Merge branch 'development' into kmeans-with-fastpair

* Remove some allocations

* Remove some more allocations

.github/CONTRIBUTING.md

@@ -37,6 +37,8 @@ $ rust-code-analysis-cli -p src/algorithm/neighbour/fastpair.rs --ls 22 --le 213
 ```
 * find more information about what happens in your binary with [`twiggy`](https://rustwasm.github.io/twiggy/install.html). This need a compiled binary so create a brief `main {}` function using `smartcore` and then point `twiggy` to that file.
 
+* Please take a look to the output of a profiler to spot most evident performance problems, see [this guide about using a profiler](http://www.codeofview.com/fix-rs/2017/01/24/how-to-optimize-rust-programs-on-linux/).
+
 ## Issue Report Process
 
 1. Go to the project's issues.

.github/workflows/coverage.yml

@@ -41,4 +41,4 @@ jobs:
       - name: Upload to codecov.io
         uses: codecov/codecov-action@v2
         with:
-          fail_ci_if_error:     true
+          fail_ci_if_error: false

Cargo.toml

@@ -2,7 +2,7 @@
 name = "smartcore"
 description = "Machine Learning in Rust."
 homepage = "https://smartcorelib.org"
-version = "0.3.0"
+version = "0.3.2"
 authors = ["smartcore Developers"]
 edition = "2021"
 license = "Apache-2.0"
@@ -42,13 +42,13 @@ std_rand = ["rand/std_rng", "rand/std"]
 js = ["getrandom/js"]
 
 [target.'cfg(target_arch = "wasm32")'.dependencies]
-getrandom = { version = "*", optional = true }
+getrandom = { version = "0.2.8", optional = true }
 
 [target.'cfg(all(target_arch = "wasm32", not(target_os = "wasi")))'.dev-dependencies]
 wasm-bindgen-test = "0.3"
 
 [dev-dependencies]
-itertools = "*"
+itertools = "0.10.5"
 serde_json = "1.0"
 bincode = "1.3.1"
 

README.md

@@ -18,4 +18,4 @@
 -----
 [![CI](https://github.com/smartcorelib/smartcore/actions/workflows/ci.yml/badge.svg)](https://github.com/smartcorelib/smartcore/actions/workflows/ci.yml)
 
-To start getting familiar with the new smartcore v0.5 API, there is now available a [**Jupyter Notebook environment repository**](https://github.com/smartcorelib/smartcore-jupyter). Please see instructions there, contributions welcome see [CONTRIBUTING](.github/CONTRIBUTING.md).
+To start getting familiar with the new smartcore v0.3 API, there is now available a [**Jupyter Notebook environment repository**](https://github.com/smartcorelib/smartcore-jupyter). Please see instructions there, contributions welcome see [CONTRIBUTING](.github/CONTRIBUTING.md).

smartcore.iml

@@ -1,15 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<module type="RUST_MODULE" version="4">
-  <component name="NewModuleRootManager" inherit-compiler-output="true">
-    <exclude-output />
-    <content url="file://$MODULE_DIR$">
-      <sourceFolder url="file://$MODULE_DIR$/src" isTestSource="false" />
-      <sourceFolder url="file://$MODULE_DIR$/examples" isTestSource="false" />
-      <sourceFolder url="file://$MODULE_DIR$/tests" isTestSource="true" />
-      <sourceFolder url="file://$MODULE_DIR$/benches" isTestSource="true" />
-      <excludeFolder url="file://$MODULE_DIR$/target" />
-    </content>
-    <orderEntry type="inheritedJdk" />
-    <orderEntry type="sourceFolder" forTests="false" />
-  </component>
-</module>

src/algorithm/neighbour/fastpair.rs

@@ -179,6 +179,21 @@ impl<'a, T: RealNumber + FloatNumber, M: Array2<T>> FastPair<'a, T, M> {
         }
     }
 
+    ///
+    /// Return order dissimilarities from closest to furthest
+    ///
+    #[allow(dead_code)]
+    pub fn ordered_pairs(&self) -> std::vec::IntoIter<&PairwiseDistance<T>> {
+        // improvement: implement this to return `impl Iterator<Item = &PairwiseDistance<T>>`
+        // need to implement trait `Iterator` for `Vec<&PairwiseDistance<T>>`
+        let mut distances = self
+            .distances
+            .values()
+            .collect::<Vec<&PairwiseDistance<T>>>();
+        distances.sort_by(|a, b| a.partial_cmp(b).unwrap());
+        distances.into_iter()
+    }
+
     //
     // Compute distances from input to all other points in data-structure.
     // input is the row index of the sample matrix
@@ -260,8 +275,8 @@ mod tests_fastpair {
         let distances = fastpair.distances;
         let neighbours = fastpair.neighbours;
 
-        assert!(distances.len() != 0);
+        assert!(!distances.is_empty());
-        assert!(neighbours.len() != 0);
+        assert!(!neighbours.is_empty());
 
         assert_eq!(10, neighbours.len());
         assert_eq!(10, distances.len());
@@ -276,17 +291,13 @@ mod tests_fastpair {
         // We expect an error when we run `FastPair` on this dataset,
         // becuase `FastPair` currently only works on a minimum of 3
         // points.
-        let _fastpair = FastPair::new(&dataset);
+        let fastpair = FastPair::new(&dataset);
+        assert!(fastpair.is_err());
 
-        match _fastpair {
+        if let Err(e) = fastpair {
-            Err(e) => {
+            let expected_error =
-                let expected_error =
+                Failed::because(FailedError::FindFailed, "min number of rows should be 3");
-                    Failed::because(FailedError::FindFailed, "min number of rows should be 3");
+            assert_eq!(e, expected_error)
-                assert_eq!(e, expected_error)
-            }
-            _ => {
-                assert!(false);
-            }
         }
     }
 
@@ -582,7 +593,7 @@ mod tests_fastpair {
         };
         for p in dissimilarities.iter() {
             if p.distance.unwrap() < min_dissimilarity.distance.unwrap() {
-                min_dissimilarity = p.clone()
+                min_dissimilarity = *p
             }
         }
 
@@ -594,4 +605,39 @@ mod tests_fastpair {
 
         assert_eq!(closest, min_dissimilarity);
     }
+
+    #[test]
+    fn fastpair_ordered_pairs() {
+        let x = DenseMatrix::<f64>::from_2d_array(&[
+            &[5.1, 3.5, 1.4, 0.2],
+            &[4.9, 3.0, 1.4, 0.2],
+            &[4.7, 3.2, 1.3, 0.2],
+            &[4.6, 3.1, 1.5, 0.2],
+            &[5.0, 3.6, 1.4, 0.2],
+            &[5.4, 3.9, 1.7, 0.4],
+            &[4.9, 3.1, 1.5, 0.1],
+            &[7.0, 3.2, 4.7, 1.4],
+            &[6.4, 3.2, 4.5, 1.5],
+            &[6.9, 3.1, 4.9, 1.5],
+            &[5.5, 2.3, 4.0, 1.3],
+            &[6.5, 2.8, 4.6, 1.5],
+            &[4.6, 3.4, 1.4, 0.3],
+            &[5.0, 3.4, 1.5, 0.2],
+            &[4.4, 2.9, 1.4, 0.2],
+        ]);
+        let fastpair = FastPair::new(&x).unwrap();
+
+        let ordered = fastpair.ordered_pairs();
+
+        let mut previous: f64 = -1.0;
+        for p in ordered {
+            if previous == -1.0 {
+                previous = p.distance.unwrap();
+            } else {
+                let current = p.distance.unwrap();
+                assert!(current >= previous);
+                previous = current;
+            }
+        }
+    }
 }

src/algorithm/neighbour/mod.rs

@@ -49,20 +49,15 @@ pub mod linear_search;
 /// Both, KNN classifier and regressor benefits from underlying search algorithms that helps to speed up queries.
 /// `KNNAlgorithmName` maintains a list of supported search algorithms, see [KNN algorithms](../algorithm/neighbour/index.html)
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Default)]
 pub enum KNNAlgorithmName {
     /// Heap Search algorithm, see [`LinearSearch`](../algorithm/neighbour/linear_search/index.html)
     LinearSearch,
     /// Cover Tree Search algorithm, see [`CoverTree`](../algorithm/neighbour/cover_tree/index.html)
+    #[default]
     CoverTree,
 }
 
-impl Default for KNNAlgorithmName {
-    fn default() -> Self {
-        KNNAlgorithmName::CoverTree
-    }
-}
-
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug)]
 pub(crate) enum KNNAlgorithm<T: Number, D: Distance<Vec<T>>> {

src/cluster/dbscan.rs

@@ -18,7 +18,7 @@
 //!
 //! Example:
 //!
