Functions
Link copied to clipboard
fun <T : DType, V> customUpsampling(inChannels: Int, outChannels: Int, upsamplingFactor: Pair<Int, Int>, kernelSize: Pair<Int, Int>? = null, padding: Pair<Int, Int>? = null, bias: Boolean = true, name: String = "TransposedConv2d_custom", initWeights: Tensor<T, V>, initBias: Tensor<T, V>? = null): TransposedConv2d<T, V>
Creates a transposed convolution for custom upsampling factor.
Link copied to clipboard
fun <T : DType, V> upsampling2x(inChannels: Int, outChannels: Int, kernelSize: Pair<Int, Int> = 4 to 4, padding: Pair<Int, Int> = 1 to 1, bias: Boolean = true, name: String = "TransposedConv2d_2x", initWeights: Tensor<T, V>, initBias: Tensor<T, V>? = null): TransposedConv2d<T, V>
Creates a standard 2x upsampling transposed convolution.