Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

rename Tensor.tensor_scatter_* to Tensor.scatter_*

This commit is contained in:
yuchaojie 2022-06-13 20:18:03 +08:00
parent 065901f416
commit d65187f1ba
5 changed files with 24 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
docs/api/api_python/mindspore/mindspore.Tensor.rst
View File

@ -1475,7 +1475,7 @@ mindspore.Tensor
Tensor。 Tensor。
.. py:method:: tensor_scatter_add(indices, updates) .. py:method:: scatter_add(indices, updates)
根据指定的更新值和输入索引，通过相加运算更新本Tensor的值。当同一索引有不同值时，更新的结果将是所有值的总和。 根据指定的更新值和输入索引，通过相加运算更新本Tensor的值。当同一索引有不同值时，更新的结果将是所有值的总和。
@ -1496,7 +1496,7 @@ mindspore.Tensor
- **TypeError** - `indices` 的数据类型既不是int32，也不是int64。 - **TypeError** - `indices` 的数据类型既不是int32，也不是int64。
- **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。 - **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。
.. py:method:: tensor_scatter_div(indices, updates) .. py:method:: scatter_div(indices, updates)
根据指定的索引, 通过除法进行计算, 将输出赋值到输出Tensor中。 根据指定的索引, 通过除法进行计算, 将输出赋值到输出Tensor中。
@ -1518,7 +1518,7 @@ mindspore.Tensor
- **TypeError** - `indices` 的数据类型不是int32，也不是int64。 - **TypeError** - `indices` 的数据类型不是int32，也不是int64。
- **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。 - **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。
.. py:method:: tensor_scatter_min(indices, updates) .. py:method:: scatter_min(indices, updates)
根据指定的更新值和输入索引，通过最小值运算，将结果赋值到输出Tensor中。 根据指定的更新值和输入索引，通过最小值运算，将结果赋值到输出Tensor中。
@ -1541,7 +1541,7 @@ mindspore.Tensor
- **TypeError** - `indices` 的数据类型既不是int32，也不是int64。 - **TypeError** - `indices` 的数据类型既不是int32，也不是int64。
- **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。 - **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。
.. py:method:: tensor_scatter_mul(indices, updates) .. py:method:: scatter_mul(indices, updates)
根据指定的索引， 通过乘法进行计算， 将输出赋值到输出Tensor中。 根据指定的索引， 通过乘法进行计算， 将输出赋值到输出Tensor中。
@ -1562,7 +1562,7 @@ mindspore.Tensor
- **TypeError** - `indices` 的数据类型不是int32，也不是int64。 - **TypeError** - `indices` 的数据类型不是int32，也不是int64。
- **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。 - **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。
.. py:method:: tensor_scatter_sub(indices, updates) .. py:method:: scatter_sub(indices, updates)
根据指定的更新值和输入索引，通过减法进行运算，将结果赋值到输出Tensor中。当同一索引有不同值时，更新的结果将是所有值的总和。此操作几乎等同于使用 :class:`mindspore.ops.ScatterNdSub` ，只是更新后的结果是通过算子output返回，而不是直接原地更新input。 根据指定的更新值和输入索引，通过减法进行运算，将结果赋值到输出Tensor中。当同一索引有不同值时，更新的结果将是所有值的总和。此操作几乎等同于使用 :class:`mindspore.ops.ScatterNdSub` ，只是更新后的结果是通过算子output返回，而不是直接原地更新input。

12
mindspore/ccsrc/pipeline/jit/resource.cc
View File

@ -112,7 +112,7 @@ BuiltInTypeMap &GetMethodMap() {
{"__lt__", prim::kPrimScalarLt}, // P.scalar_lt, {"__lt__", prim::kPrimScalarLt}, // P.scalar_lt,
{"__gt__", prim::kPrimScalarGt}, // P.scalar_gt, {"__gt__", prim::kPrimScalarGt}, // P.scalar_gt,
{"__le__", prim::kPrimScalarLe}, // P.scalar_le, {"__le__", prim::kPrimScalarLe}, // P.scalar_le,
{"__ge__", prim::kPrimScalarGe}, // P.scalar_ge,mindspore/ccsrc/pipeline/jit/resource.cc {"__ge__", prim::kPrimScalarGe}, // P.scalar_ge,
{"__bool__", std::string("float_bool")}, // C.float_bool {"__bool__", std::string("float_bool")}, // C.float_bool
{"__ms_to_array__", prim::kPrimScalarToArray}, // P.scalar_to_array, {"__ms_to_array__", prim::kPrimScalarToArray}, // P.scalar_to_array,
}}, }},
@ -234,11 +234,11 @@ BuiltInTypeMap &GetMethodMap() {
{"padding", std::string("padding")}, // padding() {"padding", std::string("padding")}, // padding()
{"searchsorted", std::string("searchsorted")}, // P.Select() {"searchsorted", std::string("searchsorted")}, // P.Select()
{"take", std::string("take")}, // P.GatherNd() {"take", std::string("take")}, // P.GatherNd()
{"tensor_scatter_add", std::string("tensor_scatter_add")}, // P.TensorScatterAdd() {"scatter_add", std::string("tensor_scatter_add")}, // P.TensorScatterAdd()
{"tensor_scatter_mul", std::string("tensor_scatter_mul")}, // tensor_scatter_mul() {"scatter_mul", std::string("tensor_scatter_mul")}, // tensor_scatter_mul()
{"tensor_scatter_sub", std::string("tensor_scatter_sub")}, // P.TensorScatterSub() {"scatter_sub", std::string("tensor_scatter_sub")}, // P.TensorScatterSub()
{"tensor_scatter_min", std::string("tensor_scatter_min")}, // P.TensorScatterMin() {"scatter_min", std::string("tensor_scatter_min")}, // P.TensorScatterMin()
{"tensor_scatter_div", std::string("tensor_scatter_div")}, // P.TensorScatterDiv() {"scatter_div", std::string("tensor_scatter_div")}, // P.TensorScatterDiv()
{"lp_norm", std::string("lp_norm")}, // lp_norm() {"lp_norm", std::string("lp_norm")}, // lp_norm()
{"renorm", std::string("renorm")}, // renorm() {"renorm", std::string("renorm")}, // renorm()
{"trace", std::string("trace")}, // P.Eye() {"trace", std::string("trace")}, // P.Eye()

