mindspore.ops.ScatterMul
- class mindspore.ops.ScatterMul(use_locking=False)[源代码]
根据指定更新值和输入索引通过乘法运算更新输入数据的值。
对于 indices.shape 的每个 i, …, j :
\[\text{input_x}[\text{indices}[i, ..., j], :] \mathrel{*}= \text{updates}[i, ..., j, :]\]输入的 input_x 和 updates 遵循隐式类型转换规则,以确保数据类型一致。如果数据类型不同,则低精度数据类型将转换为高精度的数据类型。当参数的数据类型需要转换时,则会抛出RuntimeError异常。
- 参数:
use_locking (bool) - 是否启用锁保护。默认值:
False
。
- 输入:
input_x (Union[Parameter, Tensor]) - ScatterMul的输入,任意维度的Parameter或Tensor。shape: \((N, *)\) ,其中 \(*\) 表示任意数量的附加维度。
indices (Tensor) - 指定相乘操作的索引,数据类型必须为mindspore.int32。
updates (Tensor) - 指定与 input_x 相乘的Tensor,数据类型与 input_x 相同,shape为 indices.shape + x.shape[1:] 。
- 输出:
Tensor,更新后的 input_x ,shape和类型与 input_x 相同。
- 异常:
TypeError - use_locking 不是bool。
TypeError - indices 不是int32。
ValueError - updates 的shape不等于 indices.shape + x.shape[1:] 。
RuntimeError - 当 input_x 和 updates 类型不一致,需要进行类型转换时,如果 updates 不支持转成 input_x 需要的数据类型,就会报错。
- 支持平台:
Ascend
GPU
CPU
样例:
>>> import numpy as np >>> from mindspore import dtype as mstype >>> from mindspore import Tensor, ops, Parameter >>> input_x = Parameter(Tensor(np.array([[1.0, 1.0, 1.0], [2.0, 2.0, 2.0]]), mstype.float32), name="x") >>> indices = Tensor(np.array([0, 1]), mstype.int32) >>> updates = Tensor(np.array([[2.0, 2.0, 2.0], [2.0, 2.0, 2.0]]), mstype.float32) >>> scatter_mul = ops.ScatterMul() >>> output = scatter_mul(input_x, indices, updates) >>> print(output) [[2. 2. 2.] [4. 4. 4.]] >>> # for input_x will be updated after the operation is completed. input_x need to be re-initialized. >>> input_x = Parameter(Tensor(np.array([[1.0, 1.0, 1.0], [2.0, 2.0, 2.0]]), mstype.float32), name="x") >>> # for indices = [[0, 1], [1, 1]] >>> # step 1: [0, 1] >>> # input_x[0] = [1.0, 1.0, 1.0] * [1.0, 1.0, 1.0] = [1.0, 1.0, 1.0] >>> # input_x[1] = [2.0, 2.0, 2.0] * [3.0, 3.0, 3.0] = [6.0, 6.0, 6.0] >>> # step 2: [1, 1] >>> # input_x[1] = [6.0, 6.0, 6.0] * [7.0, 7.0, 7.0] = [42.0, 42.0, 42.0] >>> # input_x[1] = [42.0, 42.0, 42.0] * [9.0, 9.0, 9.0] = [378.0, 378.0, 378.0] >>> indices = Tensor(np.array([[0, 1], [1, 1]]), mstype.int32) >>> updates = Tensor(np.array([[[1.0, 1.0, 1.0], [3.0, 3.0, 3.0]], ... [[7.0, 7.0, 7.0], [9.0, 9.0, 9.0]]]), mstype.float32) >>> scatter_mul = ops.ScatterMul() >>> output = scatter_mul(input_x, indices, updates) >>> print(output) [[ 1. 1. 1.] [378. 378. 378.]] >>> # for input_x will be updated after the operation is completed. input_x need to be re-initialized. >>> input_x = Parameter(Tensor(np.array([[1.0, 1.0, 1.0], [2.0, 2.0, 2.0]]), mstype.float32), name="x") >>> # for indices = [[1, 0], [1, 1]] >>> # step 1: [1, 0] >>> # input_x[0] = [1.0, 1.0, 1.0] * [3.0, 3.0, 3.0] = [3.0, 3.0, 3.0] >>> # input_x[1] = [2.0, 2.0, 2.0] * [1.0, 1.0, 1.0] = [2.0, 2.0, 2.0] >>> # step 2: [1, 1] >>> # input_x[1] = [2.0, 2.0, 2.0] * [7.0, 7.0, 7.0] = [14.0, 14.0, 14.0] >>> # input_x[1] = [14.0, 14.0, 14.0] * [9.0, 9.0, 9.0] = [126.0, 126.0, 126.0] >>> indices = Tensor(np.array([[1, 0], [1, 1]]), mstype.int32) >>> updates = Tensor(np.array([[[1.0, 1.0, 1.0], [3.0, 3.0, 3.0]], ... [[7.0, 7.0, 7.0], [9.0, 9.0, 9.0]]]), mstype.float32) >>> scatter_mul = ops.ScatterMul() >>> output = scatter_mul(input_x, indices, updates) >>> print(output) [[ 3. 3. 3.] [126. 126. 126.]] >>> # for input_x will be updated after the operation is completed. input_x need to be re-initialized. >>> input_x = Parameter(Tensor(np.array([[1.0, 1.0, 1.0], [2.0, 2.0, 2.0]]), mstype.float32), name="x") >>> # for indices = [[0, 1], [0, 1]] >>> # step 1: [0, 1] >>> # input_x[0] = [1.0, 1.0, 1.0] * [1.0, 1.0, 1.0] = [1.0, 1.0, 1.0] >>> # input_x[1] = [2.0, 2.0, 2.0] * [3.0, 3.0, 3.0] = [6.0, 6.0, 6.0] >>> # step 2: [0, 1] >>> # input_x[0] = [1.0, 1.0, 1.0] * [7.0, 7.0, 7.0] = [7.0, 7.0, 7.0] >>> # input_x[1] = [6.0, 6.0, 6.0] * [9.0, 9.0, 9.0] = [54.0, 54.0, 54.0] >>> indices = Tensor(np.array([[0, 1], [0, 1]]), mstype.int32) >>> updates = Tensor(np.array([[[1.0, 1.0, 1.0], [3.0, 3.0, 3.0]], ... [[7.0, 7.0, 7.0], [9.0, 9.0, 9.0]]]), mstype.float32) >>> scatter_mul = ops.ScatterMul() >>> output = scatter_mul(input_x, indices, updates) >>> print(output) [[ 7. 7. 7.] [54. 54. 54.]]