work/smartcore
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

make default params available to serde (#167)

Browse Source 
* add seed param to search params

* make default params available to serde

* lints

* create defaults for enums

* lint
This commit is contained in:
Montana Low
2022-09-21 19:48:31 -07:00
committed by GitHub
parent 403d3f2348
commit 764309e313
22 changed files with 175 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/algorithm/neighbour/mod.rs
+6
View File
@@ -59,6 +59,12 @@ pub enum KNNAlgorithmName {
CoverTree, CoverTree,
} }
impl Default for KNNAlgorithmName {
fn default() -> Self {
KNNAlgorithmName::CoverTree
}
}
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug)] #[derive(Debug)]
pub(crate) enum KNNAlgorithm<T: RealNumber, D: Distance<Vec<T>, T>> { pub(crate) enum KNNAlgorithm<T: RealNumber, D: Distance<Vec<T>, T>> {
src/cluster/dbscan.rs
+10 -1
View File
@@ -65,17 +65,22 @@ pub struct DBSCAN<T: RealNumber, D: Distance<Vec<T>, T>> {
eps: T, eps: T,
} }
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// DBSCAN clustering algorithm parameters /// DBSCAN clustering algorithm parameters
pub struct DBSCANParameters<T: RealNumber, D: Distance<Vec<T>, T>> { pub struct DBSCANParameters<T: RealNumber, D: Distance<Vec<T>, T>> {
#[cfg_attr(feature = "serde", serde(default))]
/// a function that defines a distance between each pair of point in training data. /// a function that defines a distance between each pair of point in training data.
/// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait. /// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait.
/// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions. /// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions.
pub distance: D, pub distance: D,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of samples (or total weight) in a neighborhood for a point to be considered as a core point. /// The number of samples (or total weight) in a neighborhood for a point to be considered as a core point.
pub min_samples: usize, pub min_samples: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum distance between two samples for one to be considered as in the neighborhood of the other. /// The maximum distance between two samples for one to be considered as in the neighborhood of the other.
pub eps: T, pub eps: T,
#[cfg_attr(feature = "serde", serde(default))]
/// KNN algorithm to use. /// KNN algorithm to use.
pub algorithm: KNNAlgorithmName, pub algorithm: KNNAlgorithmName,
} }
@@ -113,14 +118,18 @@ impl<T: RealNumber, D: Distance<Vec<T>, T>> DBSCANParameters<T, D> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct DBSCANSearchParameters<T: RealNumber, D: Distance<Vec<T>, T>> { pub struct DBSCANSearchParameters<T: RealNumber, D: Distance<Vec<T>, T>> {
#[cfg_attr(feature = "serde", serde(default))]
/// a function that defines a distance between each pair of point in training data. /// a function that defines a distance between each pair of point in training data.
/// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait. /// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait.
/// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions. /// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions.
pub distance: Vec<D>, pub distance: Vec<D>,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of samples (or total weight) in a neighborhood for a point to be considered as a core point. /// The number of samples (or total weight) in a neighborhood for a point to be considered as a core point.
pub min_samples: Vec<usize>, pub min_samples: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum distance between two samples for one to be considered as in the neighborhood of the other. /// The maximum distance between two samples for one to be considered as in the neighborhood of the other.
pub eps: Vec<T>, pub eps: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// KNN algorithm to use. /// KNN algorithm to use.
pub algorithm: Vec<KNNAlgorithmName>, pub algorithm: Vec<KNNAlgorithmName>,
} }
@@ -221,7 +230,7 @@ impl<T: RealNumber> Default for DBSCANParameters<T, Euclidian> {
distance: Distances::euclidian(), distance: Distances::euclidian(),
min_samples: 5, min_samples: 5,
eps: T::half(), eps: T::half(),
algorithm: KNNAlgorithmName::CoverTree, algorithm: KNNAlgorithmName::default(),
} }
} }
} }
src/cluster/kmeans.rs
+7
View File
@@ -102,13 +102,17 @@ impl<T: RealNumber> PartialEq for KMeans<T> {
} }
} }
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// K-Means clustering algorithm parameters /// K-Means clustering algorithm parameters
pub struct KMeansParameters { pub struct KMeansParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of clusters. /// Number of clusters.
pub k: usize, pub k: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Maximum number of iterations of the k-means algorithm for a single run. /// Maximum number of iterations of the k-means algorithm for a single run.
pub max_iter: usize, pub max_iter: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Determines random number generation for centroid initialization. /// Determines random number generation for centroid initialization.
/// Use an int to make the randomness deterministic /// Use an int to make the randomness deterministic
pub seed: Option<u64>, pub seed: Option<u64>,
@@ -141,10 +145,13 @@ impl Default for KMeansParameters {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct KMeansSearchParameters { pub struct KMeansSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of clusters. /// Number of clusters.
pub k: Vec<usize>, pub k: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Maximum number of iterations of the k-means algorithm for a single run. /// Maximum number of iterations of the k-means algorithm for a single run.
pub max_iter: Vec<usize>, pub max_iter: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Determines random number generation for centroid initialization. /// Determines random number generation for centroid initialization.
