Loading TOC...

cntk.sequenceUnpack

cntk.sequenceUnpack(
   $x as cntk.variable,
   $padding-value as cntk.variable,
   $supress-mask-output as cntk.variable,
   $name as String
) as cntk.function

Summary

Create an instance of the CNTK built-in operator for unpacking the specified sequence operand along the most significant static axis [-1] and padding any gaps with the specified padding value. If supressMaskOutput is false, the returned Function has 2 outputs; viz. the unpacked non-sequence data and a mask denoting the gaps in the unpacked output due to differences across lengths of the sequences in the operand.

Parameters
$x The sequence tensor (or its name) which is unpacked.
$padding-value The value to pad gaps in the unpacked tensor with.
$supress-mask-output Whether the Function has a mask tensor output denoting the gaps in the unpacked output due to differences across lengths of the sequences in the operand.
$name The name of the function instance in the network.

Example

  var inputVariable1 = cntk.inputVariable(cntk.shape([2]), "float",
    fn.false(), fn.false(), "feature")
  cntk.sequenceUnpack(inputVariable1,0.0,fn.true())
  => cntk.function(Composite UnpackSequenceOp (Input(Name(feature), Shape([2]),
  Dynamic Axes([Sequence Axis(Default Dynamic Axis), Batch Axis(Default Batch
  Axis)]))))

Stack Overflow iconStack Overflow: Get the most useful answers to questions from the MarkLogic community, or ask your own question.