torch.isclose#

torch.isclose(input,other,rtol=1e-05,atol=1e-08,equal_nan=False)→Tensor#

Returns a new tensor with boolean elements representing if each element ofinput is “close” to the corresponding element ofother.Closeness is defined as:

$$\mid {\text{input}}_i-{\text{other}}_i\mid \le \mathtt{\text{rtol}}\times \mid {\text{other}}_i\mid +\mathtt{\text{atol}}$$∣inputi​−otheri​∣≤rtol×∣otheri​∣+atol

whereinput andother are finite. Whereinputand/orother are nonfinite they are close if and only ifthey are equal, with NaNs being considered equal to each other whenequal_nan is True.

Parameters

• input (Tensor) – first tensor to compare

• other (Tensor) – second tensor to compare

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

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

• equal_nan (bool,optional) – ifTrue, then twoNaN s will be considered equal. Default:False

Examples:

>>>torch.isclose(torch.tensor((1.,2,3)),torch.tensor((1+1e-10,3,4)))tensor([ True, False, False])>>>torch.isclose(torch.tensor((float('inf'),4)),torch.tensor((float('inf'),6)),rtol=.5)tensor([True, True])