C interface for Convolution. More...

afError	af_conv_SS (float d_out, unsigned ni, const float d_in, unsigned batchi, unsigned nf, const float *d_filt, unsigned batchf, bool expand)
	Single precision, real input, real filter, one dimensional convolution. More...

afError	af_conv_CS (cfloat d_out, unsigned ni, const cfloat d_in, unsigned batchi, unsigned nf, const float *d_filt, unsigned batchf, bool expand)
	Single precision, complex input, real filter, one dimensional convolution. More...

afError	af_conv_CC (cfloat d_out, unsigned ni, const cfloat d_in, unsigned batchi, unsigned nf, const cfloat *d_filt, unsigned batchf, bool expand)
	Float precision, complex input, complex filter, one dimensional convolution. More...

afError	af_conv_DD (double d_out, unsigned ni, const double d_in, unsigned batchi, unsigned nf, const double *d_filt, unsigned batchf, bool expand)
	Single precision, real input, real filter, one dimensional convolution. More...

afError	af_conv_ZD (cdouble d_out, unsigned ni, const cdouble d_in, unsigned batchi, unsigned nf, const double *d_filt, unsigned batchf, bool expand)
	Double precision, complex input, real filter, one dimensional convolution. More...

afError	af_conv_ZZ (cdouble d_out, unsigned ni, const cdouble d_in, unsigned batchi, unsigned nf, const cdouble *d_filt, unsigned batchf, bool expand)
	Double precision, complex input, complex filter, one dimensional convolution. More...

afError	af_conv2_SS (float d_out, unsigned mi, unsigned ni, const float d_in, unsigned batch_in, unsigned mf, unsigned nf, const float *d_filt, unsigned batch_filt, bool expand)

afError	af_conv2_CS (cfloat d_out, unsigned mi, unsigned ni, const cfloat d_in, unsigned batch_in, unsigned mf, unsigned nf, const float *d_filt, unsigned batch_filt, bool expand)
	Single precision, complex input, real filter, two dimensional convolution. More...

afError	af_conv2_CC (cfloat d_out, unsigned mi, unsigned ni, const cfloat d_in, unsigned batch_in, unsigned mf, unsigned nf, const cfloat *d_filt, unsigned batch_filt, bool expand)
	Single precision, complex input, complex filter, two dimensional convolution. More...

afError	af_conv2_DD (double d_out, unsigned mi, unsigned ni, const double d_in, unsigned batch_in, unsigned mf, unsigned nf, const double *d_filt, unsigned batch_filt, bool expand)
	Double precision, real input, real filter, two dimensional convolution. More...

afError	af_conv2_ZD (cdouble d_out, unsigned mi, unsigned ni, const cdouble d_in, unsigned batch_in, unsigned mf, unsigned nf, const double *d_filt, unsigned batch_filt, bool expand)
	Double precision, complex input, real filter, two dimensional convolution. More...

afError	af_conv2_ZZ (cdouble d_out, unsigned mi, unsigned ni, const cdouble d_in, unsigned batch_in, unsigned mf, unsigned nf, const cdouble *d_filt, unsigned batch_filt, bool expand)
	Double precision, complex input, complex filter, two dimensional convolution. More...

afError	af_convn_SS (float d_out, unsigned ndimsi, unsigned dimsi, const float d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const float *d_filt, unsigned batchf, bool expand)
	Single precision, real input, real filter, multi-dimensional convolution. More...

afError	af_convn_CS (cfloat d_out, unsigned ndimsi, unsigned dimsi, const cfloat d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const float *d_filt, unsigned batchf, bool expand)
	Single precision, complex input, real filter, multi-dimensional convolution. More...

afError	af_convn_CC (cfloat d_out, unsigned ndimsi, unsigned dimsi, const cfloat d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const cfloat *d_filt, unsigned batchf, bool expand)
	Single precision, complex input, complex filter, multi-dimensional convolution. More...

afError	af_convn_DD (double d_out, unsigned ndimsi, unsigned dimsi, const double d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const double *d_filt, unsigned batchf, bool expand)
	Double precision, real input, real filter, multi-dimensional convolution. More...

afError	af_convn_ZD (cdouble d_out, unsigned ndimsi, unsigned dimsi, const cdouble d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const double *d_filt, unsigned batchf, bool expand)
	Double precision, complex input, real filter, multi-dimensional convolution. More...

afError	af_convn_ZZ (cdouble d_out, unsigned ndimsi, unsigned dimsi, const cdouble d_in, unsigned batchi, unsigned ndimsf, unsigned dimsf, const cdouble *d_filt, unsigned batchf, bool expand)
	Double precision, complex input, complex filter, multi-dimensional convolution. More...

