rmontanana/SVMClassifier
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
a1bf986365165e8446561ec3391d56af63cafa02
SVMClassifier/classsvm__classifier_1_1DataConverter.html
rmontanana a1bf986365 deploy: d6dc083a5a
2025-06-22 11:25:27 +00:00

56 KiB
Raw Blame History

<html xmlns="http://www.w3.org/1999/xhtml" lang="en-US"> <head> <script type="text/javascript" src="jquery.js"></script> <script type="text/javascript" src="dynsections.js"></script> <script type="text/javascript" src="search/searchdata.js"></script> <script type="text/javascript" src="search/search.js"></script> </head>
SVM Classifier C++ 1.0.0
High-performance Support Vector Machine classifier with scikit-learn compatible API
<script type="text/javascript"> /* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&dn=expat.txt MIT */ var searchBox = new SearchBox("searchBox", "search/",'.html'); /* @license-end */ </script> <script type="text/javascript" src="menudata.js"></script> <script type="text/javascript" src="menu.js"></script> <script type="text/javascript"> /* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&dn=expat.txt MIT */ $(function() { initMenu('',true,false,'search.php','Search'); $(document).ready(function() { init_search(); }); }); /* @license-end */ </script>
Loading...
Searching...
No Matches
Public Member Functions | List of all members
svm_classifier::DataConverter Class Reference

Data converter between libtorch tensors and SVM library formats. More...

#include <data_converter.hpp>

Public Member Functions

 DataConverter ()
 Default constructor.
 
 ~DataConverter ()
 Destructor - cleans up allocated memory.
 
std::unique_ptr< svm_problem > to_svm_problem (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Convert PyTorch tensors to libsvm format.
 
std::unique_ptr< problem > to_linear_problem (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Convert PyTorch tensors to liblinear format.
 
svm_node * to_svm_node (const torch::Tensor &sample)
 Convert single sample to libsvm format.
 
feature_node * to_feature_node (const torch::Tensor &sample)
 Convert single sample to liblinear format.
 
torch::Tensor from_predictions (const std::vector< double > &predictions)
 Convert predictions back to PyTorch tensor.
 
torch::Tensor from_probabilities (const std::vector< std::vector< double > > &probabilities)
 Convert probabilities back to PyTorch tensor.
 
torch::Tensor from_decision_values (const std::vector< std::vector< double > > &decision_values)
 Convert decision values back to PyTorch tensor.
 
void validate_tensors (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Validate input tensors.
 
int get_n_features () const
 Get number of features from last conversion.
 
int get_n_samples () const
 Get number of samples from last conversion.
 
void cleanup ()
 Clean up all allocated memory.
 
void set_sparse_threshold (double threshold)
 Set sparse threshold (features with absolute value below this are ignored)
 
double get_sparse_threshold () const
 Get sparse threshold.
 
 DataConverter ()
 Default constructor.
 
 ~DataConverter ()
 Destructor - cleans up allocated memory.
 
std::unique_ptr< svm_problem > to_svm_problem (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Convert PyTorch tensors to libsvm format.
 
std::unique_ptr< problem > to_linear_problem (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Convert PyTorch tensors to liblinear format.
 
svm_node * to_svm_node (const torch::Tensor &sample)
 Convert single sample to libsvm format.
 
feature_node * to_feature_node (const torch::Tensor &sample)
 Convert single sample to liblinear format.
 
torch::Tensor from_predictions (const std::vector< double > &predictions)
 Convert predictions back to PyTorch tensor.
 
torch::Tensor from_probabilities (const std::vector< std::vector< double > > &probabilities)
 Convert probabilities back to PyTorch tensor.
 
torch::Tensor from_decision_values (const std::vector< std::vector< double > > &decision_values)
 Convert decision values back to PyTorch tensor.
 
void validate_tensors (const torch::Tensor &X, const torch::Tensor &y=torch::Tensor())
 Validate input tensors.
 
int get_n_features () const
 Get number of features from last conversion.
 
int get_n_samples () const
 Get number of samples from last conversion.
 
void cleanup ()
 Clean up all allocated memory.
 
void set_sparse_threshold (double threshold)
 Set sparse threshold (features with absolute value below this are ignored)
 
double get_sparse_threshold () const
 Get sparse threshold.
 