-//! ```
+//! ```ignore
 //! use smartcore::linalg::basic::matrix::DenseMatrix;
 //! use smartcore::linalg::basic::arrays::Array2;
 //! use smartcore::cluster::dbscan::*;
@@ -511,6 +511,6 @@ mod tests {
             .and_then(|dbscan| dbscan.predict(&x))
             .unwrap();
 
-        println!("{:?}", labels);
+        println!("{labels:?}");
     }
 }

src/cluster/kmeans.rs

@@ -62,7 +62,7 @@ use serde::{Deserialize, Serialize};
 use crate::algorithm::neighbour::bbd_tree::BBDTree;
 use crate::api::{Predictor, UnsupervisedEstimator};
 use crate::error::Failed;
-use crate::linalg::basic::arrays::{Array1, Array2};
+use crate::linalg::basic::arrays::{Array1, Array2, Array};
 use crate::metrics::distance::euclidian::*;
 use crate::numbers::basenum::Number;
 use crate::rand_custom::get_rng_impl;
@@ -322,6 +322,109 @@ impl<TX: Number, TY: Number, X: Array2<TX>, Y: Array1<TY>> KMeans<TX, TY, X, Y>
         })
     }
 
+    /// Fit algorithm to _NxM_ matrix where _N_ is number of samples and _M_ is number of features.
+    /// * `data` - training instances to cluster    
+    /// * `parameters` - cluster parameters
+    /// * `centroids` - starting centroids
+    pub fn fit_with_centroids(
+        data: &X,
+        parameters: KMeansParameters,
+        centroids: Vec<Vec<f64>>,
+    ) -> Result<KMeans<TX, TY, X, Y>, Failed> {
+
+        // TODO: reuse existing methods in `crate::metrics`
+        fn euclidean_distance(point1: &Vec<f64>, point2: &Vec<f64>) -> f64 {
+            let mut dist = 0.0;
+            for i in 0..point1.len() {
+                dist += (point1[i] - point2[i]).powi(2);
+            }
+            dist.sqrt()
+        }
+
+        fn closest_centroid(point: &Vec<f64>, centroids: &Vec<Vec<f64>>) -> usize {
+            let mut closest_idx = 0;
+            let mut closest_dist = std::f64::MAX;
+            for (i, centroid) in centroids.iter().enumerate() {
+                let dist = euclidean_distance(point, centroid);
+                if dist < closest_dist {
+                    closest_dist = dist;
+                    closest_idx = i;
+                }
+            }
+            closest_idx
+        }
+
+        let bbd = BBDTree::new(data);
+    
+        if centroids.len() != parameters.k {
+            return Err(Failed::fit(&format!(
+                "number of centroids ({}) must be equal to k ({})",
+                centroids.len(),
+                parameters.k
+            )));
+        }
+    
+        let mut y = vec![0; data.shape().0];
+        for i in 0..data.shape().0 {
+            y[i] = closest_centroid(
+                &Vec::from_iterator(data.get_row(i).iterator(0).map(|e| e.to_f64().unwrap()),
+                data.shape().1), &centroids
+            );
+        }
+    
+        let mut size = vec![0; parameters.k];
+        let mut new_centroids = vec![vec![0f64; data.shape().1]; parameters.k];
+    
+        for i in 0..data.shape().0 {
+            size[y[i]] += 1;
+        }
+    
+        for i in 0..data.shape().0 {
+            for j in 0..data.shape().1 {
+                new_centroids[y[i]][j] += data.get((i, j)).to_f64().unwrap();
+            }
+        }
+    
+        for i in 0..parameters.k {
+            for j in 0..data.shape().1 {
+                new_centroids[i][j] /= size[i] as f64;
+            }
+        }
+    
+        let mut sums = vec![vec![0f64; data.shape().1]; parameters.k];
+        let mut distortion = std::f64::MAX;
+    
+        for _ in 1..=parameters.max_iter {
+            let dist = bbd.clustering(&new_centroids, &mut sums, &mut size, &mut y);
+            for i in 0..parameters.k {
+                if size[i] > 0 {
+                    for j in 0..data.shape().1 {
+                        new_centroids[i][j] = sums[i][j] / size[i] as f64;
+                    }
+                }
+            }
+    
+            if distortion <= dist {
+                break;
+            } else {
+                distortion = dist;
+            }
+        }
+    
+        Ok(KMeans {
+            k: parameters.k,
+            _y: y,
+            size,
+            _distortion: distortion,
+            centroids: new_centroids,
+            _phantom_tx: PhantomData,
+            _phantom_ty: PhantomData,
+            _phantom_x: PhantomData,
+            _phantom_y: PhantomData,
+        })
+    }
+
+
     /// Predict clusters for `x`
     /// * `x` - matrix with new data to transform of size _KxM_ , where _K_ is number of new samples and _M_ is number of features.
     pub fn predict(&self, x: &X) -> Result<Y, Failed> {
@@ -417,6 +520,7 @@ impl<TX: Number, TY: Number, X: Array2<TX>, Y: Array1<TY>> KMeans<TX, TY, X, Y>
 mod tests {
     use super::*;
     use crate::linalg::basic::matrix::DenseMatrix;
+    use crate::algorithm::neighbour::fastpair;
 
     #[cfg_attr(
         all(target_arch = "wasm32", not(target_os = "wasi")),
@@ -498,8 +602,80 @@ mod tests {
 
         let y: Vec<usize> = kmeans.predict(&x).unwrap();
 
-        for i in 0..y.len() {
+        for (i, _y_i) in y.iter().enumerate() {
-            assert_eq!(y[i] as usize, kmeans._y[i]);
+            assert_eq!({ y[i] }, kmeans._y[i]);
+        }
+    }
+
+    #[cfg_attr(
+        all(target_arch = "wasm32", not(target_os = "wasi")),
+        wasm_bindgen_test::wasm_bindgen_test
+    )]
+    #[test]
+    fn fit_with_centroids_predict() {
+        let x = DenseMatrix::from_2d_array(&[
+            &[5.1, 3.5, 1.4, 0.2],
+            &[4.9, 3.0, 1.4, 0.2],
+            &[4.7, 3.2, 1.3, 0.2],
+            &[4.6, 3.1, 1.5, 0.2],
+            &[5.0, 3.6, 1.4, 0.2],
+            &[5.4, 3.9, 1.7, 0.4],
+            &[4.6, 3.4, 1.4, 0.3],
+            &[5.0, 3.4, 1.5, 0.2],
+            &[4.4, 2.9, 1.4, 0.2],
+            &[4.9, 3.1, 1.5, 0.1],
+            &[7.0, 3.2, 4.7, 1.4],
+            &[6.4, 3.2, 4.5, 1.5],
+            &[6.9, 3.1, 4.9, 1.5],
+            &[5.5, 2.3, 4.0, 1.3],
+            &[6.5, 2.8, 4.6, 1.5],
+            &[5.7, 2.8, 4.5, 1.3],
+            &[6.3, 3.3, 4.7, 1.6],
+            &[4.9, 2.4, 3.3, 1.0],
+            &[6.6, 2.9, 4.6, 1.3],
+            &[5.2, 2.7, 3.9, 1.4],
+        ]);
+
+        let parameters = KMeansParameters {
+            k: 3,
+            max_iter: 50,
+            ..Default::default()
+        };
+
+        // compute pairs
+        let fastpair = fastpair::FastPair::new(&x).unwrap();
+
+        // compute centroids for N closest pairs
+        let mut n: isize = 2;
+        let mut centroids = vec![vec![0f64; x.shape().1]; n as usize + 1];
+        for p in fastpair.ordered_pairs() {
+            if n == -1 {
+                break
+            }
+
+            centroids[n as usize] = {
+                let mut result: Vec<f64> = Vec::with_capacity(x.shape().1);
+                for val1 in x.get_row(p.node).iterator(0) {
+                    for val2 in x.get_row(p.neighbour.unwrap()).iterator(0) {
+                        let sum = val1 + val2;
+                        let avg = sum * 0.5f64;
+                        result.push(avg);
+                    }
+                }
+                result
+            };
+
+            n -= 1;
+        }
+
+
+        let kmeans = KMeans::fit_with_centroids(
+            &x, parameters, centroids).unwrap();
+
+        let y: Vec<usize> = kmeans.predict(&x).unwrap();
+
+        for (i, _y_i) in y.iter().enumerate() {
+            assert_eq!({ y[i] }, kmeans._y[i]);
         }
     }
 

src/dataset/boston.rs

@@ -31,7 +31,7 @@ use crate::dataset::Dataset;
 pub fn load_dataset() -> Dataset<f32, f32> {
     let (x, y, num_samples, num_features) = match deserialize_data(std::include_bytes!("boston.xy"))
     {
-        Err(why) => panic!("Can't deserialize boston.xy. {}", why),
+        Err(why) => panic!("Can't deserialize boston.xy. {why}"),
         Ok((x, y, num_samples, num_features)) => (x, y, num_samples, num_features),
     };
 

src/dataset/breast_cancer.rs

@@ -33,7 +33,7 @@ use crate::dataset::Dataset;
 pub fn load_dataset() -> Dataset<f32, u32> {
     let (x, y, num_samples, num_features) =
         match deserialize_data(std::include_bytes!("breast_cancer.xy")) {
-            Err(why) => panic!("Can't deserialize breast_cancer.xy. {}", why),
+            Err(why) => panic!("Can't deserialize breast_cancer.xy. {why}"),
             Ok((x, y, num_samples, num_features)) => (
                 x,
                 y.into_iter().map(|x| x as u32).collect(),

src/dataset/diabetes.rs

@@ -26,7 +26,7 @@ use crate::dataset::Dataset;
 pub fn load_dataset() -> Dataset<f32, u32> {
     let (x, y, num_samples, num_features) =
         match deserialize_data(std::include_bytes!("diabetes.xy")) {
-            Err(why) => panic!("Can't deserialize diabetes.xy. {}", why),
+            Err(why) => panic!("Can't deserialize diabetes.xy. {why}"),
             Ok((x, y, num_samples, num_features)) => (
                 x,
                 y.into_iter().map(|x| x as u32).collect(),

src/dataset/digits.rs

@@ -16,7 +16,7 @@ use crate::dataset::Dataset;
 pub fn load_dataset() -> Dataset<f32, f32> {