Device pointer interface: filter is host pointer

Multi-dimensional convolution (filter on host)

Parameters

[out]	d_out	The convolved output
[in]	ndimsi	Number of dimensions in input signal
[in]	dimsi	Dimensions of input signal
[in]	d_in	Input signal
[in]	batchi	Number of input signal tiles
[in]	ndimsf	Number of dimensions in input filter
[in]	dimsf	Dimensions of input filter
[in]	h_filt	Input filter (on host)
[in]	batchf	Number of input signal tiles
[in]	expand	perform expanded convolution if true.

afError af_convh_SS (float *d_out, unsigned ndimsi, unsigned *dimsi, const float *d_in, unsigned batchi, unsigned ndimsf, unsigned *dimsf, const float *h_filt, unsigned batchf, bool expand)

Single precision, real input, real filter, multi-dimensional convolution. More...

afError af_convh_CS (cfloat *d_out, unsigned ndimsi, unsigned *dimsi, const cfloat *d_in, unsigned batchi, unsigned ndimsf, unsigned *dimsf, const float *h_filt, unsigned batchf, bool expand)

Single precision, complex input, real filter, multi-dimensional convolution. More...

afError af_convh_DD (double *d_out, unsigned ndimsi, unsigned *dimsi, const double *d_in, unsigned batchi, unsigned ndimsf, unsigned *dimsf, const double *h_filt, unsigned batchf, bool expand)

Double precision, real input, real filter, multi-dimensional convolution. More...

afError af_convh_ZD (cdouble *d_out, unsigned ndimsi, unsigned *dimsi, const cdouble *d_in, unsigned batchi, unsigned ndimsf, unsigned *dimsf, const double *h_filt, unsigned batchf, bool expand)

Double precision, complex input, real filter, multi-dimensional convolution. More...

Detailed Description

C interface for Convolution.

A convolution is a common operation between a source array, a, and a filter (or kernel) array b. The answer to the convolution is the same as computing the coefficients in polynomial multiplication, if a and b are the coefficients.

Another way to think about it is that the filter kernel is centered on each pixel in a, and the output for that pixel or data point is the sum of the products.

Consider the following example with the calculation for the first few points shown as a comment:

array a = seq(1, 8);
//a = 1.0000
//    2.0000
//    3.0000
//    4.0000
//    5.0000
//    6.0000
//    7.0000
//    8.0000
array b = seq(0, 2);
b = pow2(b);
//b = 1.0000
//    2.0000
//    4.0000
array out = convolve(a,b);
//out = 4.0000 (0*4)+(1*2)+(2*1)
//     11.0000 (1*4)+(2*2)+(3*1)
//     18.0000 (2*4)+(3*2)+(4*1)
//     25.0000 ...
//     32.0000
//     39.0000
//     46.0000
//     44.0000

Note that pixels that are outside an image or data points that are outside the domain of a signal are treated as zero.

Function Documentation

afError af_conv_SS	(	float *	d_out,
		unsigned	ni,
		const float *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const float *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, real input, real filter, one dimensional convolution.

Parameters

[out]	d_out	The convolved output
[in]	ni	Size of input signal
[in]	d_in	Input signal
[in]	batchi	Number of input signal tiles
[in]	nf	Size of input filter
[in]	d_filt	Input filter
[in]	batchf	Number of input signal tiles
[in]	expand	perform expanded convolution if true.

afError af_conv_CS	(	cfloat *	d_out,
		unsigned	ni,
		const cfloat *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const float *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, complex input, real filter, one dimensional convolution.

afError af_conv_CC	(	cfloat *	d_out,
		unsigned	ni,
		const cfloat *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const cfloat *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Float precision, complex input, complex filter, one dimensional convolution.

afError af_conv_DD	(	double *	d_out,
		unsigned	ni,
		const double *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const double *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, real input, real filter, one dimensional convolution.

