lamp-core/lamp.autograd/Conv2D

Conv2D

case class Conv2D(scope: Scope, input: Variable, weight: Variable, bias: Variable, stride: Long, padding: Long, dilation: Long, groups: Long) extends Op

2D convolution

Value parameters:
bias

out_channels

input

batch x in_channels x height x width

weight

out_channels x in_channels x kernel_size x kernel_size

Returns:

Variable with Tensor of size batch x out_channels x L' (length depends on stride/padding/dilation)

trait Serializable
trait Product
trait Equals
trait Op
class Object
trait Matchable
class Any

Value members

Inherited methods

def productElementNames: Iterator[String]
Inherited from:
Product
def productIterator: Iterator[Any]
Inherited from:
Product

Concrete fields

val batchSize: Long
val imageHeight: Long
val imageWidth: Long
val inputChannels: Long
val kernelSize: Long
val outChannels: Long
override val params: List[(Variable, (STen, STen) => Unit)]