     let (x, y, num_samples, num_features) = match deserialize_data(std::include_bytes!("digits.xy"))
     {
-        Err(why) => panic!("Can't deserialize digits.xy. {}", why),
+        Err(why) => panic!("Can't deserialize digits.xy. {why}"),
         Ok((x, y, num_samples, num_features)) => (x, y, num_samples, num_features),
     };
 

src/dataset/iris.rs

@@ -22,7 +22,7 @@ use crate::dataset::Dataset;
 pub fn load_dataset() -> Dataset<f32, u32> {
     let (x, y, num_samples, num_features): (Vec<f32>, Vec<u32>, usize, usize) =
         match deserialize_data(std::include_bytes!("iris.xy")) {
-            Err(why) => panic!("Can't deserialize iris.xy. {}", why),
+            Err(why) => panic!("Can't deserialize iris.xy. {why}"),
             Ok((x, y, num_samples, num_features)) => (
                 x,
                 y.into_iter().map(|x| x as u32).collect(),

src/dataset/mod.rs

@@ -78,7 +78,7 @@ pub(crate) fn serialize_data<X: Number + RealNumber, Y: RealNumber>(
                 .collect();
             file.write_all(&y)?;
         }
-        Err(why) => panic!("couldn't create {}: {}", filename, why),
+        Err(why) => panic!("couldn't create {filename}: {why}"),
     }
     Ok(())
 }

src/decomposition/pca.rs

@@ -231,8 +231,7 @@ impl<T: Number + RealNumber, X: Array2<T> + SVDDecomposable<T> + EVDDecomposable
 
         if parameters.n_components > n {
             return Err(Failed::fit(&format!(
-                "Number of components, n_components should be <= number of attributes ({})",
+                "Number of components, n_components should be <= number of attributes ({n})"
-                n
             )));
         }
 
@@ -374,21 +373,20 @@ mod tests {
         let parameters = PCASearchParameters {
             n_components: vec![2, 4],
             use_correlation_matrix: vec![true, false],
-            ..Default::default()
         };
         let mut iter = parameters.into_iter();
         let next = iter.next().unwrap();
         assert_eq!(next.n_components, 2);
-        assert_eq!(next.use_correlation_matrix, true);
+        assert!(next.use_correlation_matrix);
         let next = iter.next().unwrap();
         assert_eq!(next.n_components, 4);
-        assert_eq!(next.use_correlation_matrix, true);
+        assert!(next.use_correlation_matrix);
         let next = iter.next().unwrap();
         assert_eq!(next.n_components, 2);
-        assert_eq!(next.use_correlation_matrix, false);
+        assert!(!next.use_correlation_matrix);
         let next = iter.next().unwrap();
         assert_eq!(next.n_components, 4);
-        assert_eq!(next.use_correlation_matrix, false);
+        assert!(!next.use_correlation_matrix);
         assert!(iter.next().is_none());
     }
 
@@ -572,8 +570,8 @@ mod tests {
             epsilon = 1e-4
         ));
 
-        for i in 0..pca.eigenvalues.len() {
+        for (i, pca_eigenvalues_i) in pca.eigenvalues.iter().enumerate() {
-            assert!((pca.eigenvalues[i].abs() - expected_eigenvalues[i].abs()).abs() < 1e-8);
+            assert!((pca_eigenvalues_i.abs() - expected_eigenvalues[i].abs()).abs() < 1e-8);
         }
 
         let us_arrests_t = pca.transform(&us_arrests).unwrap();
@@ -694,8 +692,8 @@ mod tests {
             epsilon = 1e-4
         ));
 
-        for i in 0..pca.eigenvalues.len() {
+        for (i, pca_eigenvalues_i) in pca.eigenvalues.iter().enumerate() {
-            assert!((pca.eigenvalues[i].abs() - expected_eigenvalues[i].abs()).abs() < 1e-8);
+            assert!((pca_eigenvalues_i.abs() - expected_eigenvalues[i].abs()).abs() < 1e-8);
         }
 
         let us_arrests_t = pca.transform(&us_arrests).unwrap();

src/decomposition/svd.rs

@@ -180,8 +180,7 @@ impl<T: Number + RealNumber, X: Array2<T> + SVDDecomposable<T> + EVDDecomposable
 
         if parameters.n_components >= p {
             return Err(Failed::fit(&format!(
-                "Number of components, n_components should be < number of attributes ({})",
+                "Number of components, n_components should be < number of attributes ({p})"
-                p
             )));
         }
 
@@ -202,8 +201,7 @@ impl<T: Number + RealNumber, X: Array2<T> + SVDDecomposable<T> + EVDDecomposable
         let (p_c, k) = self.components.shape();
         if p_c != p {
             return Err(Failed::transform(&format!(
-                "Can not transform a {}x{} matrix into {}x{} matrix, incorrect input dimentions",
+                "Can not transform a {n}x{p} matrix into {n}x{k} matrix, incorrect input dimentions"
-                n, p, n, k
             )));
         }
 
@@ -227,7 +225,6 @@ mod tests {
     fn search_parameters() {
         let parameters = SVDSearchParameters {
             n_components: vec![10, 100],
-            ..Default::default()
         };
         let mut iter = parameters.into_iter();
         let next = iter.next().unwrap();

src/ensemble/random_forest_classifier.rs

@@ -454,8 +454,12 @@ impl<TX: FloatNumber + PartialOrd, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY
         y: &Y,
         parameters: RandomForestClassifierParameters,
     ) -> Result<RandomForestClassifier<TX, TY, X, Y>, Failed> {
-        let (_, num_attributes) = x.shape();
+        let (x_nrows, num_attributes) = x.shape();
         let y_ncols = y.shape();
+        if x_nrows != y_ncols {
+            return Err(Failed::fit("Number of rows in X should = len(y)"));
+        }
+
         let mut yi: Vec<usize> = vec![0; y_ncols];
         let classes = y.unique();
 
@@ -678,6 +682,30 @@ mod tests {
         assert!(accuracy(&y, &classifier.predict(&x).unwrap()) >= 0.95);
     }
 
+    #[test]
+    fn test_random_matrix_with_wrong_rownum() {
+        let x_rand: DenseMatrix<f64> = DenseMatrix::<f64>::rand(21, 200);
+
+        let y: Vec<u32> = vec![0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1];
+
+        let fail = RandomForestClassifier::fit(
+            &x_rand,
+            &y,
+            RandomForestClassifierParameters {
+                criterion: SplitCriterion::Gini,
+                max_depth: Option::None,
+                min_samples_leaf: 1,
+                min_samples_split: 2,
+                n_trees: 100,
+                m: Option::None,
+                keep_samples: false,
+                seed: 87,
+            },
+        );
+
+        assert!(fail.is_err());
+    }
+
     #[cfg_attr(
         all(target_arch = "wasm32", not(target_os = "wasi")),
         wasm_bindgen_test::wasm_bindgen_test

src/ensemble/random_forest_regressor.rs

@@ -399,6 +399,10 @@ impl<TX: Number + FloatNumber + PartialOrd, TY: Number, X: Array2<TX>, Y: Array1
     ) -> Result<RandomForestRegressor<TX, TY, X, Y>, Failed> {
         let (n_rows, num_attributes) = x.shape();
 
+        if n_rows != y.shape() {
+            return Err(Failed::fit("Number of rows in X should = len(y)"));
+        }
+
         let mtry = parameters
             .m
             .unwrap_or((num_attributes as f64).sqrt().floor() as usize);
@@ -595,6 +599,32 @@ mod tests {
         assert!(mean_absolute_error(&y, &y_hat) < 1.0);
     }
 
+    #[test]
+    fn test_random_matrix_with_wrong_rownum() {
+        let x_rand: DenseMatrix<f64> = DenseMatrix::<f64>::rand(17, 200);
+
+        let y = vec![
+            83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
+            114.2, 115.7, 116.9,
+        ];
+
+        let fail = RandomForestRegressor::fit(
+            &x_rand,
+            &y,
+            RandomForestRegressorParameters {
+                max_depth: Option::None,
+                min_samples_leaf: 1,
+                min_samples_split: 2,
+                n_trees: 1000,
+                m: Option::None,
+                keep_samples: false,
+                seed: 87,
+            },
+        );
+
+        assert!(fail.is_err());
+    }
+
     #[cfg_attr(
         all(target_arch = "wasm32", not(target_os = "wasi")),
         wasm_bindgen_test::wasm_bindgen_test

src/error/mod.rs

@@ -30,7 +30,7 @@ pub enum FailedError {
     DecompositionFailed,
     /// Can't solve for x
     SolutionFailed,
-    /// Erro in input
+    /// Error in input parameters
     ParametersError,
 }
 
@@ -98,7 +98,7 @@ impl fmt::Display for FailedError {
             FailedError::SolutionFailed => "Can't find solution",
             FailedError::ParametersError => "Error in input, check parameters",
         };
-        write!(f, "{}", failed_err_str)
+        write!(f, "{failed_err_str}")
     }
 }
 

src/lib.rs

@@ -3,7 +3,8 @@
     clippy::too_many_arguments,
     clippy::many_single_char_names,
     clippy::unnecessary_wraps,
-    clippy::upper_case_acronyms
+    clippy::upper_case_acronyms,
+    clippy::approx_constant
 )]
 #![warn(missing_docs)]
 #![warn(rustdoc::missing_doc_code_examples)]

src/linalg/basic/arrays.rs

@@ -548,7 +548,7 @@ pub trait ArrayView2<T: Debug + Display + Copy + Sized>: Array<T, (usize, usize)
         let (nrows, ncols) = self.shape();
         for r in 0..nrows {
             let row: Vec<T> = (0..ncols).map(|c| *self.get((r, c))).collect();
-            writeln!(f, "{:?}", row)?