/// Use an int to make the randomness deterministic /// Use an int to make the randomness deterministic
pub seed: Vec<Option<u64>>, pub seed: Vec<Option<u64>>,
src/decomposition/pca.rs
+5
View File
@@ -83,11 +83,14 @@ impl<T: RealNumber, M: Matrix<T>> PartialEq for PCA<T, M> {
} }
} }
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// PCA parameters /// PCA parameters
pub struct PCAParameters { pub struct PCAParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of components to keep. /// Number of components to keep.
pub n_components: usize, pub n_components: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// By default, covariance matrix is used to compute principal components. /// By default, covariance matrix is used to compute principal components.
/// Enable this flag if you want to use correlation matrix instead. /// Enable this flag if you want to use correlation matrix instead.
pub use_correlation_matrix: bool, pub use_correlation_matrix: bool,
@@ -120,8 +123,10 @@ impl Default for PCAParameters {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct PCASearchParameters { pub struct PCASearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of components to keep. /// Number of components to keep.
pub n_components: Vec<usize>, pub n_components: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// By default, covariance matrix is used to compute principal components. /// By default, covariance matrix is used to compute principal components.
/// Enable this flag if you want to use correlation matrix instead. /// Enable this flag if you want to use correlation matrix instead.
pub use_correlation_matrix: Vec<bool>, pub use_correlation_matrix: Vec<bool>,
src/decomposition/svd.rs
+3
View File
@@ -69,9 +69,11 @@ impl<T: RealNumber, M: Matrix<T>> PartialEq for SVD<T, M> {
} }
} }
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// SVD parameters /// SVD parameters
pub struct SVDParameters { pub struct SVDParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of components to keep. /// Number of components to keep.
pub n_components: usize, pub n_components: usize,
} }
@@ -94,6 +96,7 @@ impl SVDParameters {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct SVDSearchParameters { pub struct SVDSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Maximum number of iterations of the k-means algorithm for a single run. /// Maximum number of iterations of the k-means algorithm for a single run.
pub n_components: Vec<usize>, pub n_components: Vec<usize>,
} }
src/ensemble/random_forest_classifier.rs
+16
View File
@@ -67,20 +67,28 @@ use crate::tree::decision_tree_classifier::{
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct RandomForestClassifierParameters { pub struct RandomForestClassifierParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Split criteria to use when building a tree. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// Split criteria to use when building a tree. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub criterion: SplitCriterion, pub criterion: SplitCriterion,
#[cfg_attr(feature = "serde", serde(default))]
/// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub max_depth: Option<u16>, pub max_depth: Option<u16>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_leaf: usize, pub min_samples_leaf: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_split: usize, pub min_samples_split: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of trees in the forest. /// The number of trees in the forest.
pub n_trees: u16, pub n_trees: u16,
#[cfg_attr(feature = "serde", serde(default))]
/// Number of random sample of predictors to use as split candidates. /// Number of random sample of predictors to use as split candidates.
pub m: Option<usize>, pub m: Option<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Whether to keep samples used for tree generation. This is required for OOB prediction. /// Whether to keep samples used for tree generation. This is required for OOB prediction.
pub keep_samples: bool, pub keep_samples: bool,
#[cfg_attr(feature = "serde", serde(default))]
/// Seed used for bootstrap sampling and feature selection for each tree. /// Seed used for bootstrap sampling and feature selection for each tree.
pub seed: u64, pub seed: u64,
} }
@@ -198,20 +206,28 @@ impl<T: RealNumber, M: Matrix<T>> Predictor<M, M::RowVector> for RandomForestCla
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct RandomForestClassifierSearchParameters { pub struct RandomForestClassifierSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Split criteria to use when building a tree. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// Split criteria to use when building a tree. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub criterion: Vec<SplitCriterion>, pub criterion: Vec<SplitCriterion>,
#[cfg_attr(feature = "serde", serde(default))]
/// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub max_depth: Vec<Option<u16>>, pub max_depth: Vec<Option<u16>>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_leaf: Vec<usize>, pub min_samples_leaf: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_split: Vec<usize>, pub min_samples_split: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of trees in the forest. /// The number of trees in the forest.
pub n_trees: Vec<u16>, pub n_trees: Vec<u16>,
#[cfg_attr(feature = "serde", serde(default))]
/// Number of random sample of predictors to use as split candidates. /// Number of random sample of predictors to use as split candidates.
pub m: Vec<Option<usize>>, pub m: Vec<Option<usize>>,
#[cfg_attr(feature = "serde", serde(default))]
/// Whether to keep samples used for tree generation. This is required for OOB prediction. /// Whether to keep samples used for tree generation. This is required for OOB prediction.
