SVM Functions

Functions
void	arm_svm_linear_init_f32 (arm_svm_linear_instance_f32 *S, uint32_t nbOfSupportVectors, uint32_t vectorDimension, float32_t intercept, const float32_t *dualCoefficients, const float32_t *supportVectors, const int32_t *classes)
	SVM linear instance init function. More...
void	arm_svm_linear_predict_f32 (const arm_svm_linear_instance_f32 *S, const float32_t *in, int32_t *pResult)
	SVM linear prediction. More...
void	arm_svm_polynomial_init_f32 (arm_svm_polynomial_instance_f32 *S, uint32_t nbOfSupportVectors, uint32_t vectorDimension, float32_t intercept, const float32_t *dualCoefficients, const float32_t *supportVectors, const int32_t *classes, int32_t degree, float32_t coef0, float32_t gamma)
	SVM polynomial instance init function. More...
void	arm_svm_polynomial_predict_f32 (const arm_svm_polynomial_instance_f32 *S, const float32_t *in, int32_t *pResult)
	SVM polynomial prediction. More...
void	arm_svm_rbf_init_f32 (arm_svm_rbf_instance_f32 *S, uint32_t nbOfSupportVectors, uint32_t vectorDimension, float32_t intercept, const float32_t *dualCoefficients, const float32_t *supportVectors, const int32_t *classes, float32_t gamma)
	SVM radial basis function instance init function. More...
void	arm_svm_rbf_predict_f32 (const arm_svm_rbf_instance_f32 *S, const float32_t *in, int32_t *pResult)
	SVM rbf prediction. More...
void	arm_svm_sigmoid_init_f32 (arm_svm_sigmoid_instance_f32 *S, uint32_t nbOfSupportVectors, uint32_t vectorDimension, float32_t intercept, const float32_t *dualCoefficients, const float32_t *supportVectors, const int32_t *classes, float32_t coef0, float32_t gamma)
	SVM sigmoid instance init function. More...
void	arm_svm_sigmoid_predict_f32 (const arm_svm_sigmoid_instance_f32 *S, const float32_t *in, int32_t *pResult)
	SVM sigmoid prediction. More...

Description

This set of functions is implementing SVM classification on 2 classes. The training must be done from scikit-learn. The parameters can be easily generated from the scikit-learn object. Some examples are given in DSP/Testing/PatternGeneration/SVM.py

If more than 2 classes are needed, the functions in this folder will have to be used, as building blocks, to do multi-class classification.

No multi-class classification is provided in this SVM folder.

Function Documentation

void arm_svm_linear_init_f32	(	arm_svm_linear_instance_f32 *	S,
		uint32_t	nbOfSupportVectors,
		uint32_t	vectorDimension,
		float32_t	intercept,
		const float32_t *	dualCoefficients,
		const float32_t *	supportVectors,
		const int32_t *	classes
	)

Classes are integer used as output of the function (instead of having -1,1 as class values).

Parameters

[in]	S	Parameters for the SVM function
[in]	nbOfSupportVectors	Number of support vectors
[in]	vectorDimension	Dimension of vector space
[in]	intercept	Intercept
[in]	dualCoefficients	Array of dual coefficients
[in]	supportVectors	Array of support vectors
[in]	classes	Array of 2 classes ID

Returns

none.

void arm_svm_linear_predict_f32	(	const arm_svm_linear_instance_f32 *	S,
		const float32_t *	in,
		int32_t *	pResult
	)

Parameters

[in]	S	Pointer to an instance of the linear SVM structure.
[in]	in	Pointer to input vector
[out]	pResult	Decision value

Returns

none.

void arm_svm_polynomial_init_f32	(	arm_svm_polynomial_instance_f32 *	S,
		uint32_t	nbOfSupportVectors,
		uint32_t	vectorDimension,
		float32_t	intercept,
		const float32_t *	dualCoefficients,
		const float32_t *	supportVectors,
		const int32_t *	classes,
		int32_t	degree,
		float32_t	coef0,
		float32_t	gamma
	)

Classes are integer used as output of the function (instead of having -1,1 as class values).

Parameters

[in]	S	points to an instance of the polynomial SVM structure.
[in]	nbOfSupportVectors	Number of support vectors
[in]	vectorDimension	Dimension of vector space
[in]	intercept	Intercept
[in]	dualCoefficients	Array of dual coefficients
[in]	supportVectors	Array of support vectors
[in]	classes	Array of 2 classes ID
[in]	degree	Polynomial degree
[in]	coef0	coeff0 (scikit-learn terminology)
[in]	gamma	gamma (scikit-learn terminology)

Returns

none.

void arm_svm_polynomial_predict_f32	(	const arm_svm_polynomial_instance_f32 *	S,
		const float32_t *	in,
		int32_t *	pResult
	)

Parameters

[in]	S	Pointer to an instance of the polynomial SVM structure.
[in]	in	Pointer to input vector
[out]	pResult	Decision value

Returns

none.

void arm_svm_rbf_init_f32	(	arm_svm_rbf_instance_f32 *	S,
		uint32_t	nbOfSupportVectors,
		uint32_t	vectorDimension,
		float32_t	intercept,
		const float32_t *	dualCoefficients,
		const float32_t *	supportVectors,
		const int32_t *	classes,
		float32_t	gamma
	)

Classes are integer used as output of the function (instead of having -1,1 as class values).

Parameters

[in]	S	points to an instance of the polynomial SVM structure.
[in]	nbOfSupportVectors	Number of support vectors
[in]	vectorDimension	Dimension of vector space
[in]	intercept	Intercept
[in]	dualCoefficients	Array of dual coefficients
[in]	supportVectors	Array of support vectors
[in]	classes	Array of 2 classes ID
[in]	gamma	gamma (scikit-learn terminology)

Returns

none.

void arm_svm_rbf_predict_f32	(	const arm_svm_rbf_instance_f32 *	S,
		const float32_t *	in,
		int32_t *	pResult
	)

Parameters

[in]	S	Pointer to an instance of the rbf SVM structure.
[in]	in	Pointer to input vector
[out]	pResult	decision value

Returns

none.

void arm_svm_sigmoid_init_f32	(	arm_svm_sigmoid_instance_f32 *	S,
		uint32_t	nbOfSupportVectors,
		uint32_t	vectorDimension,
		float32_t	intercept,
		const float32_t *	dualCoefficients,
		const float32_t *	supportVectors,
		const int32_t *	classes,
		float32_t	coef0,
		float32_t	gamma
	)

Classes are integer used as output of the function (instead of having -1,1 as class values).

Parameters

[in]	S	points to an instance of the rbf SVM structure.
[in]	nbOfSupportVectors	Number of support vectors
[in]	vectorDimension	Dimension of vector space
[in]	intercept	Intercept
[in]	dualCoefficients	Array of dual coefficients
[in]	supportVectors	Array of support vectors
[in]	classes	Array of 2 classes ID
[in]	coef0	coeff0 (scikit-learn terminology)
[in]	gamma	gamma (scikit-learn terminology)

Returns

none.

void arm_svm_sigmoid_predict_f32	(	const arm_svm_sigmoid_instance_f32 *	S,
		const float32_t *	in,
		int32_t *	pResult
	)

Parameters

[in]	S	Pointer to an instance of the rbf SVM structure.
[in]	in	Pointer to input vector
[out]	pResult	Decision value

Returns

none.