/// <summary>
/// build a convolution function
/// </summary>
/// <param name="convolutionMap">convolution parameters (shape, type of the kernal)</param>
/// <param name="operand">input variable</param>
/// <param name="strides">strides to apply convolution</param>
/// <param name="sharing">whether to share parameters (default = true)</param>
/// <param name="autoPadding"></param>
/// <returns></returns>
public static Function Convolution(Variable convolutionMap, Variable operand, NDShape strides, IEnumerable <bool> sharing, IEnumerable <bool> autoPadding)
{
BoolVector sharingVec = Helper.AsBoolVector(sharing);
BoolVector autoPaddingVec = Helper.AsBoolVector(autoPadding);
return(CNTKLib.Convolution(convolutionMap, operand, strides, sharingVec, autoPaddingVec));
}
private static OneArgumentModule Convolution(
int nbFilters,
int[] filterShape,
int[] strides,
bool[] paddings,
int[] dilations,
bool bias,
bool[] sharing,
uint reductionRank = 1,
string name = ""
)
{
var kernelInitializer = C.GlorotUniformInitializer(
C.DefaultParamInitScale,
C.SentinelValueForInferParamInitRank,
C.SentinelValueForInferParamInitRank);
var rank = filterShape.Length;
int[] kernelShape = new int[rank + 2];
for (int i = 0; i < filterShape.Length; i++)
{
kernelShape[i] = filterShape[i];
}
kernelShape[rank] = NDShape.InferredDimension;
kernelShape[rank + 1] = nbFilters;
int[] biasShape = new int[rank + 1];
for (int i = 0; i < rank; i++)
{
biasShape[i] = 1;
}
biasShape[rank] = nbFilters;
Parameter w = new Parameter(kernelShape, DataType.Float, kernelInitializer);
Parameter b = bias ? new Parameter(biasShape, DataType.Float, 0) : null;
Function _Convolve(Variable x)
{
var r = C.Convolution(w, x, strides, new BoolVector(sharing), new BoolVector(paddings), dilations, reductionRank, 0, name);
if (bias)
{
r += b;
}
return(r);
}
return(_Convolve);
}