+            writeln!(f, "{row:?}")?
         }
         Ok(())
     }
@@ -918,8 +918,7 @@ pub trait Array1<T: Debug + Display + Copy + Sized>: MutArrayView1<T> + Sized +
         let len = self.shape();
         assert!(
             index.iter().all(|&i| i < len),
-            "All indices in `take` should be < {}",
+            "All indices in `take` should be < {len}"
-            len
         );
         Self::from_iterator(index.iter().map(move |&i| *self.get(i)), index.len())
     }
@@ -990,10 +989,7 @@ pub trait Array1<T: Debug + Display + Copy + Sized>: MutArrayView1<T> + Sized +
         };
         assert!(
             d1 == len,
-            "Can not multiply {}x{} matrix by {} vector",
+            "Can not multiply {nrows}x{ncols} matrix by {len} vector"
-            nrows,
-            ncols,
-            len
         );
         let mut result = Self::zeros(d2);
         for i in 0..d2 {
@@ -1111,11 +1107,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
 
         assert!(
             nrows * ncols == onrows * oncols,
-            "Can't reshape {}x{} array into a {}x{} array",
+            "Can't reshape {onrows}x{oncols} array into a {nrows}x{ncols} array"
-            onrows,
-            oncols,
-            nrows,
-            ncols
         );
 
         Self::from_iterator(self.iterator(0).cloned(), nrows, ncols, axis)
@@ -1129,11 +1121,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
         let (o_nrows, o_ncols) = other.shape();
         assert!(
             ncols == o_nrows,
-            "Can't multiply {}x{} and {}x{} matrices",
+            "Can't multiply {nrows}x{ncols} and {o_nrows}x{o_ncols} matrices"
-            nrows,
-            ncols,
-            o_nrows,
-            o_ncols
         );
         let inner_d = ncols;
         let mut result = Self::zeros(nrows, o_ncols);
@@ -1166,7 +1154,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
                 _ => (nrows, ncols, o_nrows, o_ncols),
             };
             if d1 != d4 {
-                panic!("Can not multiply {}x{} by {}x{} matrices", d2, d1, d4, d3);
+                panic!("Can not multiply {d2}x{d1} by {d4}x{d3} matrices");
             }
             let mut result = Self::zeros(d2, d3);
             for r in 0..d2 {
@@ -1198,10 +1186,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
         };
         assert!(
             d2 == len,
-            "Can not multiply {}x{} matrix by {} vector",
+            "Can not multiply {nrows}x{ncols} matrix by {len} vector"
-            nrows,
-            ncols,
-            len
         );
         let mut result = Self::zeros(d1, 1);
         for i in 0..d1 {
@@ -1432,8 +1417,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
             0 => {
                 assert!(
                     index.iter().all(|&i| i < nrows),
-                    "All indices in `take` should be < {}",
+                    "All indices in `take` should be < {nrows}"
-                    nrows
                 );
                 Self::from_iterator(
                     index
@@ -1448,8 +1432,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
             _ => {
                 assert!(
                     index.iter().all(|&i| i < ncols),
-                    "All indices in `take` should be < {}",
+                    "All indices in `take` should be < {ncols}"
-                    ncols
                 );
                 Self::from_iterator(
                     (0..nrows)
@@ -1587,7 +1570,7 @@ pub trait Array2<T: Debug + Display + Copy + Sized>: MutArrayView2<T> + Sized +
         mean
     }
 
-    /// copy coumn as a vector
+    /// copy column as a vector
     fn copy_col_as_vec(&self, col: usize, result: &mut Vec<T>) {
         for (r, result_r) in result.iter_mut().enumerate().take(self.shape().0) {
             *result_r = *self.get((r, col));
@@ -1736,7 +1719,7 @@ mod tests {
         let r = Vec::<f32>::rand(4);
         assert!(r.iterator(0).all(|&e| e <= 1f32));
         assert!(r.iterator(0).all(|&e| e >= 0f32));
-        assert!(r.iterator(0).map(|v| *v).sum::<f32>() > 0f32);
+        assert!(r.iterator(0).copied().sum::<f32>() > 0f32);
     }
 
     #[test]
@@ -1954,7 +1937,7 @@ mod tests {
             DenseMatrix::from_2d_array(&[&[1, 3], &[2, 4]])
         );
         assert_eq!(
-            DenseMatrix::concatenate_2d(&[&a.clone(), &b.clone()], 0),
+            DenseMatrix::concatenate_2d(&[&a, &b], 0),
             DenseMatrix::from_2d_array(&[&[1, 2], &[3, 4], &[5, 6], &[7, 8]])
         );
         assert_eq!(
@@ -2025,7 +2008,7 @@ mod tests {
         let r = DenseMatrix::<f32>::rand(2, 2);
         assert!(r.iterator(0).all(|&e| e <= 1f32));
         assert!(r.iterator(0).all(|&e| e >= 0f32));
-        assert!(r.iterator(0).map(|v| *v).sum::<f32>() > 0f32);
+        assert!(r.iterator(0).copied().sum::<f32>() > 0f32);
     }
 
     #[test]

src/linalg/basic/matrix.rs

@@ -431,9 +431,9 @@ impl<T: Number + RealNumber> SVDDecomposable<T> for DenseMatrix<T> {}
 impl<'a, T: Debug + Display + Copy + Sized> Array<T, (usize, usize)> for DenseMatrixView<'a, T> {
     fn get(&self, pos: (usize, usize)) -> &T {
         if self.column_major {
-            &self.values[(pos.0 + pos.1 * self.stride)]
+            &self.values[pos.0 + pos.1 * self.stride]
         } else {
-            &self.values[(pos.0 * self.stride + pos.1)]
+            &self.values[pos.0 * self.stride + pos.1]
         }
     }
 
@@ -495,9 +495,9 @@ impl<'a, T: Debug + Display + Copy + Sized> ArrayView1<T> for DenseMatrixView<'a
 impl<'a, T: Debug + Display + Copy + Sized> Array<T, (usize, usize)> for DenseMatrixMutView<'a, T> {
     fn get(&self, pos: (usize, usize)) -> &T {
         if self.column_major {
-            &self.values[(pos.0 + pos.1 * self.stride)]
+            &self.values[pos.0 + pos.1 * self.stride]
         } else {
-            &self.values[(pos.0 * self.stride + pos.1)]
+            &self.values[pos.0 * self.stride + pos.1]
         }
     }
 
@@ -519,9 +519,9 @@ impl<'a, T: Debug + Display + Copy + Sized> MutArray<T, (usize, usize)>
 {
     fn set(&mut self, pos: (usize, usize), x: T) {
         if self.column_major {
-            self.values[(pos.0 + pos.1 * self.stride)] = x;
+            self.values[pos.0 + pos.1 * self.stride] = x;
         } else {
-            self.values[(pos.0 * self.stride + pos.1)] = x;
+            self.values[pos.0 * self.stride + pos.1] = x;
         }
     }
 
@@ -581,9 +581,9 @@ mod tests {
             vec![4, 5, 6],
             DenseMatrix::from_slice(&(*x.slice(1..2, 0..3))).values
         );
-        let second_row: Vec<i32> = x.slice(1..2, 0..3).iterator(0).map(|x| *x).collect();
+        let second_row: Vec<i32> = x.slice(1..2, 0..3).iterator(0).copied().collect();
         assert_eq!(vec![4, 5, 6], second_row);
-        let second_col: Vec<i32> = x.slice(0..3, 1..2).iterator(0).map(|x| *x).collect();
+        let second_col: Vec<i32> = x.slice(0..3, 1..2).iterator(0).copied().collect();
