Commit 422396ef authored by peng xiao's avatar peng xiao

fix build error

parent f69ccfa4
...@@ -450,7 +450,8 @@ float* CvSVMSolver_ocl::get_row_base( int i, bool* _existed, Mat& src ) ...@@ -450,7 +450,8 @@ float* CvSVMSolver_ocl::get_row_base( int i, bool* _existed, Mat& src )
return row->data; return row->data;
} }
void matmul_sigmod(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1) #ifndef HAVE_CLAMDBLAS
static void matmul_sigmod(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1)
{ {
Context *clCxt = Context::getContext(); Context *clCxt = Context::getContext();
string kernelName = "svm_sigmod"; string kernelName = "svm_sigmod";
...@@ -489,7 +490,7 @@ void matmul_sigmod(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int ...@@ -489,7 +490,7 @@ void matmul_sigmod(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int
} }
openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1); openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1);
} }
void matmul_poly(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1, double degree1, bool flag) static void matmul_poly(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1, double degree1, bool flag)
{ {
Context *clCxt = Context::getContext(); Context *clCxt = Context::getContext();
string kernelName = "svm_poly"; string kernelName = "svm_poly";
...@@ -537,7 +538,7 @@ void matmul_poly(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int sr ...@@ -537,7 +538,7 @@ void matmul_poly(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int sr
} }
openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1, build_options); openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1, build_options);
} }
void matmul_linear(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1) static void matmul_linear(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int src2_cols, int var_count, double alpha1, double beta1)
{ {
Context *clCxt = Context::getContext(); Context *clCxt = Context::getContext();
string kernelName = "svm_linear"; string kernelName = "svm_linear";
...@@ -576,7 +577,9 @@ void matmul_linear(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int ...@@ -576,7 +577,9 @@ void matmul_linear(oclMat & src, oclMat & src2, oclMat & dst, int src_rows, int
} }
openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1); openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1);
} }
void matmul_rbf(oclMat& src, oclMat& src_e, oclMat& dst, int src_rows, int src2_cols, int var_count, double gamma1, bool flag) #endif // #ifndef HAVE_CLAMDBLAS
static void matmul_rbf(oclMat& src, oclMat& src_e, oclMat& dst, int src_rows, int src2_cols, int var_count, double gamma1, bool flag)
{ {
Context *clCxt = Context::getContext(); Context *clCxt = Context::getContext();
...@@ -621,6 +624,7 @@ void matmul_rbf(oclMat& src, oclMat& src_e, oclMat& dst, int src_rows, int src2_ ...@@ -621,6 +624,7 @@ void matmul_rbf(oclMat& src, oclMat& src_e, oclMat& dst, int src_rows, int src2_
openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1, build_options); openCLExecuteKernel(clCxt, &svm, kernelName, globalThreads, localThreads, args, -1, -1, build_options);
} }
float CvSVM_OCL::predict(const CvMat* samples, CV_OUT CvMat* results) const float CvSVM_OCL::predict(const CvMat* samples, CV_OUT CvMat* results) const
{ {
int var_count = get_var_count(); int var_count = get_var_count();
...@@ -1169,7 +1173,7 @@ void CvSVMKernel_ocl::calc_sigmoid( int vcount, const int row_idx, Qfloat* resul ...@@ -1169,7 +1173,7 @@ void CvSVMKernel_ocl::calc_sigmoid( int vcount, const int row_idx, Qfloat* resul
} }
CvSVM_OCL::CvSVM_OCL() CvSVM_OCL::CvSVM_OCL()
{ {
CvSVM::CvSVM(); CvSVM();
} }
CvSVM_OCL::CvSVM_OCL( const Mat& _train_data, const Mat& _responses, CvSVM_OCL::CvSVM_OCL( const Mat& _train_data, const Mat& _responses,
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment