pypose.optim.kernel.Tolerant¶
- class pypose.optim.kernel.Tolerant(a=1.0, b=-1.0)[source]¶
The robust Tolerant kernel cost function.
\[\bm{y}_i = b\log (1+e^{\frac{\bm{x}_i-a}{b}})-b\log (1+e^{\frac{-a}{b}}) \]where \(\bm{a}\) and \(\bm{b}\) are hyperparameters, \(\bm{x}\) and \(\bm{y}\) are the input and output tensors, respectively.
- Parameters
a (float) – Specify the Tolerant cost. The value must be positive. Default: 1.0
b (float) – Specify the Tolerant cost. The value must be negative. Default: -1.0
Note
The input has to be a non-negative tensor and the output tensor has the same shape with the input.
Example
>>> import pypose.optim.kernel as ppok >>> kernel = ppok.Tolerant() >>> input = torch.tensor([0, 0.5, 1, 2, 3]) >>> kernel(input) tensor([0.0000, 0.4636, 0.7854, 1.1071, 1.2490])
