Struct ANN_MLP

Source
pub struct ANN_MLP { /* private fields */ }
Expand description

Artificial Neural Networks - Multi-Layer Perceptrons.

Unlike many other models in ML that are constructed and trained at once, in the MLP model these steps are separated. First, a network with the specified topology is created using the non-default constructor or the method ANN_MLP::create. All the weights are set to zeros. Then, the network is trained using a set of input and output vectors. The training procedure can be repeated more than once, that is, the weights can be adjusted based on the new training data.

Additional flags for StatModel::train are available: ANN_MLP::TrainFlags.

§See also

[ml_intro_ann]

Implementations§

Source§

impl ANN_MLP

Source

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

Creates empty model

Use StatModel::train to train the model, Algorithm::load<ANN_MLP>(filename) to load the pre-trained model. Note that the train method has optional flags: ANN_MLP::TrainFlags.

Source

pub fn load(filepath: &str) -> Result<Ptr<ANN_MLP>>

Loads and creates a serialized ANN from a file

Use ANN::save to serialize and store an ANN to disk. Load the ANN from this file again, by calling this function with the path to the file.

§Parameters
  • filepath: path to serialized ANN

Trait Implementations§

Source§

impl ANN_MLPTrait for ANN_MLP

Source§

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

Source§

fn set_train_method( &mut self, method: i32, param1: f64, param2: f64, ) -> Result<()>

Sets training method and common parameters. Read more
Source§

fn set_train_method_def(&mut self, method: i32) -> Result<()>

Sets training method and common parameters. Read more
Source§

fn set_activation_function( &mut self, typ: i32, param1: f64, param2: f64, ) -> Result<()>

Initialize the activation function for each neuron. Currently the default and the only fully supported activation function is ANN_MLP::SIGMOID_SYM. Read more
Source§

fn set_activation_function_def(&mut self, typ: i32) -> Result<()>

Initialize the activation function for each neuron. Currently the default and the only fully supported activation function is ANN_MLP::SIGMOID_SYM. Read more
Source§

fn set_layer_sizes(&mut self, _layer_sizes: &impl ToInputArray) -> Result<()>

Integer vector specifying the number of neurons in each layer including the input and output layers. The very first element specifies the number of elements in the input layer. The last element - number of elements in the output layer. Default value is empty Mat. Read more
Source§

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). Default value is TermCriteria(TermCriteria::MAX_ITER + TermCriteria::EPS, 1000, 0.01). Read more
Source§

fn set_backprop_weight_scale(&mut self, val: f64) -> Result<()>

BPROP: Strength of the weight gradient term. The recommended value is about 0.1. Default value is 0.1. Read more
Source§

fn set_backprop_momentum_scale(&mut self, val: f64) -> Result<()>

BPROP: Strength of the momentum term (the difference between weights on the 2 previous iterations). This parameter provides some inertia to smooth the random fluctuations of the weights. It can vary from 0 (the feature is disabled) to 1 and beyond. The value 0.1 or so is good enough. Default value is 0.1. Read more
Source§

fn set_rprop_dw0(&mut self, val: f64) -> Result<()>

RPROP: Initial value inline formula of update-values inline formula. Default value is 0.1. Read more
Source§

fn set_rprop_dw_plus(&mut self, val: f64) -> Result<()>

RPROP: Increase factor inline formula. It must be >1. Default value is 1.2. Read more
Source§

fn set_rprop_dw_minus(&mut self, val: f64) -> Result<()>

RPROP: Decrease factor inline formula. It must be <1. Default value is 0.5. Read more
Source§

fn set_rprop_dw_min(&mut self, val: f64) -> Result<()>

RPROP: Update-values lower limit inline formula. It must be positive. Default value is FLT_EPSILON. Read more
Source§

fn set_rprop_dw_max(&mut self, val: f64) -> Result<()>

RPROP: Update-values upper limit inline formula. It must be >1. Default value is 50. Read more
Source§

fn set_anneal_initial_t(&mut self, val: f64) -> Result<()>

ANNEAL: Update initial temperature. It must be >=0. Default value is 10. Read more
Source§

fn set_anneal_final_t(&mut self, val: f64) -> Result<()>

ANNEAL: Update final temperature. It must be >=0 and less than initialT. Default value is 0.1. Read more
Source§

fn set_anneal_cooling_ratio(&mut self, val: f64) -> Result<()>

ANNEAL: Update cooling ratio. It must be >0 and less than 1. Default value is 0.95. Read more
Source§

fn set_anneal_ite_per_step(&mut self, val: i32) -> Result<()>

ANNEAL: Update iteration per step. It must be >0 . Default value is 10. Read more
Source§

fn set_anneal_energy_rng(&mut self, rng: &impl RNGTraitConst) -> Result<()>

Set/initialize anneal RNG
Source§

impl ANN_MLPTraitConst for ANN_MLP

Source§

fn as_raw_ANN_MLP(&self) -> *const c_void

Source§

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

Returns current training method
Source§

fn get_layer_sizes(&self) -> Result<Mat>

Integer vector specifying the number of neurons in each layer including the input and output layers. The very first element specifies the number of elements in the input layer. The last element - number of elements in the output layer. Read more
Source§

