mindspore.ops.approximate_equal

mindspore.ops.approximate_equal(x, y, tolerance=1e-5)

Return a boolean tensor where two tensors are element-wise equal within a tolerance.

Support implicit type conversion and type promotion.

Math function is defined as:

$$\begin{array}{r}ou{t}_i=\{\begin{array}{ll}& \text{ if }|x_i-y_i|<\text{tolerance},True\\ & \text{ if }|x_i-y_i|\ge \text{tolerance},False\end{array}\end{array}$$

Two infinite values and two NaN values are not considered equal.

Parameters

• x (Tensor) – The first input tensor.

• y (Tensor) – The second input tensor.

• tolerance (float) – The maximum deviation that two elements can be considered equal. Default 1e-5 .

Returns

Tensor

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> mindspore.ops.approximate_equal(mindspore.tensor([1e6, 2e6, float("inf"), float("-inf"), float("nan")]),
...                                 mindspore.tensor([1e6, 2e7, float("inf"), float("-inf"), float("nan")]))
Tensor(shape=[6], dtype=Bool, value= [ True, False, False, False, False])
>>>
>>> mindspore.ops.approximate_equal(mindspore.tensor([1e6, 2e6, 3e6]),
...                                 mindspore.tensor([1.00001e6, 2.00002e6, 3.00009e6]), tolerance=1e3)
Tensor(shape=[3], dtype=Bool, value= [ True,  True,  True])
>>>
>>> # If `x` and `y` have different datatypes, the lower precision data type will be converted to the
    relatively highest precision data type.
>>> mindspore.ops.approximate_equal(mindspore.tensor([1, 2], mindspore.int32),
...                                 mindspore.tensor([1., 2], mindspore.float32))
Tensor(shape=[2], dtype=Bool, value= [ True,  True])