         assert_eq!(vec![2, 5, 8], second_col);
     }
 
@@ -640,12 +640,12 @@ mod tests {
         let x = DenseMatrix::<&str>::from_2d_array(&[&["1", "2", "3"], &["4", "5", "6"]]);
 
         assert_eq!(vec!["1", "4", "2", "5", "3", "6"], x.values);
-        assert!(x.column_major == true);
+        assert!(x.column_major);
 
         // transpose
         let x = x.transpose();
         assert_eq!(vec!["1", "4", "2", "5", "3", "6"], x.values);
-        assert!(x.column_major == false); // should change column_major
+        assert!(!x.column_major); // should change column_major
     }
 
     #[test]
@@ -659,7 +659,7 @@ mod tests {
             vec![1, 2, 3, 4, 5, 6],
             m.values.iter().map(|e| **e).collect::<Vec<i32>>()
         );
-        assert!(m.column_major == false);
+        assert!(!m.column_major);
     }
 
     #[test]
@@ -667,10 +667,10 @@ mod tests {
         let a = DenseMatrix::from_2d_array(&[&[1, 2, 3], &[4, 5, 6]]);
         let b = DenseMatrix::from_2d_array(&[&[1, 2], &[3, 4], &[5, 6]]);
 
-        println!("{}", a);
+        println!("{a}");
         // take column 0 and 2
         assert_eq!(vec![1, 3, 4, 6], a.take(&[0, 2], 1).values);
-        println!("{}", b);
+        println!("{b}");
         // take rows 0 and 2
         assert_eq!(vec![1, 2, 5, 6], b.take(&[0, 2], 0).values);
     }
@@ -692,11 +692,11 @@ mod tests {
 
         let a = a.reshape(2, 6, 0);
         assert_eq!(vec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12], a.values);
-        assert!(a.ncols == 6 && a.nrows == 2 && a.column_major == false);
+        assert!(a.ncols == 6 && a.nrows == 2 && !a.column_major);
 
         let a = a.reshape(3, 4, 1);
         assert_eq!(vec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12], a.values);
-        assert!(a.ncols == 4 && a.nrows == 3 && a.column_major == true);
+        assert!(a.ncols == 4 && a.nrows == 3 && a.column_major);
     }
 
     #[test]

src/linalg/basic/vector.rs

@@ -15,6 +15,25 @@ pub struct VecView<'a, T: Debug + Display + Copy + Sized> {
     ptr: &'a [T],
 }
 
+impl<T: Debug + Display + Copy + Sized> Array<T, usize> for &[T] {
+    fn get(&self, i: usize) -> &T {
+        &self[i]
+    }
+
+    fn shape(&self) -> usize {
+        self.len()
+    }
+
+    fn is_empty(&self) -> bool {
+        self.len() > 0
+    }
+
+    fn iterator<'b>(&'b self, axis: u8) -> Box<dyn Iterator<Item = &'b T> + 'b> {
+        assert!(axis == 0, "For one dimensional array `axis` should == 0");
+        Box::new(self.iter())
+    }
+}
+
 impl<T: Debug + Display + Copy + Sized> Array<T, usize> for Vec<T> {
     fn get(&self, i: usize) -> &T {
         &self[i]
@@ -46,6 +65,7 @@ impl<T: Debug + Display + Copy + Sized> MutArray<T, usize> for Vec<T> {
 }
 
 impl<T: Debug + Display + Copy + Sized> ArrayView1<T> for Vec<T> {}
+impl<T: Debug + Display + Copy + Sized> ArrayView1<T> for &[T] {}
 
 impl<T: Debug + Display + Copy + Sized> MutArrayView1<T> for Vec<T> {}
 
@@ -160,8 +180,8 @@ mod tests {
     fn dot_product<T: Number, V: Array1<T>>(v: &V) -> T {
         let vv = V::zeros(10);
         let v_s = vv.slice(0..3);
-        let dot = v_s.dot(v);
+
-        dot
+        v_s.dot(v)
     }
 
     fn vector_ops<T: Number + PartialOrd, V: Array1<T>>(_: &V) -> T {
@@ -216,7 +236,7 @@ mod tests {
     #[test]
     fn test_mut_iterator() {
         let mut x = vec![1, 2, 3];
-        x.iterator_mut(0).for_each(|v| *v = *v * 2);
+        x.iterator_mut(0).for_each(|v| *v *= 2);
         assert_eq!(vec![2, 4, 6], x);
     }
 

src/linalg/ndarray/matrix.rs

@@ -217,7 +217,7 @@ mod tests {
     fn test_iterator() {
         let a = arr2(&[[1, 2, 3], [4, 5, 6]]);
 
-        let v: Vec<i32> = a.iterator(0).map(|&v| v).collect();
+        let v: Vec<i32> = a.iterator(0).copied().collect();
         assert_eq!(v, vec!(1, 2, 3, 4, 5, 6));
     }
 
@@ -236,7 +236,7 @@ mod tests {
         let x = arr2(&[[1, 2, 3], [4, 5, 6]]);
         let x_slice = Array2::slice(&x, 0..2, 1..2);
         assert_eq!((2, 1), x_slice.shape());
-        let v: Vec<i32> = x_slice.iterator(0).map(|&v| v).collect();
+        let v: Vec<i32> = x_slice.iterator(0).copied().collect();
         assert_eq!(v, [2, 5]);
     }
 
@@ -245,11 +245,11 @@ mod tests {
         let x = arr2(&[[1, 2, 3], [4, 5, 6]]);
         let x_slice = Array2::slice(&x, 0..2, 0..3);
         assert_eq!(
-            x_slice.iterator(0).map(|&v| v).collect::<Vec<i32>>(),
+            x_slice.iterator(0).copied().collect::<Vec<i32>>(),
             vec![1, 2, 3, 4, 5, 6]
         );
         assert_eq!(
-            x_slice.iterator(1).map(|&v| v).collect::<Vec<i32>>(),
+            x_slice.iterator(1).copied().collect::<Vec<i32>>(),
             vec![1, 4, 2, 5, 3, 6]
         );
     }
@@ -279,8 +279,8 @@ mod tests {
     fn test_c_from_iterator() {
         let data = vec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12];
         let a: NDArray2<i32> = Array2::from_iterator(data.clone().into_iter(), 4, 3, 0);
-        println!("{}", a);
+        println!("{a}");
         let a: NDArray2<i32> = Array2::from_iterator(data.into_iter(), 4, 3, 1);
-        println!("{}", a);
+        println!("{a}");
     }
 }

src/linalg/ndarray/vector.rs

@@ -152,7 +152,7 @@ mod tests {
     fn test_iterator() {
         let a = arr1(&[1, 2, 3]);
 
-        let v: Vec<i32> = a.iterator(0).map(|&v| v).collect();
+        let v: Vec<i32> = a.iterator(0).copied().collect();
         assert_eq!(v, vec!(1, 2, 3));
     }
 

src/linalg/traits/evd.rs

@@ -66,7 +66,7 @@ pub trait EVDDecomposable<T: Number + RealNumber>: Array2<T> {
     fn evd_mut(mut self, symmetric: bool) -> Result<EVD<T, Self>, Failed> {
         let (nrows, ncols) = self.shape();
         if ncols != nrows {
-            panic!("Matrix is not square: {} x {}", nrows, ncols);
+            panic!("Matrix is not square: {nrows} x {ncols}");
         }
 
         let n = nrows;
@@ -837,10 +837,8 @@ mod tests {
             evd.V.abs(),
             epsilon = 1e-4
         ));
-        for i in 0..eigen_values.len() {
+        for (i, eigen_values_i) in eigen_values.iter().enumerate() {
-            assert!((eigen_values[i] - evd.d[i]).abs() < 1e-4);
+            assert!((eigen_values_i - evd.d[i]).abs() < 1e-4);
-        }
-        for i in 0..eigen_values.len() {
             assert!((0f64 - evd.e[i]).abs() < std::f64::EPSILON);
         }
     }
@@ -871,10 +869,8 @@ mod tests {
             evd.V.abs(),
             epsilon = 1e-4
         ));
-        for i in 0..eigen_values.len() {
+        for (i, eigen_values_i) in eigen_values.iter().enumerate() {
-            assert!((eigen_values[i] - evd.d[i]).abs() < 1e-4);
+            assert!((eigen_values_i - evd.d[i]).abs() < 1e-4);
-        }
-        for i in 0..eigen_values.len() {
             assert!((0f64 - evd.e[i]).abs() < std::f64::EPSILON);
         }
     }
@@ -908,11 +904,11 @@ mod tests {
             evd.V.abs(),
             epsilon = 1e-4
         ));
-        for i in 0..eigen_values_d.len() {