Detailed Description

Data converter between libtorch tensors and SVM library formats.

This class handles the conversion between PyTorch tensors and the data structures required by libsvm and liblinear libraries. It manages memory allocation and provides efficient conversion methods.

Definition at line 23 of file data_converter.hpp.

Constructor & Destructor Documentation

◆ DataConverter()

svm_classifier::DataConverter::DataConverter ( )

Default constructor.

Definition at line 10 of file data_converter.cpp.

◆ ~DataConverter()

svm_classifier::DataConverter::~DataConverter ( )

Destructor - cleans up allocated memory.

Definition at line 17 of file data_converter.cpp.

Member Function Documentation

◆ cleanup()

void svm_classifier::DataConverter::cleanup ( )

Clean up all allocated memory.

Definition at line 198 of file data_converter.cpp.

◆ from_decision_values() [1/2]

torch::Tensor svm_classifier::DataConverter::from_decision_values ( const std::vector< std::vector< double > > &  decision_values)

Convert decision values back to PyTorch tensor.

Parameters
decision_values2D vector of decision function values
Returns
PyTorch tensor with decision values

Definition at line 152 of file data_converter.cpp.

◆ from_decision_values() [2/2]

torch::Tensor svm_classifier::DataConverter::from_decision_values ( const std::vector< std::vector< double > > &  decision_values)

Convert decision values back to PyTorch tensor.

Parameters
decision_values2D vector of decision function values
Returns
PyTorch tensor with decision values

◆ from_predictions() [1/2]

torch::Tensor svm_classifier::DataConverter::from_predictions ( const std::vector< double > &  predictions)

Convert predictions back to PyTorch tensor.

Parameters
predictionsVector of predictions
Returns
PyTorch tensor with predictions

Definition at line 120 of file data_converter.cpp.

◆ from_predictions() [2/2]

torch::Tensor svm_classifier::DataConverter::from_predictions ( const std::vector< double > &  predictions)

Convert predictions back to PyTorch tensor.

Parameters
predictionsVector of predictions
Returns
PyTorch tensor with predictions

◆ from_probabilities() [1/2]

torch::Tensor svm_classifier::DataConverter::from_probabilities ( const std::vector< std::vector< double > > &  probabilities)

Convert probabilities back to PyTorch tensor.

Parameters
probabilities2D vector of class probabilities
Returns
PyTorch tensor with probabilities of shape (n_samples, n_classes)

Definition at line 132 of file data_converter.cpp.

◆ from_probabilities() [2/2]

torch::Tensor svm_classifier::DataConverter::from_probabilities ( const std::vector< std::vector< double > > &  probabilities)

Convert probabilities back to PyTorch tensor.

Parameters
probabilities2D vector of class probabilities
Returns
PyTorch tensor with probabilities of shape (n_samples, n_classes)

◆ get_n_features() [1/2]

int svm_classifier::DataConverter::get_n_features ( ) const
inline

Get number of features from last conversion.

Returns
Number of features

Definition at line 100 of file data_converter.hpp.

◆ get_n_features() [2/2]

int svm_classifier::DataConverter::get_n_features ( ) const
inline

Get number of features from last conversion.

Returns
Number of features

Definition at line 100 of file kernel_parameters.hpp.

◆ get_n_samples() [1/2]

int svm_classifier::DataConverter::get_n_samples ( ) const
inline

Get number of samples from last conversion.

Returns
Number of samples

Definition at line 106 of file data_converter.hpp.

◆ get_n_samples() [2/2]

int svm_classifier::DataConverter::get_n_samples ( ) const
inline

Get number of samples from last conversion.

Returns
Number of samples

Definition at line 106 of file kernel_parameters.hpp.

◆ get_sparse_threshold() [1/2]

double svm_classifier::DataConverter::get_sparse_threshold ( ) const
inline

Get sparse threshold.

Returns
Current sparse threshold

Definition at line 123 of file data_converter.hpp.

◆ get_sparse_threshold() [2/2]

double svm_classifier::DataConverter::get_sparse_threshold ( ) const
inline

Get sparse threshold.