20
mindspore/python/mindspore/common/tensor.py
View File

@ -756,7 +756,7 @@ class Tensor(Tensor_):
self._init_check() self._init_check()
return tensor_operator_registry.get('bitwise_xor')(self, x) return tensor_operator_registry.get('bitwise_xor')(self, x)
def tensor_scatter_mul(self, indices, updates): def scatter_mul(self, indices, updates):
""" """
Creates a new tensor by multiplying the values from the positions in self tensor indicated by Creates a new tensor by multiplying the values from the positions in self tensor indicated by
`indices`, with values from `updates`. When divided values are provided for the same `indices`, with values from `updates`. When divided values are provided for the same
@ -802,7 +802,7 @@ class Tensor(Tensor_):
>>> # first_input_x = input_x[0][0] * updates[0] = [[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]] >>> # first_input_x = input_x[0][0] * updates[0] = [[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]
>>> # 6, Perform the multiply operation for the second time: >>> # 6, Perform the multiply operation for the second time:
>>> # second_input_x = input_x[0][0] * updates[1] = [[-0.22, 0.3, 3.6], [0.4, 0.5, -3.2]] >>> # second_input_x = input_x[0][0] * updates[1] = [[-0.22, 0.3, 3.6], [0.4, 0.5, -3.2]]
>>> output = input_x.tensor_scatter_mul(indices, updates) >>> output = input_x.scatter_mul(indices, updates)
>>> print(output) >>> print(output)
[[-0.22 0.3 3.6 ] [[-0.22 0.3 3.6 ]
[ 0.4 0.5 -3.2 ]] [ 0.4 0.5 -3.2 ]]
@ -810,7 +810,7 @@ class Tensor(Tensor_):
self._init_check() self._init_check()
return tensor_operator_registry.get('tensor_scatter_mul')(self, indices, updates) return tensor_operator_registry.get('tensor_scatter_mul')(self, indices, updates)
def tensor_scatter_div(self, indices, updates): def scatter_div(self, indices, updates):
""" """
Creates a new tensor by dividing the values from the positions in self tensor indicated by Creates a new tensor by dividing the values from the positions in self tensor indicated by
`indices`, with values from `updates`. When divided values are provided for the same `indices`, with values from `updates`. When divided values are provided for the same
@ -850,7 +850,7 @@ class Tensor(Tensor_):
>>> input_x = Tensor(np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]]).astype('float32')) >>> input_x = Tensor(np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]]).astype('float32'))
>>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32')) >>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32'))
>>> updates = Tensor(np.array([1.0, 2.0]).astype('float32')) >>> updates = Tensor(np.array([1.0, 2.0]).astype('float32'))
>>> output = input_x.tensor_scatter_div(indices, updates) >>> output = input_x.scatter_div(indices, updates)
>>> print(output) >>> print(output)
[[-0.05, 0.3, 3.6 ] [[-0.05, 0.3, 3.6 ]
[ 0.4, 0.5, -3.2 ]] [ 0.4, 0.5, -3.2 ]]
@ -2389,7 +2389,7 @@ class Tensor(Tensor_):
return reduce_(self, reduce_min(keepdims), cmp_fn=minimum(), axis=axis, keepdims=keepdims, return reduce_(self, reduce_min(keepdims), cmp_fn=minimum(), axis=axis, keepdims=keepdims,
initial=initial, where=where) initial=initial, where=where)
def tensor_scatter_add(self, indices, updates): def scatter_add(self, indices, updates):
""" """
Creates a new tensor by adding the values from the positions in self tensor indicated by Creates a new tensor by adding the values from the positions in self tensor indicated by
`indices`, with values from `updates`. When multiple values are given for the same `indices`, with values from `updates`. When multiple values are given for the same
@ -2427,7 +2427,7 @@ class Tensor(Tensor_):
>>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32')) >>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32'))
>>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32')) >>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32'))
>>> updates = Tensor(np.array([1.0, 2.2]).astype('float32')) >>> updates = Tensor(np.array([1.0, 2.2]).astype('float32'))
>>> output = x.tensor_scatter_add(indices, updates) >>> output = x.scatter_add(indices, updates)
