potok/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-07-29 15:23:05 +04:00
Code Packages Releases Wiki Activity

Partially back-port #25075 to 4.x

Browse Source
This commit is contained in:
Alexander Smorkalov
2024-03-04 15:51:05 +03:00
parent ef611df09b
commit daa8f7dfc6
111 changed files with 1124 additions and 1124 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/ml/src/svm.cpp
+23 -23
View File
@@ -95,11 +95,11 @@ const int QFLOAT_TYPE = DataDepth<Qfloat>::value;
static void checkParamGrid(const ParamGrid& pg)
{
if( pg.minVal > pg.maxVal )
CV_Error( CV_StsBadArg, "Lower bound of the grid must be less then the upper one" );
CV_Error( cv::Error::StsBadArg, "Lower bound of the grid must be less then the upper one" );
if( pg.minVal < DBL_EPSILON )
CV_Error( CV_StsBadArg, "Lower bound of the grid must be positive" );
CV_Error( cv::Error::StsBadArg, "Lower bound of the grid must be positive" );
if( pg.logStep < 1. + FLT_EPSILON )
CV_Error( CV_StsBadArg, "Grid step must greater than 1" );
CV_Error( cv::Error::StsBadArg, "Grid step must greater than 1" );
}
// SVM training parameters
@@ -325,7 +325,7 @@ public:
calc_intersec(vcount, var_count, vecs, another, results);
break;
default:
CV_Error(CV_StsBadArg, "Unknown kernel type");
CV_Error(cv::Error::StsBadArg, "Unknown kernel type");
}
const Qfloat max_val = (Qfloat)(FLT_MAX*1e-3);
for( int j = 0; j < vcount; j++ )
@@ -410,7 +410,7 @@ ParamGrid SVM::getDefaultGrid( int param_id )
grid.logStep = 7; // total iterations = 3
}
else
cvError( CV_StsBadArg, "SVM::getDefaultGrid", "Invalid type of parameter "
cvError( cv::Error::StsBadArg, "SVM::getDefaultGrid", "Invalid type of parameter "
"(use one of SVM::C, SVM::GAMMA et al.)", __FILE__, __LINE__ );
return grid;
}
@@ -1297,12 +1297,12 @@ public:
if( kernelType != LINEAR && kernelType != POLY &&
kernelType != SIGMOID && kernelType != RBF &&
kernelType != INTER && kernelType != CHI2)
CV_Error( CV_StsBadArg, "Unknown/unsupported kernel type" );
CV_Error( cv::Error::StsBadArg, "Unknown/unsupported kernel type" );
if( kernelType == LINEAR )
params.gamma = 1;
else if( params.gamma <= 0 )
CV_Error( CV_StsOutOfRange, "gamma parameter of the kernel must be positive" );
CV_Error( cv::Error::StsOutOfRange, "gamma parameter of the kernel must be positive" );
if( kernelType != SIGMOID && kernelType != POLY )
params.coef0 = 0;
@@ -1310,14 +1310,14 @@ public:
if( kernelType != POLY )
params.degree = 0;
else if( params.degree <= 0 )
CV_Error( CV_StsOutOfRange, "The kernel parameter <degree> must be positive" );
CV_Error( cv::Error::StsOutOfRange, "The kernel parameter <degree> must be positive" );
kernel = makePtr<SVMKernelImpl>(params);
}
else
{
if (!kernel)
CV_Error( CV_StsBadArg, "Custom kernel is not set" );
CV_Error( cv::Error::StsBadArg, "Custom kernel is not set" );
}
int svmType = params.svmType;
@@ -1325,22 +1325,22 @@ public:
if( svmType != C_SVC && svmType != NU_SVC &&
svmType != ONE_CLASS && svmType != EPS_SVR &&
svmType != NU_SVR )
CV_Error( CV_StsBadArg, "Unknown/unsupported SVM type" );
CV_Error( cv::Error::StsBadArg, "Unknown/unsupported SVM type" );
if( svmType == ONE_CLASS || svmType == NU_SVC )
params.C = 0;
else if( params.C <= 0 )
CV_Error( CV_StsOutOfRange, "The parameter C must be positive" );
CV_Error( cv::Error::StsOutOfRange, "The parameter C must be positive" );
if( svmType == C_SVC || svmType == EPS_SVR )
params.nu = 0;
else if( params.nu <= 0 || params.nu >= 1 )
CV_Error( CV_StsOutOfRange, "The parameter nu must be between 0 and 1" );
CV_Error( cv::Error::StsOutOfRange, "The parameter nu must be between 0 and 1" );
if( svmType != EPS_SVR )
params.p = 0;
else if( params.p <= 0 )
