Xavier normal initialization — nn_init_xavier_normal

Fills the input Tensor with values according to the method described in Understanding the difficulty of training deep feedforward neural networks - Glorot, X. & Bengio, Y. (2010), using a normal distribution.

Usage

nn_init_xavier_normal_(tensor, gain = 1)

Arguments

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

Examples

if (torch_is_installed()) {
w <- torch_empty(3, 5)
nn_init_xavier_normal_(w)
}
#> torch_tensor
#>  0.2242 -1.1448  0.4269  0.3308 -0.4103
#> -0.2068 -0.5633 -0.2517  0.5765  0.2002
#> -0.6237 -0.5038 -0.0536 -0.6221 -0.2528
#> [ CPUFloatType{3,5} ]