masking

• namespace: Rindow\NeuralNetworks\Gradient\Func
• classname: Masking

Differentiable masking function.

Applies a boolean mask to the data. The mask itself is not differentiable.

Methods

masking

$g->masking(
    Variable|NDArray $mask,
    Variable|NDArray $data,
    ?int $batchDims=null,
    ?int $axis=null,
    ?float $fill=null,
    ?int $mode=null,
) : Variable

Create and execute the function in the builder method

Arguments

• mask: A boolean Variable or NDArray. Implicitly create Variable for NDArray. Gradients are not propagated to the mask.
• data: A Variable or NDArray. Implicitly create Variable for NDArray.

Options

• batchDims: Number of batch dimensions.
• axis: Axis to apply the mask.
• fill: Fill value for masked positions.
• mode: Masking mode.

use Rindow\Math\Matrix\MatrixOperator;
use Rindow\NeuralNetworks\Builder\NeuralNetworks;
$mo = new MatrixOperator();
$nn = new NeuralNetworks($mo);
$g = $nn->gradient();
$mask = $g->Variable([[true,false],[false,true]]);
$data = $g->Variable([[1,2],[3,4]]);
$c = $nn->with($tape=$g->GradientTape(),function() use ($g,$mask,$data) {
    return $g->masking($mask,$data,fill:0);
});
$dData = $tape->gradient($c,$data);
echo $mo->toString($c)."\n";
echo $mo->toString($dData)."\n";