CV_Error( CV_StsOutOfRange, "The parameter p must be positive" );
CV_Error( cv::Error::StsOutOfRange, "The parameter p must be positive" );
if( svmType != C_SVC )
params.classWeights.release();
@@ -1431,7 +1431,7 @@ public:
if( (cw.cols != 1 && cw.rows != 1) ||
(int)cw.total() != class_count ||
(cw.type() != CV_32F && cw.type() != CV_64F) )
CV_Error( CV_StsBadArg, "params.class_weights must be 1d floating-point vector "
CV_Error( cv::Error::StsBadArg, "params.class_weights must be 1d floating-point vector "
"containing as many elements as the number of classes" );
cw.convertTo(class_weights, CV_64F, params.C);
@@ -1446,7 +1446,7 @@ public:
//check that while cross-validation there were the samples from all the classes
if ((int)class_ranges.size() < class_count + 1)
CV_Error( CV_StsBadArg, "While cross-validation one or more of the classes have "
CV_Error( cv::Error::StsBadArg, "While cross-validation one or more of the classes have "
"been fell out of the sample. Try to reduce <Params::k_fold>" );
if( svmType == NU_SVC )
@@ -1620,7 +1620,7 @@ public:
{
responses = data->getTrainNormCatResponses();
if( responses.empty() )
CV_Error(CV_StsBadArg, "in the case of classification problem the responses must be categorical; "
CV_Error(cv::Error::StsBadArg, "in the case of classification problem the responses must be categorical; "
"either specify varType when creating TrainData, or pass integer responses");
class_labels = data->getClassLabels();
}
@@ -1969,7 +1969,7 @@ public:
}
}
else
CV_Error( CV_StsBadArg, "INTERNAL ERROR: Unknown SVM type, "
CV_Error( cv::Error::StsBadArg, "INTERNAL ERROR: Unknown SVM type, "
"the SVM structure is probably corrupted" );
}
@@ -2112,7 +2112,7 @@ public:
int class_count = !class_labels.empty() ? (int)class_labels.total() :
params.svmType == ONE_CLASS ? 1 : 0;
if( !isTrained() )
CV_Error( CV_StsParseError, "SVM model data is invalid, check sv_count, var_* and class_count tags" );
CV_Error( cv::Error::StsParseError, "SVM model data is invalid, check sv_count, var_* and class_count tags" );
writeFormat(fs);
write_params( fs );
@@ -2197,11 +2197,11 @@ public:
svm_type_str == "NU_SVR" ? NU_SVR : -1;
if( svmType < 0 )
CV_Error( CV_StsParseError, "Missing or invalid SVM type" );
CV_Error( cv::Error::StsParseError, "Missing or invalid SVM type" );
FileNode kernel_node = fn["kernel"];
if( kernel_node.empty() )
CV_Error( CV_StsParseError, "SVM kernel tag is not found" );
CV_Error( cv::Error::StsParseError, "SVM kernel tag is not found" );
String kernel_type_str = (String)kernel_node["type"];
int kernelType =
@@ -2213,7 +2213,7 @@ public:
kernel_type_str == "INTER" ? INTER : CUSTOM;
if( kernelType == CUSTOM )
CV_Error( CV_StsParseError, "Invalid SVM kernel type (or custom kernel)" );
CV_Error( cv::Error::StsParseError, "Invalid SVM kernel type (or custom kernel)" );
_params.svmType = svmType;
_params.kernelType = kernelType;
@@ -2253,7 +2253,7 @@ public:
int class_count = (int)fn["class_count"];
if( sv_total <= 0 || var_count <= 0 )
CV_Error( CV_StsParseError, "SVM model data is invalid, check sv_count, var_* and class_count tags" );
CV_Error( cv::Error::StsParseError, "SVM model data is invalid, check sv_count, var_* and class_count tags" );
FileNode m = fn["class_labels"];
if( !m.empty() )
@@ -2263,7 +2263,7 @@ public:
m >> params.classWeights;
if( class_count > 1 && (class_labels.empty() || (int)class_labels.total() != class_count))
CV_Error( CV_StsParseError, "Array of class labels is missing or invalid" );
CV_Error( cv::Error::StsParseError, "Array of class labels is missing or invalid" );
// read support vectors
FileNode sv_node = fn["support_vectors"];