Min positive value for each Tensor type?

1

I’d like to avoid NaNs in division by evaluating (x + eps) / (y + eps) for a type-dependent number eps. Does torch provide an equivalent to C’s DBL_MIN/FLT_MIN or C++'s std::numeric_limits<T>::min or Numpy’s numpy.finfo(...).tiny? It would ideally be a method of each Tensor class like Tensor.finfo.tiny.

1 Like
2

PyTorch doesn’t have such functionality yet, but we use standard floating point tensors. You can use np methods to get the minimum, or easily compute yourself.

3

I just created a PR to get the numpy.dtype for torch.Tensors: https://github.com/pytorch/pytorch/pull/4256

1 Like