Orthogonal initialization — nn_init_orthogonal

Fills the input Tensor with a (semi) orthogonal matrix, as described in Exact solutions to the nonlinear dynamics of learning in deep linear neural networks - Saxe, A. et al. (2013). The input tensor must have at least 2 dimensions, and for tensors with more than 2 dimensions the trailing dimensions are flattened.

Usage

nn_init_orthogonal_(tensor, gain = 1)

Arguments

tensor: an n-dimensional Tensor
gain: optional scaling factor

Examples

if (torch_is_installed()) {
w <- torch_empty(3, 5)
nn_init_orthogonal_(w)
}
#> torch_tensor
#>  0.7160  0.3257  0.3196  0.5157 -0.1145
#>  0.0487 -0.4434  0.6289 -0.3018 -0.5607
#> -0.0492 -0.6043  0.3151  0.3915  0.6163
#> [ CPUFloatType{3,5} ]