) - The data to be merged. All tuple elements should have the same data type.
Outputs:
tuple. Output is tuple(`data`, `output_index`). The `data` has the same shape of `inputs` element.

2
mindspore/ops/operations/debug_ops.py
Unescape View File

@ -42,7 +42,7 @@ SUMMARY_RETURN_VALUE = {'dtype': mstype.int32, 'shape': [1], 'value': None}
class ScalarSummary(PrimitiveWithInfer):
"""
Output scalar to protocol buffer through scalar summary operator.
Output a scalar to a protocol buffer through a scalar summary operator.
Inputs:
- **name** (str) - The name of the input variable, it should not be an empty string.

2
mindspore/ops/operations/inner_ops.py
Unescape View File

@ -28,7 +28,7 @@ class ScalarCast(PrimitiveWithInfer):
Inputs:
- **input_x** (scalar) - The input scalar. Only constant value is allowed.
- **input_y** (mindspore.dtype) - The type should cast to be. Only constant value is allowed.
- **input_y** (mindspore.dtype) - The type to be cast. Only constant value is allowed.
Outputs:
Scalar. The type is the same as the python type corresponding to `input_y`.

112
mindspore/ops/operations/math_ops.py
View File

File diff suppressed because it is too large Load Diff

60
mindspore/ops/operations/nn_ops.py
Unescape View File

@ -298,7 +298,7 @@ class ReLU6(PrimitiveWithInfer):
It returns :math:`\min(\max(0,x), 6)` element-wise.
Inputs:
- **input_x** (Tensor) - The input tensor. With float16 or float32 data type.
- **input_x** (Tensor) - The input tensor, with float16 or float32 data type.
Outputs:
Tensor, with the same type and shape as the `input_x`.
@ -1238,7 +1238,7 @@ class MaxPool(_Pool):
class MaxPoolWithArgmax(_Pool):
r"""
Performs max pooling on the input Tensor and return both max values and indices.
Perform max pooling on the input Tensor and return both max values and indices.
Typically the input is of shape :math:`(N_{in}, C_{in}, H_{in}, W_{in})`, MaxPool outputs
regional maximum in the :math:`(H_{in}, W_{in})`-dimension. Given kernel size
@ -1272,7 +1272,7 @@ class MaxPoolWithArgmax(_Pool):
Data type should be float16 or float32.
Outputs:
Tuple of 2 Tensor, the maxpool result and where max values from.
Tuple of 2 Tensors, representing the maxpool result and where the max values are generated.
- **output** (Tensor) - Maxpooling result, with shape :math:`(N, C_{out}, H_{out}, W_{out})`.
- **mask** (Tensor) - Max values' index represented by the mask.
@ -1557,7 +1557,7 @@ class TopK(PrimitiveWithInfer):
- **k** (int) - Number of top elements to be computed along the last dimension, constant input is needed.
Outputs:
Tuple of 2 Tensor, the values and the indices.
Tuple of 2 Tensors, the values and the indices.
- **values** (Tensor) - The `k` largest elements along each last dimensional slice.
- **indices** (Tensor) - The indices of values within the last dimension of input.
@ -1609,7 +1609,7 @@ class SoftmaxCrossEntropyWithLogits(PrimitiveWithInfer):
- **labels** (Tensor) - Ground truth labels, with shape :math:`(N, C)`, has the same data type with `logits`.
Outputs:
Tuple of 2 Tensor, the loss shape is `(N,)`, and the dlogits with the same shape as `logits`.
Tuple of 2 Tensors, the loss shape is `(N,)`, and the dlogits with the same shape as `logits`.
Examples:
>>> logits = Tensor([[2, 4, 1, 4, 5], [2, 1, 2, 4, 3]], mindspore.float32)
@ -1961,7 +1961,7 @@ class SGD(PrimitiveWithInfer):
- **accum** (Tensor) - Accum(velocity) to be updated. With float16 or float32 data type.
- **momentum** (Tensor) - Momentum, a scalar tensor with float16 or float32 data type.