afError af_conv_ZD	(	cdouble *	d_out,
		unsigned	ni,
		const cdouble *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const double *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, complex input, real filter, one dimensional convolution.

afError af_conv_ZZ	(	cdouble *	d_out,
		unsigned	ni,
		const cdouble *	d_in,
		unsigned	batchi,
		unsigned	nf,
		const cdouble *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, complex input, complex filter, one dimensional convolution.

afError af_conv2_SS	(	float *	d_out,
		unsigned	mi,
		unsigned	ni,
		const float *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const float *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Parameters

[out]	d_out	The convolved output
[in]	mi	Rows in input signal
[in]	ni	Columns in input signal
[in]	d_in	Input signal
[in]	batch_in	Number of input signal tiles
[in]	mf	Rows in the input filter
[in]	nf	Columns in the input filter
[in]	d_filt	Input filter
[in]	batch_filt	Number of input signal tiles
[in]	expand	perform expanded convolution if true.Single precision, real input, real filter, two dimensional convolution

afError af_conv2_CS	(	cfloat *	d_out,
		unsigned	mi,
		unsigned	ni,
		const cfloat *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const float *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Single precision, complex input, real filter, two dimensional convolution.

afError af_conv2_CC	(	cfloat *	d_out,
		unsigned	mi,
		unsigned	ni,
		const cfloat *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const cfloat *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Single precision, complex input, complex filter, two dimensional convolution.

afError af_conv2_DD	(	double *	d_out,
		unsigned	mi,
		unsigned	ni,
		const double *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const double *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Double precision, real input, real filter, two dimensional convolution.

afError af_conv2_ZD	(	cdouble *	d_out,
		unsigned	mi,
		unsigned	ni,
		const cdouble *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const double *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Double precision, complex input, real filter, two dimensional convolution.

afError af_conv2_ZZ	(	cdouble *	d_out,
		unsigned	mi,
		unsigned	ni,
		const cdouble *	d_in,
		unsigned	batch_in,
		unsigned	mf,
		unsigned	nf,
		const cdouble *	d_filt,
		unsigned	batch_filt,
		bool	expand
	)

Double precision, complex input, complex filter, two dimensional convolution.

afError af_convn_SS	(	float *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const float *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const float *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, real input, real filter, multi-dimensional convolution.

Parameters

[out]	d_out	The convolved output
[in]	ndimsi	Number of dimensions in input signal
[in]	dimsi	Dimensions of input signal
[in]	d_in	Input signal
[in]	batchi	Number of input signal tiles
[in]	ndimsf	Number of dimensions in input filter
[in]	dimsf	Dimensions of input filter
[in]	d_filt	Input filter
[in]	batchf	Number of input signal tiles
[in]	expand	perform expanded convolution if true.

afError af_convn_CS	(	cfloat *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cfloat *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const float *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, complex input, real filter, multi-dimensional convolution.

afError af_convn_CC	(	cfloat *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cfloat *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const cfloat *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, complex input, complex filter, multi-dimensional convolution.

afError af_convn_DD	(	double *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const double *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const double *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, real input, real filter, multi-dimensional convolution.

afError af_convn_ZD	(	cdouble *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cdouble *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const double *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, complex input, real filter, multi-dimensional convolution.

afError af_convn_ZZ	(	cdouble *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cdouble *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const cdouble *	d_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, complex input, complex filter, multi-dimensional convolution.

afError af_convh_SS	(	float *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const float *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const float *	h_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, real input, real filter, multi-dimensional convolution.

afError af_convh_CS	(	cfloat *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cfloat *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const float *	h_filt,
		unsigned	batchf,
		bool	expand
	)

Single precision, complex input, real filter, multi-dimensional convolution.

afError af_convh_DD	(	double *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const double *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const double *	h_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, real input, real filter, multi-dimensional convolution.

afError af_convh_ZD	(	cdouble *	d_out,
		unsigned	ndimsi,
		unsigned *	dimsi,
		const cdouble *	d_in,
		unsigned	batchi,
		unsigned	ndimsf,
		unsigned *	dimsf,
		const double *	h_filt,
		unsigned	batchf,
		bool	expand
	)

Double precision, complex input, real filter, multi-dimensional convolution.