pub keep_samples: Vec<bool>, pub keep_samples: Vec<bool>,
#[cfg_attr(feature = "serde", serde(default))]
/// Seed used for bootstrap sampling and feature selection for each tree. /// Seed used for bootstrap sampling and feature selection for each tree.
pub seed: Vec<u64>, pub seed: Vec<u64>,
} }
src/ensemble/random_forest_regressor.rs
+14
View File
@@ -65,18 +65,25 @@ use crate::tree::decision_tree_regressor::{
/// Parameters of the Random Forest Regressor /// Parameters of the Random Forest Regressor
/// Some parameters here are passed directly into base estimator. /// Some parameters here are passed directly into base estimator.
pub struct RandomForestRegressorParameters { pub struct RandomForestRegressorParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Tree max depth. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// Tree max depth. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub max_depth: Option<u16>, pub max_depth: Option<u16>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// The minimum number of samples required to be at a leaf node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub min_samples_leaf: usize, pub min_samples_leaf: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// The minimum number of samples required to split an internal node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub min_samples_split: usize, pub min_samples_split: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of trees in the forest. /// The number of trees in the forest.
pub n_trees: usize, pub n_trees: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Number of random sample of predictors to use as split candidates. /// Number of random sample of predictors to use as split candidates.
pub m: Option<usize>, pub m: Option<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Whether to keep samples used for tree generation. This is required for OOB prediction. /// Whether to keep samples used for tree generation. This is required for OOB prediction.
pub keep_samples: bool, pub keep_samples: bool,
#[cfg_attr(feature = "serde", serde(default))]
/// Seed used for bootstrap sampling and feature selection for each tree. /// Seed used for bootstrap sampling and feature selection for each tree.
pub seed: u64, pub seed: u64,
} }
@@ -181,18 +188,25 @@ impl<T: RealNumber, M: Matrix<T>> Predictor<M, M::RowVector> for RandomForestReg
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct RandomForestRegressorSearchParameters { pub struct RandomForestRegressorSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// Tree max depth. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub max_depth: Vec<Option<u16>>, pub max_depth: Vec<Option<u16>>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to be at a leaf node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_leaf: Vec<usize>, pub min_samples_leaf: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html) /// The minimum number of samples required to split an internal node. See [Decision Tree Classifier](../../tree/decision_tree_classifier/index.html)
pub min_samples_split: Vec<usize>, pub min_samples_split: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The number of trees in the forest. /// The number of trees in the forest.
pub n_trees: Vec<usize>, pub n_trees: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Number of random sample of predictors to use as split candidates. /// Number of random sample of predictors to use as split candidates.
pub m: Vec<Option<usize>>, pub m: Vec<Option<usize>>,
#[cfg_attr(feature = "serde", serde(default))]
/// Whether to keep samples used for tree generation. This is required for OOB prediction. /// Whether to keep samples used for tree generation. This is required for OOB prediction.
pub keep_samples: Vec<bool>, pub keep_samples: Vec<bool>,
#[cfg_attr(feature = "serde", serde(default))]
/// Seed used for bootstrap sampling and feature selection for each tree. /// Seed used for bootstrap sampling and feature selection for each tree.
pub seed: Vec<u64>, pub seed: Vec<u64>,
} }
src/linear/elastic_net.rs
+10
View File
@@ -71,16 +71,21 @@ use crate::linear::lasso_optimizer::InteriorPointOptimizer;
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct ElasticNetParameters<T: RealNumber> { pub struct ElasticNetParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub alpha: T, pub alpha: T,
#[cfg_attr(feature = "serde", serde(default))]
/// The elastic net mixing parameter, with 0 <= l1_ratio <= 1. /// The elastic net mixing parameter, with 0 <= l1_ratio <= 1.
/// For l1_ratio = 0 the penalty is an L2 penalty. /// For l1_ratio = 0 the penalty is an L2 penalty.
/// For l1_ratio = 1 it is an L1 penalty. For 0 < l1_ratio < 1, the penalty is a combination of L1 and L2. /// For l1_ratio = 1 it is an L1 penalty. For 0 < l1_ratio < 1, the penalty is a combination of L1 and L2.
pub l1_ratio: T, pub l1_ratio: T,
#[cfg_attr(feature = "serde", serde(default))]
/// If True, the regressors X will be normalized before regression by subtracting the mean and dividing by the standard deviation. /// If True, the regressors X will be normalized before regression by subtracting the mean and dividing by the standard deviation.
pub normalize: bool, pub normalize: bool,
#[cfg_attr(feature = "serde", serde(default))]
/// The tolerance for the optimization /// The tolerance for the optimization
pub tol: T, pub tol: T,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum number of iterations /// The maximum number of iterations
pub max_iter: usize, pub max_iter: usize,
} }
@@ -139,16 +144,21 @@ impl<T: RealNumber> Default for ElasticNetParameters<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct ElasticNetSearchParameters<T: RealNumber> { pub struct ElasticNetSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub alpha: Vec<T>, pub alpha: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// The elastic net mixing parameter, with 0 <= l1_ratio <= 1. /// The elastic net mixing parameter, with 0 <= l1_ratio <= 1.
