oneflow.allclose

oneflow.allclose(input, other, rtol=1e-05, atol=1e-08, equal_nan=False)bool

The documentation is referenced from: https://pytorch.org/docs/1.10/generated/torch.allclose.html

This function checks if input and other satisfy the condition:

\[\lvert \text{input} - \text{other} \rvert \leq \texttt{atol} + \texttt{rtol} \times \lvert \text{other} \rvert\]

elementwise, for all elements of input and other. The behaviour of this function is analogous to numpy.allclose

Parameters
  • input (oneflow.Tensor) – first tensor to compare

  • other (oneflow.Tensor) – second tensor to compare

  • atol (float, optional) – absolute tolerance. Default: 1e-08

  • rtol (float, optional) – relative tolerance. Default: 1e-05

  • equal_nan (bool, optional) – if True, then two NaN s will be considered equal. Default: False

Returns

A Tensor with bool type.

Return type

oneflow.Tensor

For example:

>>> import oneflow as flow

>>> flow.allclose(flow.tensor([10000., 1e-07]), flow.tensor([10000.1, 1e-08]))
False
>>> flow.allclose(flow.tensor([10000., 1e-08]), flow.tensor([10000.1, 1e-09]))
True
>>> flow.allclose(flow.tensor([1.0, float('nan')]), flow.tensor([1.0, float('nan')]))
False
>>> flow.allclose(flow.tensor([1.0, float('nan')]), flow.tensor([1.0, float('nan')]), equal_nan=True)
True