>>> print(output) >>> print(output)
[[ 3.1 0.3 3.6] [[ 3.1 0.3 3.6]
[ 0.4 0.5 -3.2]] [ 0.4 0.5 -3.2]]
@ -2435,7 +2435,7 @@ class Tensor(Tensor_):
self._init_check() self._init_check()
return tensor_operator_registry.get("tensor_scatter_add")()(self, indices, updates) return tensor_operator_registry.get("tensor_scatter_add")()(self, indices, updates)
def tensor_scatter_sub(self, indices, updates): def scatter_sub(self, indices, updates):
""" """
Creates a new tensor by subtracting the values from the positions in self tensor indicated by Creates a new tensor by subtracting the values from the positions in self tensor indicated by
`indices`, with values from `updates`. When multiple values are provided for the same `indices`, with values from `updates`. When multiple values are provided for the same
@ -2473,7 +2473,7 @@ class Tensor(Tensor_):
>>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32')) >>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32'))
>>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32')) >>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32'))
>>> updates = Tensor(np.array([1.0, 2.2]).astype('float32')) >>> updates = Tensor(np.array([1.0, 2.2]).astype('float32'))
>>> output = x.tensor_scatter_sub(indices, updates) >>> output = x.scatter_sub(indices, updates)
>>> print(output) >>> print(output)
[[-3.3000002 0.3 3.6 ] [[-3.3000002 0.3 3.6 ]
[ 0.4 0.5 -3.2 ]] [ 0.4 0.5 -3.2 ]]
@ -2481,7 +2481,7 @@ class Tensor(Tensor_):
self._init_check() self._init_check()
return tensor_operator_registry.get('tensor_scatter_sub')()(self, indices, updates) return tensor_operator_registry.get('tensor_scatter_sub')()(self, indices, updates)
def tensor_scatter_min(self, indices, updates): def scatter_min(self, indices, updates):
""" """
By comparing the value at the position indicated by `indices` in self tensor with the value in the `updates`, By comparing the value at the position indicated by `indices` in self tensor with the value in the `updates`,
the value at the index will eventually be equal to the smallest one to create a new tensor. the value at the index will eventually be equal to the smallest one to create a new tensor.
@ -2516,7 +2516,7 @@ class Tensor(Tensor_):
>>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32')) >>> x = Tensor(np.array([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]]).astype('float32'))
>>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32')) >>> indices = Tensor(np.array([[0, 0], [0, 0]]).astype('int32'))
>>> updates = Tensor(np.array([1.0, 2.2]).astype('float32')) >>> updates = Tensor(np.array([1.0, 2.2]).astype('float32'))
>>> output = x.tensor_scatter_min(indices, updates) >>> output = x.scatter_min(indices, updates)
>>> print(output) >>> print(output)
[[ -0.1 0.3 3.6] [[ -0.1 0.3 3.6]
[ 0.4 0.5 -3.2]] [ 0.4 0.5 -3.2]]

4
tests/st/ops/ascend/test_tensor_scatter_arithmetic.py
View File

@ -197,9 +197,9 @@ def test_tensor_scatter_arithmetic_tensor_op(func, data_type, index_type):
expected = tensor_scatter_np(func, input_x, indices, updates) expected = tensor_scatter_np(func, input_x, indices, updates)
if func == 'add': if func == 'add':
output = input_x.tensor_scatter_add(indices, updates) output = input_x.scatter_add(indices, updates)
elif func == 'sub': elif func == 'sub':
output = input_x.tensor_scatter_sub(indices, updates) output = input_x.scatter_sub(indices, updates)
np.testing.assert_allclose(output.asnumpy(), expected, rtol=1e-6) np.testing.assert_allclose(output.asnumpy(), expected, rtol=1e-6)

2
tests/st/ops/cpu/test_tensor_scatter_arithmetic.py
View File

@ -266,7 +266,7 @@ def test_tensor_scatter_arithmetic_tensor_func_check(func, data_type, index_type
expected = tensor_scatter_np(func, input_x, indices, updates) expected = tensor_scatter_np(func, input_x, indices, updates)
if func == 'div': if func == 'div':
output = input_x.tensor_scatter_div(indices, updates) output = input_x.scatter_div(indices, updates)
np.testing.assert_allclose(output.asnumpy(), expected, rtol=1e-6) np.testing.assert_allclose(output.asnumpy(), expected, rtol=1e-6)