e.g. Tensor(0.1, mindspore.float32).
- **stat** (Tensor) - States to be updated with the same shape as gradient. With float16 or float32 data type.
- **stat** (Tensor) - States to be updated with the same shape as gradient, with float16 or float32 data type.
Outputs:
Tensor, parameters to be updated.
@ -2397,9 +2397,9 @@ class ResizeBilinear(PrimitiveWithInfer):
can be represented by different data types, but the data types of output images are always float32.
Args:
size (tuple[int]): A tuple of 2 int elements `(new_height, new_width)`, the new size for the images.
align_corners (bool): If it's true, rescale input by `(new_height - 1) / (height - 1)`,
which exactly aligns the 4 corners of images and resized images. If it's false,
size (tuple[int]): A tuple of 2 int elements `(new_height, new_width)`, the new size of the images.
align_corners (bool): If True, rescale input by `(new_height - 1) / (height - 1)`,
which exactly aligns the 4 corners of images and resized images. If False,
rescale by `new_height / height`. Default: False.
Inputs:
@ -2456,7 +2456,7 @@ class OneHot(PrimitiveWithInfer):
Has the same data type with as `on_value`.
Outputs:
Tensor, one_hot tensor. Tensor of shape :math:`(X_0, \ldots, X_{axis}, \text{depth} ,X_{axis+1}, \ldots, X_n)`.
Tensor, one-hot tensor. Tensor of shape :math:`(X_0, \ldots, X_{axis}, \text{depth} ,X_{axis+1}, \ldots, X_n)`.
Examples:
>>> indices = Tensor(np.array([0, 1, 2]), mindspore.int32)
@ -2590,13 +2590,13 @@ class PReLU(PrimitiveWithInfer):
Inputs:
- **input_x** (Tensor) - Float tensor, representing the output of the preview layer.
With data type of float16 or float32.
- **weight** (Tensor) - Float Tensor, w > 0, there is only two shapes are legitimate,
1 or the number of channels at input. With data type of float16 or float32.
- **weight** (Tensor) - Float Tensor, w > 0, there are only two shapes are legitimate,
1 or the number of channels of the input. With data type of float16 or float32.
Outputs:
Tensor, with the same type as `input_x`.
Detailed information, please refer to `nn.PReLU`.
For detailed information, please refer to `nn.PReLU`.
Examples:
>>> import mindspore
@ -2783,7 +2783,7 @@ class Pad(PrimitiveWithInfer):
paddings (tuple): The shape of parameter `paddings` is (N, 2). N is the rank of input data. All elements of
paddings are int type. For the input in `D` th dimension, paddings[D, 0] indicates how many sizes to be
extended ahead of the input tensor in the `D` th dimension, and paddings[D, 1] indicates how many sizes to
be extended behind of the input tensor in the `D` th dimension.
be extended behind the input tensor in the `D` th dimension.
Inputs:
- **input_x** (Tensor) - The input tensor.
@ -2833,21 +2833,21 @@ class MirrorPad(PrimitiveWithInfer):
Pads the input tensor according to the paddings and mode.
Args:
mode (str): Specifies padding mode. The optional values are "REFLECT", "SYMMETRIC".
mode (str): Specifies the padding mode. The optional values are "REFLECT" and "SYMMETRIC".
Default: "REFLECT".
Inputs:
- **input_x** (Tensor) - The input tensor.
- **paddings** (Tensor) - The paddings tensor. The value of `paddings` is a matrix(list),
and its shape is (N, 2). N is the rank of input data. All elements of paddings
are int type. For the input in `D` th dimension, paddings[D, 0] indicates how many sizes to be
are int type. For the input in the `D` th dimension, paddings[D, 0] indicates how many sizes to be
extended ahead of the input tensor in the `D` th dimension, and paddings[D, 1] indicates how many sizes to
