Neural Architecture Search (NAS) recurrent network cell.
layer_nas_cell(
object,
units,
projection = NULL,
use_bias = FALSE,
kernel_initializer = "glorot_uniform",
recurrent_initializer = "glorot_uniform",
projection_initializer = "glorot_uniform",
bias_initializer = "zeros",
...
)Model or layer object
int, The number of units in the NAS cell.
(optional) int, The output dimensionality for the projection matrices. If None, no projection is performed.
(optional) bool, If `TRUE` then use biases within the cell. This is `FALSE` by default.
Initializer for kernel weight.
Initializer for recurrent kernel weight.
Initializer for projection weight, used when projection is not `NULL`.
Initializer for bias, used when `use_bias` is `TRUE`.
Additional keyword arguments.
A tensor
This implements the recurrent cell from the paper: https://arxiv.org/abs/1611.01578 Barret Zoph and Quoc V. Le. "Neural Architecture Search with Reinforcement Learning" Proc. ICLR 2017. The class uses an optional projection layer.