create defaults for enums

Author: Montana Low

src/algorithm/neighbour/mod.rs

@@ -59,6 +59,12 @@ pub enum KNNAlgorithmName {
     CoverTree,
 }
 
+impl Default for KNNAlgorithmName {
+    fn default() -> Self {
+        KNNAlgorithmName::CoverTree
+    }
+}
+
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug)]
 pub(crate) enum KNNAlgorithm<T: RealNumber, D: Distance<Vec<T>, T>> {

src/cluster/dbscan.rs

@@ -65,6 +65,7 @@ pub struct DBSCAN<T: RealNumber, D: Distance<Vec<T>, T>> {
     eps: T,
 }
 
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 /// DBSCAN clustering algorithm parameters
 pub struct DBSCANParameters<T: RealNumber, D: Distance<Vec<T>, T>> {
@@ -229,7 +230,7 @@ impl<T: RealNumber> Default for DBSCANParameters<T, Euclidian> {
             distance: Distances::euclidian(),
             min_samples: 5,
             eps: T::half(),
-            algorithm: KNNAlgorithmName::CoverTree,
+            algorithm: KNNAlgorithmName::default(),
         }
     }
 }

src/cluster/kmeans.rs

@@ -102,6 +102,7 @@ impl<T: RealNumber> PartialEq for KMeans<T> {
     }
 }
 
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 /// K-Means clustering algorithm parameters
 pub struct KMeansParameters {

src/decomposition/pca.rs

@@ -83,6 +83,7 @@ impl<T: RealNumber, M: Matrix<T>> PartialEq for PCA<T, M> {
     }
 }
 
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 /// PCA parameters
 pub struct PCAParameters {

src/decomposition/svd.rs

@@ -69,6 +69,7 @@ impl<T: RealNumber, M: Matrix<T>> PartialEq for SVD<T, M> {
     }
 }
 
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 /// SVD parameters
 pub struct SVDParameters {

src/linear/linear_regression.rs

@@ -80,6 +80,12 @@ pub enum LinearRegressionSolverName {
     SVD,
 }
 
+impl Default for LinearRegressionSolverName {
+    fn default() -> Self {
+        LinearRegressionSolverName::SVD
+    }
+}
+
 /// Linear Regression parameters
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
@@ -109,7 +115,7 @@ impl LinearRegressionParameters {
 impl Default for LinearRegressionParameters {
     fn default() -> Self {
         LinearRegressionParameters {
-            solver: LinearRegressionSolverName::SVD,
+            solver: LinearRegressionSolverName::default(),
         }
     }
 }

src/linear/logistic_regression.rs

@@ -75,6 +75,12 @@ pub enum LogisticRegressionSolverName {
     LBFGS,
 }
 
+impl Default for LogisticRegressionSolverName {
+    fn default() -> Self {
+        LogisticRegressionSolverName::LBFGS
+    }
+}
+
 /// Logistic Regression parameters
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
@@ -208,7 +214,7 @@ impl<T: RealNumber> LogisticRegressionParameters<T> {
 impl<T: RealNumber> Default for LogisticRegressionParameters<T> {
     fn default() -> Self {
         LogisticRegressionParameters {
-            solver: LogisticRegressionSolverName::LBFGS,
+            solver: LogisticRegressionSolverName::default(),
             alpha: T::zero(),
         }
     }

src/linear/ridge_regression.rs

@@ -71,14 +71,18 @@ use crate::math::num::RealNumber;
 #[derive(Debug, Clone, Eq, PartialEq)]
 /// Approach to use for estimation of regression coefficients. Cholesky is more efficient but SVD is more stable.
 pub enum RidgeRegressionSolverName {
-    #[cfg_attr(feature = "serde", serde(default))]
     /// Cholesky decomposition, see [Cholesky](../../linalg/cholesky/index.html)
     Cholesky,
-    #[cfg_attr(feature = "serde", serde(default))]
     /// SVD decomposition, see [SVD](../../linalg/svd/index.html)
     SVD,
 }
 
+impl Default for RidgeRegressionSolverName {
+    fn default() -> Self {
+        RidgeRegressionSolverName::Cholesky
+    }
+}
+
 /// Ridge Regression parameters
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
@@ -209,7 +213,7 @@ impl<T: RealNumber> RidgeRegressionParameters<T> {
 impl<T: RealNumber> Default for RidgeRegressionParameters<T> {
     fn default() -> Self {
         RidgeRegressionParameters {
-            solver: RidgeRegressionSolverName::Cholesky,
+            solver: RidgeRegressionSolverName::default(),
             alpha: T::one(),
             normalize: true,
         }

src/neighbors/knn_classifier.rs

@@ -116,8 +116,8 @@ impl<T: RealNumber> Default for KNNClassifierParameters<T, Euclidian> {
     fn default() -> Self {
         KNNClassifierParameters {
             distance: Distances::euclidian(),
-            algorithm: KNNAlgorithmName::CoverTree,
-            weight: KNNWeightFunction::Uniform,
+            algorithm: KNNAlgorithmName::default(),
+            weight: KNNWeightFunction::default(),
             k: 3,
             t: PhantomData,
         }

src/neighbors/knn_regressor.rs

@@ -118,8 +118,8 @@ impl<T: RealNumber> Default for KNNRegressorParameters<T, Euclidian> {
     fn default() -> Self {
         KNNRegressorParameters {
             distance: Distances::euclidian(),
-            algorithm: KNNAlgorithmName::CoverTree,
-            weight: KNNWeightFunction::Uniform,
+            algorithm: KNNAlgorithmName::default(),
+            weight: KNNWeightFunction::default(),
             k: 3,
             t: PhantomData,
         }

src/neighbors/mod.rs

@@ -58,6 +58,12 @@ pub enum KNNWeightFunction {
     Distance,
 }
 
+impl Default for KNNWeightFunction {
+    fn default() -> Self {
+        KNNWeightFunction::Uniform
+    }
+}
+
 impl KNNWeightFunction {
     fn calc_weights<T: RealNumber>(&self, distances: Vec<T>) -> std::vec::Vec<T> {
         match *self {

src/tree/decision_tree_classifier.rs

@@ -123,6 +123,12 @@ pub enum SplitCriterion {
     ClassificationError,
 }
 
+impl Default for SplitCriterion {
+    fn default() -> Self {
+        SplitCriterion::Gini
+    }
+}
+
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug)]
 struct Node<T: RealNumber> {
@@ -201,7 +207,7 @@ impl DecisionTreeClassifierParameters {
 impl Default for DecisionTreeClassifierParameters {
     fn default() -> Self {
         DecisionTreeClassifierParameters {
-            criterion: SplitCriterion::Gini,
+            criterion: SplitCriterion::default(),
             max_depth: None,
             min_samples_leaf: 1,
             min_samples_split: 2,