be extended behind of the input tensor in the `D` th dimension.
be extended behind the input tensor in the `D` th dimension.
Outputs:
Tensor, the tensor after padding.
- If `mode` is "REFLECT", it uses a way of symmetrical copying throught the axis of symmetry to fill in.
- If `mode` is "REFLECT", it uses a way of symmetrical copying through the axis of symmetry to fill in.
If the `input_x` is [[1,2,3],[4,5,6],[7,8,9]] and `paddings` is [[1,1],[2,2]], then the
Outputs is [[6,5,4,5,6,5,4],[3,2,1,2,3,2,1],[6,5,4,5,6,5,4],[9,8,7,8,9,8,7],[6,5,4,5,6,5,4]].
- If `mode` is "SYMMETRIC", the filling method is similar to the "REFLECT". It is also copied
@ -3929,7 +3929,7 @@ class ApplyAdagrad(PrimitiveWithInfer):
With float32 or float16 data type.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `accum`.
@ -4012,7 +4012,7 @@ class ApplyAdagradV2(PrimitiveWithInfer):
With float16 or float32 data type.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `m`.
@ -4096,7 +4096,7 @@ class SparseApplyAdagrad(PrimitiveWithInfer):
The shape of `indices` must be the same as `grad` in first dimension, the type must be int32.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `accum`.
@ -4183,7 +4183,7 @@ class SparseApplyAdagradV2(PrimitiveWithInfer):
The shape of `indices` must be the same as `grad` in first dimension, the type must be int32.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `accum`.
@ -4273,7 +4273,7 @@ class ApplyProximalAdagrad(PrimitiveWithInfer):
- **grad** (Tensor) - Gradient with the same shape and dtype as `var`.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `accum`.
@ -4377,7 +4377,7 @@ class SparseApplyProximalAdagrad(PrimitiveWithCheck):
- **indices** (Tensor) - A vector of indices into the first dimension of `var` and `accum`.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **accum** (Tensor) - The same shape and data type as `accum`.
@ -4468,7 +4468,7 @@ class ApplyAddSign(PrimitiveWithInfer):
- **grad** (Tensor) - A tensor of the same type as `var`, for the gradient.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **m** (Tensor) - The same shape and data type as `m`.
@ -4576,7 +4576,7 @@ class ApplyPowerSign(PrimitiveWithInfer):
- **grad** (Tensor) - A tensor of the same type as `var`, for the gradient.
Outputs:
Tuple of 2 Tensor, the updated parameters.
Tuple of 2 Tensors, the updated parameters.
- **var** (Tensor) - The same shape and data type as `var`.
- **m** (Tensor) - The same shape and data type as `m`.
@ -5162,7 +5162,7 @@ class ConfusionMulGrad(PrimitiveWithInfer):
Default:(), reduce all dimensions. Only constant value is allowed.
keep_dims (bool):
- If true, keep these reduced dimensions and the length as 1.
- If false, don't keep these dimensions. Default:False.
- If false, don't keep these dimensions. Default: False.
Inputs:
- **input_0** (Tensor) - The input Tensor.
@ -5173,11 +5173,11 @@ class ConfusionMulGrad(PrimitiveWithInfer):
- **output_0** (Tensor) - The same shape as `input0`.
- **output_1** (Tensor)
- If axis is (), and keep_dims is false, the output is a 0-D array representing
- If axis is (), and keep_dims is False, the output is a 0-D array representing
the sum of all elements in the input array.
- If axis is int, set as 2, and keep_dims is false,
- If axis is int, set as 2, and keep_dims is False,
the shape of output is :math:`(x_1,x_3,...,x_R)`.
- If axis is tuple(int), set as (2,3), and keep_dims is false,
- If axis is tuple(int), set as (2,3), and keep_dims is False,
the shape of output is :math:`(x_1,x_4,...x_R)`.
Examples:

