r""" Computes the smooth L1 loss. The smooth L1 loss is a variant of the L1 loss which replaces the absolute difference with a squared di

r"""Applies the Softmin function. Applies :math:`\frac{e^{-x_i}}{\sum_j e^{-x_j}}` element-wise.

r"""Applies the Softsign function. Applies :math:`\frac{x}{1 + |x|}` element wise.

r"""An initializer that returns a sparse matrix. Args: sparsity (float): The fraction of elements in each column to be set to zer

The optimizer's state dictionary.