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). Default value is TermCriteria(TermCriteria::MAX_ITER + TermCriteria::EPS, 1000, 0.01). Read more
Source§

fn get_backprop_weight_scale(&self) -> Result<f64>

BPROP: Strength of the weight gradient term. The recommended value is about 0.1. Default value is 0.1. Read more
Source§

fn get_backprop_momentum_scale(&self) -> Result<f64>

BPROP: Strength of the momentum term (the difference between weights on the 2 previous iterations). This parameter provides some inertia to smooth the random fluctuations of the weights. It can vary from 0 (the feature is disabled) to 1 and beyond. The value 0.1 or so is good enough. Default value is 0.1. Read more
Source§

fn get_rprop_dw0(&self) -> Result<f64>

RPROP: Initial value inline formula of update-values inline formula. Default value is 0.1. Read more
Source§

fn get_rprop_dw_plus(&self) -> Result<f64>

RPROP: Increase factor inline formula. It must be >1. Default value is 1.2. Read more
Source§

fn get_rprop_dw_minus(&self) -> Result<f64>

RPROP: Decrease factor inline formula. It must be <1. Default value is 0.5. Read more
Source§

fn get_rprop_dw_min(&self) -> Result<f64>

RPROP: Update-values lower limit inline formula. It must be positive. Default value is FLT_EPSILON. Read more
Source§

fn get_rprop_dw_max(&self) -> Result<f64>

RPROP: Update-values upper limit inline formula. It must be >1. Default value is 50. Read more
Source§

fn get_anneal_initial_t(&self) -> Result<f64>

ANNEAL: Update initial temperature. It must be >=0. Default value is 10. Read more
Source§

fn get_anneal_final_t(&self) -> Result<f64>

ANNEAL: Update final temperature. It must be >=0 and less than initialT. Default value is 0.1. Read more
Source§

fn get_anneal_cooling_ratio(&self) -> Result<f64>

ANNEAL: Update cooling ratio. It must be >0 and less than 1. Default value is 0.95. Read more
Source§

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

ANNEAL: Update iteration per step. It must be >0 . Default value is 10. Read more
Source§

fn get_weights(&self, layer_idx: i32) -> Result<Mat>

Source§

impl AlgorithmTrait for ANN_MLP

Source§

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

Source§

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

Clears the algorithm state
Source§

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

Reads algorithm parameters from a file storage
Source§

impl AlgorithmTraitConst for ANN_MLP

Source§

fn as_raw_Algorithm(&self) -> *const c_void

Source§

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

Stores algorithm parameters in a file storage
Source§

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

Stores algorithm parameters in a file storage Read more
Source§

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

@deprecated Read more
Source§

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: Read more
Source§

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

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

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).
Source§

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.
Source§

impl Boxed for ANN_MLP

Source§

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

Wrap the specified raw pointer Read more
Source§

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

Return the underlying raw pointer while consuming this wrapper. Read more
Source§

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

Return the underlying raw pointer. Read more
Source§

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

Return the underlying mutable raw pointer Read more
Source§

impl Debug for ANN_MLP

Source§

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

Formats the value using the given formatter. Read more
Source§

impl Drop for ANN_MLP

Source§

fn drop(&mut self)

Executes the destructor for this type. Read more
Source§

impl From<ANN_MLP> for Algorithm

Source§

fn from(s: ANN_MLP) -> Self

Converts to this type from the input type.
Source§

impl From<ANN_MLP> for StatModel

Source§

fn from(s: ANN_MLP) -> Self

Converts to this type from the input type.
Source§

impl StatModelTrait for ANN_MLP

Source§

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

Source§

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

Trains the statistical model Read more
Source§

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

Trains the statistical model Read more
Source§

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

Trains the statistical model Read more
Source§

impl StatModelTraitConst for ANN_MLP

Source§

fn as_raw_StatModel(&self) -> *const c_void

Source§

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

Returns the number of variables in training samples
Source§

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

Source§

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

Returns true if the model is trained
Source§

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

Returns true if the model is classifier
Source§

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

Computes error on the training or test dataset Read more
Source§

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

Predicts response(s) for the provided sample(s) Read more
Source§

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

Predicts response(s) for the provided sample(s) Read more
Source§

impl TryFrom<StatModel> for ANN_MLP

Source§

type Error = Error

The type returned in the event of a conversion error.
Source§

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

Performs the conversion.
Source§

impl Send for ANN_MLP

Auto Trait Implementations§

Blanket Implementations§

Source§

impl<T> Any for T
where T: 'static + ?Sized,

Source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
Source§

impl<T> Borrow<T> for T
where T: ?Sized,

Source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
Source§

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source§

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more
Source§

impl<T> From<T> for T

Source§

fn from(t: T) -> T

Returns the argument unchanged.

Source§

impl<T, U> Into<U> for T
where U: From<T>,

Source§

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source§

impl<Mat> ModifyInplace for Mat
where Mat: Boxed,

Source§

unsafe fn modify_inplace<Res>( &mut self, f: impl FnOnce(&Mat, &mut Mat) -> Res, ) -> Res

Helper function to call OpenCV functions that allow in-place modification of a Mat or another similar object. By passing a mutable reference to the Mat to this function your closure will get called with the read reference and a write references to the same Mat. This is unsafe in a general case as it leads to having non-exclusive mutable access to the internal data, but it can be useful for some performance sensitive operations. One example of an OpenCV function that allows such in-place modification is imgproc::threshold. Read more
Source§

impl<T, U> TryFrom<U> for T
where U: Into<T>,

Source§

type Error = Infallible

The type returned in the event of a conversion error.
Source§

fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

Performs the conversion.
Source§

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

Source§

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.
Source§

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.