Struct SVMSGD

pub struct SVMSGD { /* private fields */ }

! Stochastic Gradient Descent SVM classifier

SVMSGD provides a fast and easy-to-use implementation of the SVM classifier using the Stochastic Gradient Descent approach, as presented in bottou2010large.

The classifier has following parameters:

• model type,
• margin type,
• margin regularization (),
• initial step size (),
• step decreasing power (),
• and termination criteria.

The model type may have one of the following values: \ref SGD and \ref ASGD.

• \ref SGD is the classic version of SVMSGD classifier: every next step is calculated by the formula where

  • is the weights vector for decision function at step ,
  • is the step size of model parameters at the iteration , it is decreased on each step by the formula
  • is the target functional from SVM task for sample with number , this sample is chosen stochastically on each step of the algorithm.

• \ref ASGD is Average Stochastic Gradient Descent SVM Classifier. ASGD classifier averages weights vector on each step of algorithm by the formula

The recommended model type is ASGD (following bottou2010large).

The margin type may have one of the following values: \ref SOFT_MARGIN or \ref HARD_MARGIN.

• You should use \ref HARD_MARGIN type, if you have linearly separable sets.
• You should use \ref SOFT_MARGIN type, if you have non-linearly separable sets or sets with outliers.
• In the general case (if you know nothing about linear separability of your sets), use SOFT_MARGIN.

The other parameters may be described as follows:

• Margin regularization parameter is responsible for weights decreasing at each step and for the strength of restrictions on outliers (the less the parameter, the less probability that an outlier will be ignored). Recommended value for SGD model is 0.0001, for ASGD model is 0.00001.

• Initial step size parameter is the initial value for the step size . You will have to find the best initial step for your problem.

• Step decreasing power is the power parameter for decreasing by the formula, mentioned above. Recommended value for SGD model is 1, for ASGD model is 0.75.

• Termination criteria can be TermCriteria::COUNT, TermCriteria::EPS or TermCriteria::COUNT + TermCriteria::EPS. You will have to find the best termination criteria for your problem.

Note that the parameters margin regularization, initial step size, and step decreasing power should be positive.

To use SVMSGD algorithm do as follows:

• first, create the SVMSGD object. The algorithm will set optimal parameters by default, but you can set your own parameters via functions setSvmsgdType(), setMarginType(), setMarginRegularization(), setInitialStepSize(), and setStepDecreasingPower().

• then the SVM model can be trained using the train features and the correspondent labels by the method train().

• after that, the label of a new feature vector can be predicted using the method predict().

// Create empty object
cv::Ptr<SVMSGD> svmsgd = SVMSGD::create();

// Train the Stochastic Gradient Descent SVM
svmsgd->train(trainData);

// Predict labels for the new samples
svmsgd->predict(samples, responses);

Implementations

impl SVMSGD

pub fn create() -> Result<Ptr<SVMSGD>>

Creates empty model. Use StatModel::train to train the model. Since %SVMSGD has several parameters, you may want to find the best parameters for your problem or use setOptimalParameters() to set some default parameters.

pub fn load(filepath: &str, node_name: &str) -> Result<Ptr<SVMSGD>>

Loads and creates a serialized SVMSGD from a file

Use SVMSGD::save to serialize and store an SVMSGD to disk. Load the SVMSGD from this file again, by calling this function with the path to the file. Optionally specify the node for the file containing the classifier

Parameters

• filepath: path to serialized SVMSGD
• nodeName: name of node containing the classifier

C++ default parameters

• node_name: String()

pub fn load_def(filepath: &str) -> Result<Ptr<SVMSGD>>

Loads and creates a serialized SVMSGD from a file

Use SVMSGD::save to serialize and store an SVMSGD to disk. Load the SVMSGD from this file again, by calling this function with the path to the file. Optionally specify the node for the file containing the classifier

Parameters

• filepath: path to serialized SVMSGD
• nodeName: name of node containing the classifier

Note

This alternative version of SVMSGD::load function uses the following default values for its arguments:

• node_name: String()

Trait Implementations

impl AlgorithmTrait for SVMSGD

fn as_raw_mut_Algorithm(&mut self) -> *mut c_void

fn clear(&mut self) -> Result<()>

Clears the algorithm state

fn read(&mut self, fn_: &impl FileNodeTraitConst) -> Result<()>

Reads algorithm parameters from a file storage

impl AlgorithmTraitConst for SVMSGD

fn as_raw_Algorithm(&self) -> *const c_void

fn write(&self, fs: &mut impl FileStorageTrait) -> Result<()>

Stores algorithm parameters in a file storage

fn write_1(&self, fs: &mut impl FileStorageTrait, name: &str) -> Result<()>

Stores algorithm parameters in a file storage

fn write_with_name(&self, fs: &Ptr<FileStorage>, name: &str) -> Result<()>

@deprecated

fn write_with_name_def(&self, fs: &Ptr<FileStorage>) -> Result<()>

👎Deprecated:

Note

Deprecated: ## Note This alternative version of AlgorithmTraitConst::write_with_name function uses the following default values for its arguments:

fn empty(&self) -> Result<bool>

Returns true if the Algorithm is empty (e.g. in the very beginning or after unsuccessful read

fn save(&self, filename: &str) -> Result<()>

Saves the algorithm to a file. In order to make this method work, the derived class must implement Algorithm::write(FileStorage& fs).