+        for (i, eigen_values_d_i) in eigen_values_d.iter().enumerate() {
-            assert!((eigen_values_d[i] - evd.d[i]).abs() < 1e-4);
+            assert!((eigen_values_d_i - evd.d[i]).abs() < 1e-4);
         }
-        for i in 0..eigen_values_e.len() {
+        for (i, eigen_values_e_i) in eigen_values_e.iter().enumerate() {
-            assert!((eigen_values_e[i] - evd.e[i]).abs() < 1e-4);
+            assert!((eigen_values_e_i - evd.e[i]).abs() < 1e-4);
         }
     }
 }

src/linalg/traits/lu.rs

@@ -126,7 +126,7 @@ impl<T: Number + RealNumber, M: Array2<T>> LU<T, M> {
         let (m, n) = self.LU.shape();
 
         if m != n {
-            panic!("Matrix is not square: {}x{}", m, n);
+            panic!("Matrix is not square: {m}x{n}");
         }
 
         let mut inv = M::zeros(n, n);
@@ -143,10 +143,7 @@ impl<T: Number + RealNumber, M: Array2<T>> LU<T, M> {
         let (b_m, b_n) = b.shape();
 
         if b_m != m {
-            panic!(
+            panic!("Row dimensions do not agree: A is {m} x {n}, but B is {b_m} x {b_n}");
-                "Row dimensions do not agree: A is {} x {}, but B is {} x {}",
-                m, n, b_m, b_n
-            );
         }
 
         if self.singular {

src/linalg/traits/qr.rs

@@ -102,10 +102,7 @@ impl<T: Number + RealNumber, M: Array2<T>> QR<T, M> {
         let (b_nrows, b_ncols) = b.shape();
 
         if b_nrows != m {
-            panic!(
+            panic!("Row dimensions do not agree: A is {m} x {n}, but B is {b_nrows} x {b_ncols}");
-                "Row dimensions do not agree: A is {} x {}, but B is {} x {}",
-                m, n, b_nrows, b_ncols
-            );
         }
 
         if self.singular {

src/linalg/traits/stats.rs

@@ -286,7 +286,7 @@ mod tests {
         }
 
         {
-            let mut m = m.clone();
+            let mut m = m;
             m.standard_scale_mut(&m.mean(1), &m.std(1), 1);
             assert_eq!(&m, &expected_1);
         }

src/linalg/traits/svd.rs

@@ -509,8 +509,8 @@ mod tests {
 
         assert!(relative_eq!(V.abs(), svd.V.abs(), epsilon = 1e-4));
         assert!(relative_eq!(U.abs(), svd.U.abs(), epsilon = 1e-4));
-        for i in 0..s.len() {
+        for (i, s_i) in s.iter().enumerate() {
-            assert!((s[i] - svd.s[i]).abs() < 1e-4);
+            assert!((s_i - svd.s[i]).abs() < 1e-4);
         }
     }
     #[cfg_attr(
@@ -713,8 +713,8 @@ mod tests {
 
         assert!(relative_eq!(V.abs(), svd.V.abs(), epsilon = 1e-4));
         assert!(relative_eq!(U.abs(), svd.U.abs(), epsilon = 1e-4));
-        for i in 0..s.len() {
+        for (i, s_i) in s.iter().enumerate() {
-            assert!((s[i] - svd.s[i]).abs() < 1e-4);
+            assert!((s_i - svd.s[i]).abs() < 1e-4);
         }
     }
     #[cfg_attr(

src/linear/elastic_net.rs

@@ -425,10 +425,7 @@ impl<TX: FloatNumber + RealNumber, TY: Number, X: Array2<TX>, Y: Array1<TY>>
 
         for (i, col_std_i) in col_std.iter().enumerate() {
             if (*col_std_i - TX::zero()).abs() < TX::epsilon() {
-                return Err(Failed::fit(&format!(
+                return Err(Failed::fit(&format!("Cannot rescale constant column {i}")));
-                    "Cannot rescale constant column {}",
-                    i
-                )));
             }
         }
 

src/linear/lasso.rs

@@ -356,10 +356,7 @@ impl<TX: FloatNumber + RealNumber, TY: Number, X: Array2<TX>, Y: Array1<TY>> Las
 
         for (i, col_std_i) in col_std.iter().enumerate() {
             if (*col_std_i - TX::zero()).abs() < TX::epsilon() {
-                return Err(Failed::fit(&format!(
+                return Err(Failed::fit(&format!("Cannot rescale constant column {i}")));
-                    "Cannot rescale constant column {}",
-                    i
-                )));
             }
         }
 

src/linear/logistic_regression.rs

@@ -71,19 +71,14 @@ use crate::optimization::line_search::Backtracking;
 use crate::optimization::FunctionOrder;
 
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Debug, Clone, Eq, PartialEq)]
+#[derive(Debug, Clone, Eq, PartialEq, Default)]
 /// Solver options for Logistic regression. Right now only LBFGS solver is supported.
 pub enum LogisticRegressionSolverName {
     /// Limited-memory Broyden–Fletcher–Goldfarb–Shanno method, see [LBFGS paper](http://users.iems.northwestern.edu/~nocedal/lbfgsb.html)
+    #[default]
     LBFGS,
 }
 
-impl Default for LogisticRegressionSolverName {
-    fn default() -> Self {
-        LogisticRegressionSolverName::LBFGS
-    }
-}
-
 /// Logistic Regression parameters
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
@@ -449,8 +444,7 @@ impl<TX: Number + FloatNumber + RealNumber, TY: Number + Ord, X: Array2<TX>, Y:
 
         match k.cmp(&2) {
             Ordering::Less => Err(Failed::fit(&format!(
-                "incorrect number of classes: {}. Should be >= 2.",
+                "incorrect number of classes: {k}. Should be >= 2."
-                k
             ))),
             Ordering::Equal => {
                 let x0 = Vec::zeros(num_attributes + 1);
@@ -636,19 +630,19 @@ mod tests {
 
         assert!((g[0] + 33.000068218163484).abs() < std::f64::EPSILON);
 
-        let f = objective.f(&vec![1., 2., 3., 4., 5., 6., 7., 8., 9.]);
+        let f = objective.f(&[1., 2., 3., 4., 5., 6., 7., 8., 9.]);
 
         assert!((f - 408.0052230582765).abs() < std::f64::EPSILON);
 
         let objective_reg = MultiClassObjectiveFunction {
             x: &x,
-            y: y.clone(),
+            y,
             k: 3,
             alpha: 1.0,
             _phantom_t: PhantomData,
         };
 
-        let f = objective_reg.f(&vec![1., 2., 3., 4., 5., 6., 7., 8., 9.]);
+        let f = objective_reg.f(&[1., 2., 3., 4., 5., 6., 7., 8., 9.]);
         assert!((f - 487.5052).abs() < 1e-4);
 
         objective_reg.df(&mut g, &vec![1., 2., 3., 4., 5., 6., 7., 8., 9.]);
@@ -697,18 +691,18 @@ mod tests {
         assert!((g[1] - 10.239000702928523).abs() < std::f64::EPSILON);
         assert!((g[2] - 3.869294270156324).abs() < std::f64::EPSILON);
 
-        let f = objective.f(&vec![1., 2., 3.]);
+        let f = objective.f(&[1., 2., 3.]);
 
         assert!((f - 59.76994756647412).abs() < std::f64::EPSILON);
 
         let objective_reg = BinaryObjectiveFunction {
             x: &x,
-            y: y.clone(),
+            y,
             alpha: 1.0,
             _phantom_t: PhantomData,
         };
 
-        let f = objective_reg.f(&vec![1., 2., 3.]);
+        let f = objective_reg.f(&[1., 2., 3.]);
         assert!((f - 62.2699).abs() < 1e-4);
 
         objective_reg.df(&mut g, &vec![1., 2., 3.]);

src/linear/ridge_regression.rs

@@ -71,21 +71,16 @@ use crate::numbers::basenum::Number;
 use crate::numbers::realnum::RealNumber;
 
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Debug, Clone, Eq, PartialEq)]
+#[derive(Debug, Clone, Eq, PartialEq, Default)]
 /// Approach to use for estimation of regression coefficients. Cholesky is more efficient but SVD is more stable.