Returns
Current sparse threshold

Definition at line 123 of file kernel_parameters.hpp.

◆ set_sparse_threshold() [1/2]

void svm_classifier::DataConverter::set_sparse_threshold ( double  threshold)
inline

Set sparse threshold (features with absolute value below this are ignored)

Parameters
thresholdSparse threshold (default: 1e-8)

Definition at line 117 of file data_converter.hpp.

◆ set_sparse_threshold() [2/2]

void svm_classifier::DataConverter::set_sparse_threshold ( double  threshold)
inline

Set sparse threshold (features with absolute value below this are ignored)

Parameters
thresholdSparse threshold (default: 1e-8)

Definition at line 117 of file kernel_parameters.hpp.

◆ to_feature_node() [1/2]

feature_node * svm_classifier::DataConverter::to_feature_node ( const torch::Tensor &  sample)

Convert single sample to liblinear format.

Parameters
sampleFeature tensor of shape (n_features,)
Returns
Pointer to feature_node array

Definition at line 110 of file data_converter.cpp.

◆ to_feature_node() [2/2]

feature_node * svm_classifier::DataConverter::to_feature_node ( const torch::Tensor &  sample)

Convert single sample to liblinear format.

Parameters
sampleFeature tensor of shape (n_features,)
Returns
Pointer to feature_node array

◆ to_linear_problem() [1/2]

std::unique_ptr< problem > svm_classifier::DataConverter::to_linear_problem ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Convert PyTorch tensors to liblinear format.

Parameters
XFeature tensor of shape (n_samples, n_features)
yTarget tensor of shape (n_samples,) - optional for prediction
Returns
Pointer to problem structure

Definition at line 60 of file data_converter.cpp.

◆ to_linear_problem() [2/2]

std::unique_ptr< problem > svm_classifier::DataConverter::to_linear_problem ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Convert PyTorch tensors to liblinear format.

Parameters
XFeature tensor of shape (n_samples, n_features)
yTarget tensor of shape (n_samples,) - optional for prediction
Returns
Pointer to problem structure

◆ to_svm_node() [1/2]

svm_node * svm_classifier::DataConverter::to_svm_node ( const torch::Tensor &  sample)

Convert single sample to libsvm format.

Parameters
sampleFeature tensor of shape (n_features,)
Returns
Pointer to svm_node array

Definition at line 100 of file data_converter.cpp.

◆ to_svm_node() [2/2]

svm_node * svm_classifier::DataConverter::to_svm_node ( const torch::Tensor &  sample)

Convert single sample to libsvm format.

Parameters
sampleFeature tensor of shape (n_features,)
Returns
Pointer to svm_node array

◆ to_svm_problem() [1/2]

std::unique_ptr< svm_problem > svm_classifier::DataConverter::to_svm_problem ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Convert PyTorch tensors to libsvm format.

Parameters
XFeature tensor of shape (n_samples, n_features)
yTarget tensor of shape (n_samples,) - optional for prediction
Returns
Pointer to svm_problem structure

Definition at line 22 of file data_converter.cpp.

◆ to_svm_problem() [2/2]

std::unique_ptr< svm_problem > svm_classifier::DataConverter::to_svm_problem ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Convert PyTorch tensors to libsvm format.

Parameters
XFeature tensor of shape (n_samples, n_features)
yTarget tensor of shape (n_samples,) - optional for prediction
Returns
Pointer to svm_problem structure

◆ validate_tensors() [1/2]

void svm_classifier::DataConverter::validate_tensors ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Validate input tensors.

Parameters
XFeature tensor
yTarget tensor (optional)
Exceptions
std::invalid_argumentif tensors are invalid

Definition at line 172 of file data_converter.cpp.

◆ validate_tensors() [2/2]

void svm_classifier::DataConverter::validate_tensors ( const torch::Tensor &  X,
const torch::Tensor &  y = torch::Tensor() 
)

Validate input tensors.

Parameters
XFeature tensor
yTarget tensor (optional)
Exceptions
std::invalid_argumentif tensors are invalid
The documentation for this class was generated from the following files:
Generated on Sun Jun 22 2025 11:25:27 for SVM Classifier C++ by doxygen 1.9.8 </html>