/// For l1_ratio = 0 the penalty is an L2 penalty. /// For l1_ratio = 0 the penalty is an L2 penalty.
/// For l1_ratio = 1 it is an L1 penalty. For 0 < l1_ratio < 1, the penalty is a combination of L1 and L2. /// For l1_ratio = 1 it is an L1 penalty. For 0 < l1_ratio < 1, the penalty is a combination of L1 and L2.
pub l1_ratio: Vec<T>, pub l1_ratio: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// If True, the regressors X will be normalized before regression by subtracting the mean and dividing by the standard deviation. /// If True, the regressors X will be normalized before regression by subtracting the mean and dividing by the standard deviation.
pub normalize: Vec<bool>, pub normalize: Vec<bool>,
#[cfg_attr(feature = "serde", serde(default))]
/// The tolerance for the optimization /// The tolerance for the optimization
pub tol: Vec<T>, pub tol: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum number of iterations /// The maximum number of iterations
pub max_iter: Vec<usize>, pub max_iter: Vec<usize>,
} }
src/linear/lasso.rs
+8
View File
@@ -38,13 +38,17 @@ use crate::math::num::RealNumber;
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct LassoParameters<T: RealNumber> { pub struct LassoParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the strength of the penalty to the loss function. /// Controls the strength of the penalty to the loss function.
pub alpha: T, pub alpha: T,
#[cfg_attr(feature = "serde", serde(default))]
/// If true the regressors X will be normalized before regression /// If true the regressors X will be normalized before regression
/// by subtracting the mean and dividing by the standard deviation. /// by subtracting the mean and dividing by the standard deviation.
pub normalize: bool, pub normalize: bool,
#[cfg_attr(feature = "serde", serde(default))]
/// The tolerance for the optimization /// The tolerance for the optimization
pub tol: T, pub tol: T,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum number of iterations /// The maximum number of iterations
pub max_iter: usize, pub max_iter: usize,
} }
@@ -116,13 +120,17 @@ impl<T: RealNumber, M: Matrix<T>> Predictor<M, M::RowVector> for Lasso<T, M> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct LassoSearchParameters<T: RealNumber> { pub struct LassoSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the strength of the penalty to the loss function. /// Controls the strength of the penalty to the loss function.
pub alpha: Vec<T>, pub alpha: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// If true the regressors X will be normalized before regression /// If true the regressors X will be normalized before regression
/// by subtracting the mean and dividing by the standard deviation. /// by subtracting the mean and dividing by the standard deviation.
pub normalize: Vec<bool>, pub normalize: Vec<bool>,
#[cfg_attr(feature = "serde", serde(default))]
/// The tolerance for the optimization /// The tolerance for the optimization
pub tol: Vec<T>, pub tol: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum number of iterations /// The maximum number of iterations
pub max_iter: Vec<usize>, pub max_iter: Vec<usize>,
} }
src/linear/linear_regression.rs
+9 -10
View File
@@ -71,19 +71,21 @@ use crate::linalg::Matrix;
use crate::math::num::RealNumber; use crate::math::num::RealNumber;
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone, Eq, PartialEq)] #[derive(Debug, Default, Clone, Eq, PartialEq)]
/// Approach to use for estimation of regression coefficients. QR is more efficient but SVD is more stable. /// Approach to use for estimation of regression coefficients. QR is more efficient but SVD is more stable.
pub enum LinearRegressionSolverName { pub enum LinearRegressionSolverName {
/// QR decomposition, see [QR](../../linalg/qr/index.html) /// QR decomposition, see [QR](../../linalg/qr/index.html)
QR, QR,
#[default]
/// SVD decomposition, see [SVD](../../linalg/svd/index.html) /// SVD decomposition, see [SVD](../../linalg/svd/index.html)
SVD, SVD,
} }
/// Linear Regression parameters /// Linear Regression parameters
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Default, Clone)]
pub struct LinearRegressionParameters { pub struct LinearRegressionParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Solver to use for estimation of regression coefficients. /// Solver to use for estimation of regression coefficients.
pub solver: LinearRegressionSolverName, pub solver: LinearRegressionSolverName,
} }
@@ -105,18 +107,11 @@ impl LinearRegressionParameters {
} }
} }
impl Default for LinearRegressionParameters {
fn default() -> Self {
LinearRegressionParameters {
solver: LinearRegressionSolverName::SVD,
}
}
}
/// Linear Regression grid search parameters /// Linear Regression grid search parameters
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct LinearRegressionSearchParameters { pub struct LinearRegressionSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Solver to use for estimation of regression coefficients. /// Solver to use for estimation of regression coefficients.