 pub enum RidgeRegressionSolverName {
     /// Cholesky decomposition, see [Cholesky](../../linalg/cholesky/index.html)
+    #[default]
     Cholesky,
     /// 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)]
@@ -384,10 +379,7 @@ impl<
 
         for (i, col_std_i) in col_std.iter().enumerate() {
             if (*col_std_i - TX::zero()).abs() < TX::epsilon() {
-                return Err(Failed::fit(&format!(
+                return Err(Failed::fit(&format!("Cannot rescale constant column {i}")));
-                    "Cannot rescale constant column {}",
-                    i
-                )));
             }
         }
 

src/metrics/cluster_hcv.rs

@@ -98,8 +98,8 @@ mod tests {
         let mut scores = HCVScore::new();
         scores.compute(&v1, &v2);
 
-        assert!((0.2548 - scores.homogeneity.unwrap() as f64).abs() < 1e-4);
+        assert!((0.2548 - scores.homogeneity.unwrap()).abs() < 1e-4);
-        assert!((0.5440 - scores.completeness.unwrap() as f64).abs() < 1e-4);
+        assert!((0.5440 - scores.completeness.unwrap()).abs() < 1e-4);
-        assert!((0.3471 - scores.v_measure.unwrap() as f64).abs() < 1e-4);
+        assert!((0.3471 - scores.v_measure.unwrap()).abs() < 1e-4);
     }
 }

src/metrics/cluster_helpers.rs

@@ -125,7 +125,7 @@ mod tests {
     fn entropy_test() {
         let v1 = vec![0, 0, 1, 1, 2, 0, 4];
 
-        assert!((1.2770 - entropy(&v1).unwrap() as f64).abs() < 1e-4);
+        assert!((1.2770 - entropy(&v1).unwrap()).abs() < 1e-4);
     }
 
     #[cfg_attr(

src/metrics/f1.rs

@@ -95,8 +95,8 @@ mod tests {
         let score1: f64 = F1::new_with(beta).get_score(&y_true, &y_pred);
         let score2: f64 = F1::new_with(beta).get_score(&y_true, &y_true);
 
-        println!("{:?}", score1);
+        println!("{score1:?}");
-        println!("{:?}", score2);
+        println!("{score2:?}");
 
         assert!((score1 - 0.57142857).abs() < 1e-8);
         assert!((score2 - 1.0).abs() < 1e-8);

src/model_selection/kfold.rs

@@ -213,17 +213,17 @@ mod tests {
 
         for t in &test_masks[0][0..11] {
             // TODO: this can be prob done better
-            assert_eq!(*t, true)
+            assert!(*t)
         }
         for t in &test_masks[0][11..22] {
-            assert_eq!(*t, false)
+            assert!(!*t)
         }
 
         for t in &test_masks[1][0..11] {
-            assert_eq!(*t, false)
+            assert!(!*t)
         }
         for t in &test_masks[1][11..22] {
-            assert_eq!(*t, true)
+            assert!(*t)
         }
     }
 
@@ -283,9 +283,7 @@ mod tests {
             (vec![0, 1, 2, 3, 7, 8, 9], vec![4, 5, 6]),
             (vec![0, 1, 2, 3, 4, 5, 6], vec![7, 8, 9]),
         ];
-        for ((train, test), (expected_train, expected_test)) in
+        for ((train, test), (expected_train, expected_test)) in k.split(&x).zip(expected) {
-            k.split(&x).into_iter().zip(expected)
-        {
             assert_eq!(test, expected_test);
             assert_eq!(train, expected_train);
         }
@@ -307,9 +305,7 @@ mod tests {
             (vec![0, 1, 2, 3, 7, 8, 9], vec![4, 5, 6]),
             (vec![0, 1, 2, 3, 4, 5, 6], vec![7, 8, 9]),
         ];
-        for ((train, test), (expected_train, expected_test)) in
+        for ((train, test), (expected_train, expected_test)) in k.split(&x).zip(expected) {
-            k.split(&x).into_iter().zip(expected)
-        {
             assert_eq!(test.len(), expected_test.len());
             assert_eq!(train.len(), expected_train.len());
         }

src/model_selection/mod.rs

@@ -169,7 +169,7 @@ pub fn train_test_split<
     let n_test = ((n as f32) * test_size) as usize;
 
     if n_test < 1 {
-        panic!("number of sample is too small {}", n);
+        panic!("number of sample is too small {n}");
     }
 
     let mut indices: Vec<usize> = (0..n).collect();
@@ -553,6 +553,6 @@ mod tests {
             &accuracy,
         )
         .unwrap();
-        println!("{:?}", results);
+        println!("{results:?}");
     }
 }

src/naive_bayes/bernoulli.rs

@@ -271,21 +271,18 @@ impl<TY: Number + Ord + Unsigned> BernoulliNBDistribution<TY> {
         let y_samples = y.shape();
         if y_samples != n_samples {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{n_samples}], |y|=[{y_samples}]"
-                n_samples, y_samples
             )));
         }
 
         if n_samples == 0 {
             return Err(Failed::fit(&format!(
-                "Size of x and y should greater than 0; |x|=[{}]",
+                "Size of x and y should greater than 0; |x|=[{n_samples}]"
-                n_samples
             )));
         }
         if alpha < 0f64 {
             return Err(Failed::fit(&format!(
-                "Alpha should be greater than 0; |alpha|=[{}]",
+                "Alpha should be greater than 0; |alpha|=[{alpha}]"
-                alpha
             )));
         }
 
@@ -318,8 +315,7 @@ impl<TY: Number + Ord + Unsigned> BernoulliNBDistribution<TY> {
                 feature_in_class_counter[class_index][idx] +=
                     row_i.to_usize().ok_or_else(|| {
                         Failed::fit(&format!(
-                            "Elements of the matrix should be 1.0 or 0.0 |found|=[{}]",
+                            "Elements of the matrix should be 1.0 or 0.0 |found|=[{row_i}]"
-                            row_i
                         ))
                     })?;
             }

src/naive_bayes/categorical.rs

@@ -158,8 +158,7 @@ impl<T: Number + Unsigned> CategoricalNBDistribution<T> {
     pub fn fit<X: Array2<T>, Y: Array1<T>>(x: &X, y: &Y, alpha: f64) -> Result<Self, Failed> {
         if alpha < 0f64 {
             return Err(Failed::fit(&format!(
-                "alpha should be >= 0, alpha=[{}]",
+                "alpha should be >= 0, alpha=[{alpha}]"
-                alpha
             )));
         }
 
@@ -167,15 +166,13 @@ impl<T: Number + Unsigned> CategoricalNBDistribution<T> {
         let y_samples = y.shape();
         if y_samples != n_samples {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{n_samples}], |y|=[{y_samples}]"
-                n_samples, y_samples
             )));
         }
 
         if n_samples == 0 {
             return Err(Failed::fit(&format!(
-                "Size of x and y should greater than 0; |x|=[{}]",
+                "Size of x and y should greater than 0; |x|=[{n_samples}]"
-                n_samples
             )));
         }
         let y: Vec<usize> = y.iterator(0).map(|y_i| y_i.to_usize().unwrap()).collect();
@@ -202,8 +199,7 @@ impl<T: Number + Unsigned> CategoricalNBDistribution<T> {
                 .max()
                 .ok_or_else(|| {
                     Failed::fit(&format!(
-                        "Failed to get the categories for feature = {}",
+                        "Failed to get the categories for feature = {feature}"
-                        feature
                     ))
                 })?;
             n_categories.push(feature_max + 1);
@@ -429,7 +425,6 @@ mod tests {
     fn search_parameters() {
         let parameters = CategoricalNBSearchParameters {
             alpha: vec![1., 2.],
-            ..Default::default()
         };
         let mut iter = parameters.into_iter();
         let next = iter.next().unwrap();

src/naive_bayes/gaussian.rs

@@ -185,15 +185,13 @@ impl<TY: Number + Ord + Unsigned> GaussianNBDistribution<TY> {
         let y_samples = y.shape();
         if y_samples != n_samples {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{n_samples}], |y|=[{y_samples}]"
-                n_samples, y_samples
             )));
         }
 
         if n_samples == 0 {
             return Err(Failed::fit(&format!(
-                "Size of x and y should greater than 0; |x|=[{}]",
+                "Size of x and y should greater than 0; |x|=[{n_samples}]"
-                n_samples
             )));
         }
         let (class_labels, indices) = y.unique_with_indices();
@@ -375,7 +373,6 @@ mod tests {
     fn search_parameters() {
         let parameters = GaussianNBSearchParameters {