2
mindspore/ops/operations/other_ops.py
Unescape View File

@ -490,7 +490,7 @@ class PopulationCount(PrimitiveWithInfer):
- **input** (Tensor) - The data type should be int16 or uint16.
Outputs:
Tensor, with shape same as the input.
Tensor, with the sam shape as the input.
Examples:
>>> population_count = P.PopulationCount()

25
mindspore/ops/operations/random_ops.py
Unescape View File

@ -185,11 +185,11 @@ class Poisson(PrimitiveWithInfer):
Inputs:
- **shape** (tuple) - The shape of random tensor to be generated. Only constant value is allowed.
- **mean** (Tensor) - μ parameter the distribution was constructed with.
The parameter defines mean number of occurrences of the event. With float32 data type.
- **mean** (Tensor) - μ, parameter which the distribution was constructed with.
The parameter defines the mean number of occurrences of the event, with float32 data type.
Outputs:
Tensor. The shape should be the broadcasted shape of Input "shape" and shape of mean.
Tensor. Its shape should be the broadcasted shape of `shape` and the shape of `mean`.
The dtype is int32.
Examples:
@ -325,9 +325,10 @@ class UniformReal(PrimitiveWithInfer):
class RandomChoiceWithMask(PrimitiveWithInfer):
"""
Generates a random samply as index tensor with a mask tensor from a given tensor.
Generates a random sample as index tensor with a mask tensor from a given tensor.
The input must be a tensor of rank >= 1. If its rank >= 2, the first dimension specify the number of sample.
The input must be a tensor of rank not less than 1. If its rank is greater than or equal to 2,
the first dimension specifies the number of samples.
The index tensor and the mask tensor have the fixed shapes. The index tensor denotes the index of the nonzero
sample, while the mask tensor denotes which elements in the index tensor are valid.
@ -337,7 +338,8 @@ class RandomChoiceWithMask(PrimitiveWithInfer):
seed2 (int): Random seed2. Default: 0.
Inputs:
- **input_x** (Tensor[bool]) - The input tensor. The input tensor rank should be >= 1 and <= 5.
- **input_x** (Tensor[bool]) - The input tensor.
The input tensor rank should be greater than or equal to 1 and less than or equal to 5.
Outputs:
Two tensors, the first one is the index tensor and the other one is the mask tensor.
@ -374,8 +376,8 @@ class RandomCategorical(PrimitiveWithInfer):
Generates random samples from a given categorical distribution tensor.
Args:
dtype (mindspore.dtype): The type of output. Its value should be one of [mindspore.int16,
mindspore.int32, mindspore.int64]. Default: mindspore.int64.
dtype (mindspore.dtype): The type of output. Its value should be one of mindspore.int16,
mindspore.int32 and mindspore.int64. Default: mindspore.int64.
Inputs:
- **logits** (Tensor) - The input tensor. 2-D Tensor with shape [batch_size, num_classes].
@ -437,16 +439,17 @@ class Multinomial(PrimitiveWithInfer):
The rows of input do not need to sum to one (in which case we use the values as weights),
but must be non-negative, finite and have a non-zero sum.
Args:
seed (int): Seed data is used as entropy source for Random number engines generating pseudo-random numbers.
seed (int): Seed data is used as entropy source for Random number engines to generate pseudo-random numbers.
Must be non-negative. Default: 0.
replacement(bool): Whether to draw with replacement or not.
Inputs:
- **input** (Tensor[float32]) - the input tensor containing the cumsum of probabilities, must be 1 or 2 dims.
- **input** (Tensor[float32]) - the input tensor containing the cumsum of probabilities, must be 1 or 2
dimensions.
- **num_samples** (int32) - number of samples to draw.
Outputs:
Tensor. have the same rows with input, each row has num_samples sampled indices.
Tensor with the same rows as input, each row has num_samples sampled indices.
Examples:
>>> input = Tensor([0., 9., 4., 0.], mstype.float32)

2
mindspore/train/serialization.py
Unescape View File

@ -149,7 +149,7 @@ def save_checkpoint(save_obj, ckpt_file_name, integrated_save=True, async_save=F
save_obj (nn.Cell or list): The cell object or parameters list(each element is a dictionary,
like {"name": param_name, "data": param_data}.)
ckpt_file_name (str): Checkpoint file name. If the file name already exists, it will be overwritten.
integrated_save (bool): Whether to integrated save in automatic model parallel scene.
integrated_save (bool): Whether to integrated save in automatic model parallel scene. Default: True
async_save (bool): Whether asynchronous execution saves the checkpoint to a file. Default: False
Raises:

Write Preview
Loading…
Cancel
Save