pub solver: Vec<LinearRegressionSolverName>, pub solver: Vec<LinearRegressionSolverName>,
} }
@@ -353,5 +348,9 @@ mod tests {
serde_json::from_str(&serde_json::to_string(&lr).unwrap()).unwrap(); serde_json::from_str(&serde_json::to_string(&lr).unwrap()).unwrap();
assert_eq!(lr, deserialized_lr); assert_eq!(lr, deserialized_lr);
let default = LinearRegressionParameters::default();
let parameters: LinearRegressionParameters = serde_json::from_str("{}").unwrap();
assert_eq!(parameters.solver, default.solver);
} }
} }
src/linear/logistic_regression.rs
+11 -1
View File
@@ -75,12 +75,20 @@ pub enum LogisticRegressionSolverName {
LBFGS, LBFGS,
} }
impl Default for LogisticRegressionSolverName {
fn default() -> Self {
LogisticRegressionSolverName::LBFGS
}
}
/// Logistic Regression parameters /// Logistic Regression parameters
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct LogisticRegressionParameters<T: RealNumber> { pub struct LogisticRegressionParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Solver to use for estimation of regression coefficients. /// Solver to use for estimation of regression coefficients.
pub solver: LogisticRegressionSolverName, pub solver: LogisticRegressionSolverName,
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub alpha: T, pub alpha: T,
} }
@@ -89,8 +97,10 @@ pub struct LogisticRegressionParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct LogisticRegressionSearchParameters<T: RealNumber> { pub struct LogisticRegressionSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Solver to use for estimation of regression coefficients. /// Solver to use for estimation of regression coefficients.
pub solver: Vec<LogisticRegressionSolverName>, pub solver: Vec<LogisticRegressionSolverName>,
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub alpha: Vec<T>, pub alpha: Vec<T>,
} }
@@ -204,7 +214,7 @@ impl<T: RealNumber> LogisticRegressionParameters<T> {
impl<T: RealNumber> Default for LogisticRegressionParameters<T> { impl<T: RealNumber> Default for LogisticRegressionParameters<T> {
fn default() -> Self { fn default() -> Self {
LogisticRegressionParameters { LogisticRegressionParameters {
solver: LogisticRegressionSolverName::LBFGS, solver: LogisticRegressionSolverName::default(),
alpha: T::zero(), alpha: T::zero(),
} }
} }
src/linear/ridge_regression.rs
+10 -1
View File
@@ -77,6 +77,12 @@ pub enum RidgeRegressionSolverName {
SVD, SVD,
} }
impl Default for RidgeRegressionSolverName {
fn default() -> Self {
RidgeRegressionSolverName::Cholesky
}
}
/// Ridge Regression parameters /// Ridge Regression parameters
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
@@ -94,10 +100,13 @@ pub struct RidgeRegressionParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct RidgeRegressionSearchParameters<T: RealNumber> { pub struct RidgeRegressionSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Solver to use for estimation of regression coefficients. /// Solver to use for estimation of regression coefficients.
pub solver: Vec<RidgeRegressionSolverName>, pub solver: Vec<RidgeRegressionSolverName>,
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub alpha: Vec<T>, pub alpha: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// If true the regressors X will be normalized before regression /// If true the regressors X will be normalized before regression
/// by subtracting the mean and dividing by the standard deviation. /// by subtracting the mean and dividing by the standard deviation.
pub normalize: Vec<bool>, pub normalize: Vec<bool>,
@@ -204,7 +213,7 @@ impl<T: RealNumber> RidgeRegressionParameters<T> {
impl<T: RealNumber> Default for RidgeRegressionParameters<T> { impl<T: RealNumber> Default for RidgeRegressionParameters<T> {
fn default() -> Self { fn default() -> Self {
RidgeRegressionParameters { RidgeRegressionParameters {
solver: RidgeRegressionSolverName::Cholesky, solver: RidgeRegressionSolverName::default(),
alpha: T::one(), alpha: T::one(),
normalize: true, normalize: true,
} }
src/naive_bayes/bernoulli.rs
+6
View File
@@ -114,10 +114,13 @@ impl<T: RealNumber, M: Matrix<T>> NBDistribution<T, M> for BernoulliNBDistributi
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct BernoulliNBParameters<T: RealNumber> { pub struct BernoulliNBParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: T, pub alpha: T,
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Option<Vec<T>>, pub priors: Option<Vec<T>>,
#[cfg_attr(feature = "serde", serde(default))]
/// Threshold for binarizing (mapping to booleans) of sample features. If None, input is presumed to already consist of binary vectors. /// Threshold for binarizing (mapping to booleans) of sample features. If None, input is presumed to already consist of binary vectors.