fn get_default_name(&self) -> Result<String>

Returns the algorithm string identifier. This string is used as top level xml/yml node tag when the object is saved to a file or string.

impl Boxed for SVMSGD

unsafe fn from_raw(ptr: <SVMSGD as OpenCVFromExtern>::ExternReceive) -> Self

Wrap the specified raw pointer

fn into_raw(self) -> <SVMSGD as OpenCVTypeExternContainer>::ExternSendMut

Return the underlying raw pointer while consuming this wrapper.

fn as_raw(&self) -> <SVMSGD as OpenCVTypeExternContainer>::ExternSend

Return the underlying raw pointer.

fn as_raw_mut(&mut self) -> <SVMSGD as OpenCVTypeExternContainer>::ExternSendMut

Return the underlying mutable raw pointer

impl Debug for SVMSGD

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Drop for SVMSGD

fn drop(&mut self)

Executes the destructor for this type.

impl From<SVMSGD> for Algorithm

fn from(s: SVMSGD) -> Self

Converts to this type from the input type.

impl From<SVMSGD> for StatModel

fn from(s: SVMSGD) -> Self

Converts to this type from the input type.

impl SVMSGDTrait for SVMSGD

fn as_raw_mut_SVMSGD(&mut self) -> *mut c_void

fn get_weights(&mut self) -> Result<Mat>

Returns

fn get_shift(&mut self) -> Result<f32>

Returns

fn set_optimal_parameters( &mut self, svmsgd_type: i32, margin_type: i32, ) -> Result<()>

Function sets optimal parameters values for chosen SVM SGD model.

fn set_optimal_parameters_def(&mut self) -> Result<()>

Function sets optimal parameters values for chosen SVM SGD model.

fn set_svmsgd_type(&mut self, svmsgd_type: i32) -> Result<()>

%Algorithm type, one of SVMSGD::SvmsgdType.

fn set_margin_type(&mut self, margin_type: i32) -> Result<()>

%Margin type, one of SVMSGD::MarginType.

fn set_margin_regularization( &mut self, margin_regularization: f32, ) -> Result<()>

Parameter marginRegularization of a %SVMSGD optimization problem.

fn set_initial_step_size(&mut self, initial_step_size: f32) -> Result<()>

Parameter initialStepSize of a %SVMSGD optimization problem.

fn set_step_decreasing_power( &mut self, step_decreasing_power: f32, ) -> Result<()>

Parameter stepDecreasingPower of a %SVMSGD optimization problem.

fn set_term_criteria(&mut self, val: TermCriteria) -> Result<()>

Termination criteria of the training algorithm. You can specify the maximum number of iterations (maxCount) and/or how much the error could change between the iterations to make the algorithm continue (epsilon).

impl SVMSGDTraitConst for SVMSGD

fn as_raw_SVMSGD(&self) -> *const c_void

fn get_svmsgd_type(&self) -> Result<i32>

%Algorithm type, one of SVMSGD::SvmsgdType.

fn get_margin_type(&self) -> Result<i32>

%Margin type, one of SVMSGD::MarginType.

fn get_margin_regularization(&self) -> Result<f32>

Parameter marginRegularization of a %SVMSGD optimization problem.

fn get_initial_step_size(&self) -> Result<f32>

Parameter initialStepSize of a %SVMSGD optimization problem.

fn get_step_decreasing_power(&self) -> Result<f32>

Parameter stepDecreasingPower of a %SVMSGD optimization problem.

fn get_term_criteria(&self) -> Result<TermCriteria>

Termination criteria of the training algorithm. You can specify the maximum number of iterations (maxCount) and/or how much the error could change between the iterations to make the algorithm continue (epsilon).

impl StatModelTrait for SVMSGD

fn as_raw_mut_StatModel(&mut self) -> *mut c_void

fn train_with_data( &mut self, train_data: &Ptr<TrainData>, flags: i32, ) -> Result<bool>

Trains the statistical model

fn train_with_data_def(&mut self, train_data: &Ptr<TrainData>) -> Result<bool>

Trains the statistical model

fn train( &mut self, samples: &impl ToInputArray, layout: i32, responses: &impl ToInputArray, ) -> Result<bool>

Trains the statistical model

impl StatModelTraitConst for SVMSGD

fn as_raw_StatModel(&self) -> *const c_void

fn get_var_count(&self) -> Result<i32>

Returns the number of variables in training samples

fn empty(&self) -> Result<bool>

fn is_trained(&self) -> Result<bool>

Returns true if the model is trained

fn is_classifier(&self) -> Result<bool>

Returns true if the model is classifier

fn calc_error( &self, data: &Ptr<TrainData>, test: bool, resp: &mut impl ToOutputArray, ) -> Result<f32>

Computes error on the training or test dataset

fn predict( &self, samples: &impl ToInputArray, results: &mut impl ToOutputArray, flags: i32, ) -> Result<f32>

Predicts response(s) for the provided sample(s)

fn predict_def(&self, samples: &impl ToInputArray) -> Result<f32>

Predicts response(s) for the provided sample(s)

impl TryFrom<StatModel> for SVMSGD

type Error = Error

The type returned in the event of a conversion error.

fn try_from(s: StatModel) -> Result<Self>

Performs the conversion.

impl Send for SVMSGD