pub binarize: Option<T>, pub binarize: Option<T>,
} }
@@ -154,10 +157,13 @@ impl<T: RealNumber> Default for BernoulliNBParameters<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct BernoulliNBSearchParameters<T: RealNumber> { pub struct BernoulliNBSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: Vec<T>, pub alpha: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Vec<Option<Vec<T>>>, pub priors: Vec<Option<Vec<T>>>,
#[cfg_attr(feature = "serde", serde(default))]
/// Threshold for binarizing (mapping to booleans) of sample features. If None, input is presumed to already consist of binary vectors. /// Threshold for binarizing (mapping to booleans) of sample features. If None, input is presumed to already consist of binary vectors.
pub binarize: Vec<Option<T>>, pub binarize: Vec<Option<T>>,
} }
src/naive_bayes/categorical.rs
+2
View File
@@ -243,6 +243,7 @@ impl<T: RealNumber> CategoricalNBDistribution<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct CategoricalNBParameters<T: RealNumber> { pub struct CategoricalNBParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: T, pub alpha: T,
} }
@@ -265,6 +266,7 @@ impl<T: RealNumber> Default for CategoricalNBParameters<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct CategoricalNBSearchParameters<T: RealNumber> { pub struct CategoricalNBSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: Vec<T>, pub alpha: Vec<T>,
} }
src/naive_bayes/gaussian.rs
+2
View File
@@ -78,6 +78,7 @@ impl<T: RealNumber, M: Matrix<T>> NBDistribution<T, M> for GaussianNBDistributio
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct GaussianNBParameters<T: RealNumber> { pub struct GaussianNBParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Option<Vec<T>>, pub priors: Option<Vec<T>>,
} }
@@ -100,6 +101,7 @@ impl<T: RealNumber> Default for GaussianNBParameters<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct GaussianNBSearchParameters<T: RealNumber> { pub struct GaussianNBSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Vec<Option<Vec<T>>>, pub priors: Vec<Option<Vec<T>>>,
} }
src/naive_bayes/multinomial.rs
+4
View File
@@ -86,8 +86,10 @@ impl<T: RealNumber, M: Matrix<T>> NBDistribution<T, M> for MultinomialNBDistribu
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct MultinomialNBParameters<T: RealNumber> { pub struct MultinomialNBParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: T, pub alpha: T,
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Option<Vec<T>>, pub priors: Option<Vec<T>>,
} }
@@ -118,8 +120,10 @@ impl<T: RealNumber> Default for MultinomialNBParameters<T> {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct MultinomialNBSearchParameters<T: RealNumber> { pub struct MultinomialNBSearchParameters<T: RealNumber> {
#[cfg_attr(feature = "serde", serde(default))]
/// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing). /// Additive (Laplace/Lidstone) smoothing parameter (0 for no smoothing).
pub alpha: Vec<T>, pub alpha: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// Prior probabilities of the classes. If specified the priors are not adjusted according to the data /// Prior probabilities of the classes. If specified the priors are not adjusted according to the data
pub priors: Vec<Option<Vec<T>>>, pub priors: Vec<Option<Vec<T>>>,
} }
src/neighbors/knn_classifier.rs
+7 -2
View File
@@ -49,16 +49,21 @@ use crate::neighbors::KNNWeightFunction;
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct KNNClassifierParameters<T: RealNumber, D: Distance<Vec<T>, T>> { pub struct KNNClassifierParameters<T: RealNumber, D: Distance<Vec<T>, T>> {
#[cfg_attr(feature = "serde", serde(default))]
/// a function that defines a distance between each pair of point in training data. /// a function that defines a distance between each pair of point in training data.
/// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait. /// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait.
/// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions. /// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions.
pub distance: D, pub distance: D,
#[cfg_attr(feature = "serde", serde(default))]
/// backend search algorithm. See [`knn search algorithms`](../../algorithm/neighbour/index.html). `CoverTree` is default. /// backend search algorithm. See [`knn search algorithms`](../../algorithm/neighbour/index.html). `CoverTree` is default.
pub algorithm: KNNAlgorithmName, pub algorithm: KNNAlgorithmName,
#[cfg_attr(feature = "serde", serde(default))]
/// weighting function that is used to calculate estimated class value. Default function is `KNNWeightFunction::Uniform`. /// weighting function that is used to calculate estimated class value. Default function is `KNNWeightFunction::Uniform`.
pub weight: KNNWeightFunction, pub weight: KNNWeightFunction,
#[cfg_attr(feature = "serde", serde(default))]
/// number of training samples to consider when estimating class for new point. Default value is 3. /// number of training samples to consider when estimating class for new point. Default value is 3.
pub k: usize, pub k: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// this parameter is not used /// this parameter is not used
t: PhantomData<T>, t: PhantomData<T>,
} }
@@ -111,8 +116,8 @@ impl<T: RealNumber> Default for KNNClassifierParameters<T, Euclidian> {
fn default() -> Self { fn default() -> Self {
KNNClassifierParameters { KNNClassifierParameters {
distance: Distances::euclidian(), distance: Distances::euclidian(),
algorithm: KNNAlgorithmName::CoverTree, algorithm: KNNAlgorithmName::default(),
weight: KNNWeightFunction::Uniform, weight: KNNWeightFunction::default(),
k: 3, k: 3,
t: PhantomData, t: PhantomData,
} }
src/neighbors/knn_regressor.rs
+7 -2
View File
@@ -52,16 +52,21 @@ use crate::neighbors::KNNWeightFunction;
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct KNNRegressorParameters<T: RealNumber, D: Distance<Vec<T>, T>> { pub struct KNNRegressorParameters<T: RealNumber, D: Distance<Vec<T>, T>> {
#[cfg_attr(feature = "serde", serde(default))]
/// a function that defines a distance between each pair of point in training data. /// a function that defines a distance between each pair of point in training data.
/// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait. /// This function should extend [`Distance`](../../math/distance/trait.Distance.html) trait.
/// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions. /// See [`Distances`](../../math/distance/struct.Distances.html) for a list of available functions.
distance: D, distance: D,
#[cfg_attr(feature = "serde", serde(default))]
/// backend search algorithm. See [`knn search algorithms`](../../algorithm/neighbour/index.html). `CoverTree` is default. /// backend search algorithm. See [`knn search algorithms`](../../algorithm/neighbour/index.html). `CoverTree` is default.
pub algorithm: KNNAlgorithmName, pub algorithm: KNNAlgorithmName,
#[cfg_attr(feature = "serde", serde(default))]
/// weighting function that is used to calculate estimated class value. Default function is `KNNWeightFunction::Uniform`. /// weighting function that is used to calculate estimated class value. Default function is `KNNWeightFunction::Uniform`.
pub weight: KNNWeightFunction, pub weight: KNNWeightFunction,
#[cfg_attr(feature = "serde", serde(default))]
/// number of training samples to consider when estimating class for new point. Default value is 3. /// number of training samples to consider when estimating class for new point. Default value is 3.
pub k: usize, pub k: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// this parameter is not used /// this parameter is not used
t: PhantomData<T>, t: PhantomData<T>,
} }
@@ -113,8 +118,8 @@ impl<T: RealNumber> Default for KNNRegressorParameters<T, Euclidian> {
fn default() -> Self { fn default() -> Self {
KNNRegressorParameters { KNNRegressorParameters {
distance: Distances::euclidian(), distance: Distances::euclidian(),
algorithm: KNNAlgorithmName::CoverTree, algorithm: KNNAlgorithmName::default(),
weight: KNNWeightFunction::Uniform, weight: KNNWeightFunction::default(),
k: 3, k: 3,
t: PhantomData, t: PhantomData,
} }
src/neighbors/mod.rs
+6
View File
@@ -58,6 +58,12 @@ pub enum KNNWeightFunction {
Distance, Distance,
} }
impl Default for KNNWeightFunction {
fn default() -> Self {
KNNWeightFunction::Uniform
}
}
impl KNNWeightFunction { impl KNNWeightFunction {
fn calc_weights<T: RealNumber>(&self, distances: Vec<T>) -> std::vec::Vec<T> { fn calc_weights<T: RealNumber>(&self, distances: Vec<T>) -> std::vec::Vec<T> {
match *self { match *self {
src/svm/svc.rs
+12
View File
@@ -91,16 +91,22 @@ use crate::svm::{Kernel, Kernels, LinearKernel};
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// SVC Parameters /// SVC Parameters
pub struct SVCParameters<T: RealNumber, M: Matrix<T>, K: Kernel<T, M::RowVector>> { pub struct SVCParameters<T: RealNumber, M: Matrix<T>, K: Kernel<T, M::RowVector>> {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of epochs. /// Number of epochs.
pub epoch: usize, pub epoch: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub c: T, pub c: T,
#[cfg_attr(feature = "serde", serde(default))]
/// Tolerance for stopping criterion. /// Tolerance for stopping criterion.
pub tol: T, pub tol: T,
#[cfg_attr(feature = "serde", serde(default))]
/// The kernel function. /// The kernel function.
pub kernel: K, pub kernel: K,
#[cfg_attr(feature = "serde", serde(default))]
/// Unused parameter. /// Unused parameter.
m: PhantomData<M>, m: PhantomData<M>,
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the pseudo random number generation for shuffling the data for probability estimates /// Controls the pseudo random number generation for shuffling the data for probability estimates
seed: Option<u64>, seed: Option<u64>,
} }
@@ -109,16 +115,22 @@ pub struct SVCParameters<T: RealNumber, M: Matrix<T>, K: Kernel<T, M::RowVector>
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct SVCSearchParameters<T: RealNumber, M: Matrix<T>, K: Kernel<T, M::RowVector>> { pub struct SVCSearchParameters<T: RealNumber, M: Matrix<T>, K: Kernel<T, M::RowVector>> {
#[cfg_attr(feature = "serde", serde(default))]
/// Number of epochs. /// Number of epochs.
pub epoch: Vec<usize>, pub epoch: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Regularization parameter. /// Regularization parameter.
pub c: Vec<T>, pub c: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// Tolerance for stopping epoch. /// Tolerance for stopping epoch.
pub tol: Vec<T>, pub tol: Vec<T>,
#[cfg_attr(feature = "serde", serde(default))]
/// The kernel function. /// The kernel function.
pub kernel: Vec<K>, pub kernel: Vec<K>,
#[cfg_attr(feature = "serde", serde(default))]
/// Unused parameter. /// Unused parameter.
m: PhantomData<M>, m: PhantomData<M>,
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the pseudo random number generation for shuffling the data for probability estimates /// Controls the pseudo random number generation for shuffling the data for probability estimates
seed: Vec<Option<u64>>, seed: Vec<Option<u64>>,
} }
src/tree/decision_tree_classifier.rs
+12 -1
View File
@@ -83,14 +83,19 @@ use crate::rand::get_rng_impl;
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// Parameters of Decision Tree /// Parameters of Decision Tree
pub struct DecisionTreeClassifierParameters { pub struct DecisionTreeClassifierParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Split criteria to use when building a tree. /// Split criteria to use when building a tree.
pub criterion: SplitCriterion, pub criterion: SplitCriterion,
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum depth of the tree. /// The maximum depth of the tree.
pub max_depth: Option<u16>, pub max_depth: Option<u16>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. /// The minimum number of samples required to be at a leaf node.
pub min_samples_leaf: usize, pub min_samples_leaf: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. /// The minimum number of samples required to split an internal node.
pub min_samples_split: usize, pub min_samples_split: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the randomness of the estimator /// Controls the randomness of the estimator
pub seed: Option<u64>, pub seed: Option<u64>,
} }
@@ -118,6 +123,12 @@ pub enum SplitCriterion {
ClassificationError, ClassificationError,
} }
impl Default for SplitCriterion {
fn default() -> Self {
SplitCriterion::Gini
}
}
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug)] #[derive(Debug)]
struct Node<T: RealNumber> { struct Node<T: RealNumber> {
@@ -196,7 +207,7 @@ impl DecisionTreeClassifierParameters {
impl Default for DecisionTreeClassifierParameters { impl Default for DecisionTreeClassifierParameters {
fn default() -> Self { fn default() -> Self {
DecisionTreeClassifierParameters { DecisionTreeClassifierParameters {
criterion: SplitCriterion::Gini, criterion: SplitCriterion::default(),
max_depth: None, max_depth: None,
min_samples_leaf: 1, min_samples_leaf: 1,
min_samples_split: 2, min_samples_split: 2,
src/tree/decision_tree_regressor.rs
+8
View File
@@ -78,12 +78,16 @@ use crate::rand::get_rng_impl;
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
/// Parameters of Regression Tree /// Parameters of Regression Tree
pub struct DecisionTreeRegressorParameters { pub struct DecisionTreeRegressorParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// The maximum depth of the tree. /// The maximum depth of the tree.
pub max_depth: Option<u16>, pub max_depth: Option<u16>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. /// The minimum number of samples required to be at a leaf node.
pub min_samples_leaf: usize, pub min_samples_leaf: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. /// The minimum number of samples required to split an internal node.
pub min_samples_split: usize, pub min_samples_split: usize,
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the randomness of the estimator /// Controls the randomness of the estimator
pub seed: Option<u64>, pub seed: Option<u64>,
} }
@@ -142,12 +146,16 @@ impl Default for DecisionTreeRegressorParameters {
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct DecisionTreeRegressorSearchParameters { pub struct DecisionTreeRegressorSearchParameters {
#[cfg_attr(feature = "serde", serde(default))]
/// Tree max depth. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// Tree max depth. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub max_depth: Vec<Option<u16>>, pub max_depth: Vec<Option<u16>>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to be at a leaf node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// The minimum number of samples required to be at a leaf node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub min_samples_leaf: Vec<usize>, pub min_samples_leaf: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// The minimum number of samples required to split an internal node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html) /// The minimum number of samples required to split an internal node. See [Decision Tree Regressor](../../tree/decision_tree_regressor/index.html)
pub min_samples_split: Vec<usize>, pub min_samples_split: Vec<usize>,
#[cfg_attr(feature = "serde", serde(default))]
/// Controls the randomness of the estimator /// Controls the randomness of the estimator
pub seed: Vec<Option<u64>>, pub seed: Vec<Option<u64>>,
} }