             priors: vec![Some(vec![1.]), Some(vec![2.])],
-            ..Default::default()
         };
         let mut iter = parameters.into_iter();
         let next = iter.next().unwrap();

src/naive_bayes/multinomial.rs

@@ -220,21 +220,18 @@ impl<TY: Number + Ord + Unsigned> MultinomialNBDistribution<TY> {
         let y_samples = y.shape();
         if y_samples != n_samples {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{n_samples}], |y|=[{y_samples}]"
-                n_samples, y_samples
             )));
         }
 
         if n_samples == 0 {
             return Err(Failed::fit(&format!(
-                "Size of x and y should greater than 0; |x|=[{}]",
+                "Size of x and y should greater than 0; |x|=[{n_samples}]"
-                n_samples
             )));
         }
         if alpha < 0f64 {
             return Err(Failed::fit(&format!(
-                "Alpha should be greater than 0; |alpha|=[{}]",
+                "Alpha should be greater than 0; |alpha|=[{alpha}]"
-                alpha
             )));
         }
 
@@ -266,8 +263,7 @@ impl<TY: Number + Ord + Unsigned> MultinomialNBDistribution<TY> {
                 feature_in_class_counter[class_index][idx] +=
                     row_i.to_usize().ok_or_else(|| {
                         Failed::fit(&format!(
-                            "Elements of the matrix should be convertible to usize |found|=[{}]",
+                            "Elements of the matrix should be convertible to usize |found|=[{row_i}]"
-                            row_i
                         ))
                     })?;
             }

src/neighbors/knn_classifier.rs

@@ -236,8 +236,7 @@ impl<TX: Number, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>, D: Distance<Vec
 
         if x_n != y_n {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{x_n}], |y|=[{y_n}]"
-                x_n, y_n
             )));
         }
 

src/neighbors/knn_regressor.rs

@@ -224,8 +224,7 @@ impl<TX: Number, TY: Number, X: Array2<TX>, Y: Array1<TY>, D: Distance<Vec<TX>>>
 
         if x_n != y_n {
             return Err(Failed::fit(&format!(
-                "Size of x should equal size of y; |x|=[{}], |y|=[{}]",
+                "Size of x should equal size of y; |x|=[{x_n}], |y|=[{y_n}]"
-                x_n, y_n
             )));
         }
 

src/neighbors/mod.rs

@@ -49,20 +49,15 @@ pub type KNNAlgorithmName = crate::algorithm::neighbour::KNNAlgorithmName;
 
 /// Weight function that is used to determine estimated value.
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Default)]
 pub enum KNNWeightFunction {
     /// All k nearest points are weighted equally
+    #[default]
     Uniform,
     /// k nearest points are weighted by the inverse of their distance. Closer neighbors will have a greater influence than neighbors which are further away.
     Distance,
 }
 
-impl Default for KNNWeightFunction {
-    fn default() -> Self {
-        KNNWeightFunction::Uniform
-    }
-}
-
 impl KNNWeightFunction {
     fn calc_weights(&self, distances: Vec<f64>) -> std::vec::Vec<f64> {
         match *self {

src/numbers/realnum.rs

@@ -2,9 +2,13 @@
 //! Most algorithms in `smartcore` rely on basic linear algebra operations like dot product, matrix decomposition and other subroutines that are defined for a set of real numbers, ℝ.
 //! This module defines real number and some useful functions that are used in [Linear Algebra](../../linalg/index.html) module.
 
+use rand::rngs::SmallRng;
+use rand::{Rng, SeedableRng};
+
 use num_traits::Float;
 
 use crate::numbers::basenum::Number;
+use crate::rand_custom::get_rng_impl;
 
 /// Defines real number
 /// <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS_CHTML"></script>
@@ -63,8 +67,12 @@ impl RealNumber for f64 {
     }
 
     fn rand() -> f64 {
-        // TODO: to be implemented, see issue smartcore#214
+        let mut small_rng = get_rng_impl(None);
-        1.0
+
+        let mut rngs: Vec<SmallRng> = (0..3)
+            .map(|_| SmallRng::from_rng(&mut small_rng).unwrap())
+            .collect();
+        rngs[0].gen::<f64>()
     }
 
     fn two() -> Self {
@@ -108,7 +116,12 @@ impl RealNumber for f32 {
     }
 
     fn rand() -> f32 {
-        1.0
+        let mut small_rng = get_rng_impl(None);
+
+        let mut rngs: Vec<SmallRng> = (0..3)
+            .map(|_| SmallRng::from_rng(&mut small_rng).unwrap())
+            .collect();
+        rngs[0].gen::<f32>()
     }
 
     fn two() -> Self {
@@ -149,4 +162,14 @@ mod tests {
     fn f64_from_string() {
         assert_eq!(f64::from_str("1.111111111").unwrap(), 1.111111111)
     }
+
+    #[test]
+    fn f64_rand() {
+        f64::rand();
+    }
+
+    #[test]
+    fn f32_rand() {
+        f32::rand();
+    }
 }

src/optimization/first_order/gradient_descent.rs

@@ -113,12 +113,13 @@ mod tests {
             g[1] = 200. * (x[1] - x[0].powf(2.));
         };
 
-        let mut ls: Backtracking<f64> = Default::default();
+        let ls: Backtracking<f64> = Backtracking::<f64> {
-        ls.order = FunctionOrder::THIRD;
+            order: FunctionOrder::THIRD,
+            ..Default::default()
+        };
         let optimizer: GradientDescent = Default::default();
 
         let result = optimizer.optimize(&f, &df, &x0, &ls);
-        println!("{:?}", result);
 
         assert!((result.f_x - 0.0).abs() < 1e-5);
         assert!((result.x[0] - 1.0).abs() < 1e-2);

src/optimization/first_order/lbfgs.rs

@@ -196,9 +196,9 @@ impl LBFGS {
     }
 
     ///
-    fn update_hessian<'a, T: FloatNumber, X: Array1<T>>(
+    fn update_hessian<T: FloatNumber, X: Array1<T>>(
         &self,
-        _: &'a DF<'_, X>,
+        _: &DF<'_, X>,
         state: &mut LBFGSState<T, X>,
     ) {
         state.dg = state.x_df.sub(&state.x_df_prev);
@@ -291,8 +291,10 @@ mod tests {
             g[0] = -2. * (1. - x[0]) - 400. * (x[1] - x[0].powf(2.)) * x[0];
             g[1] = 200. * (x[1] - x[0].powf(2.));
         };
-        let mut ls: Backtracking<f64> = Default::default();
+        let ls: Backtracking<f64> = Backtracking::<f64> {
-        ls.order = FunctionOrder::THIRD;
+            order: FunctionOrder::THIRD,
+            ..Default::default()
+        };
         let optimizer: LBFGS = Default::default();
 
         let result = optimizer.optimize(&f, &df, &x0, &ls);

src/preprocessing/categorical.rs

@@ -132,8 +132,7 @@ impl OneHotEncoder {
                     data.copy_col_as_vec(idx, &mut col_buf);
                     if !validate_col_is_categorical(&col_buf) {
                         let msg = format!(
-                            "Column {} of data matrix containts non categorizable (integer) values",
+                            "Column {idx} of data matrix containts non categorizable (integer) values"
-                            idx
                         );
                         return Err(Failed::fit(&msg[..]));
                     }
@@ -182,7 +181,7 @@ impl OneHotEncoder {
                 match oh_vec {
                     None => {
                         // Since we support T types, bad value in a series causes in to be invalid
-                        let msg = format!("At least one value in column {} doesn't conform to category definition", old_cidx);
+                        let msg = format!("At least one value in column {old_cidx} doesn't conform to category definition");
                         return Err(Failed::transform(&msg[..]));
                     }
                     Some(v) => {
@@ -338,11 +337,7 @@ mod tests {
         ]);
 
         let params = OneHotEncoderParams::from_cat_idx(&[1]);
-        match OneHotEncoder::fit(&m, params) {
+        let result = OneHotEncoder::fit(&m, params);
-            Err(_) => {
+        assert!(result.is_err());
-                assert!(true);
-            }
-            _ => assert!(false),
-        }
     }
 }

src/preprocessing/numerical.rs

@@ -294,7 +294,7 @@ mod tests {
                     &[0.5708488802, 0.1846414616, 0.9590802982, 0.5591871046],
                     &[0.8387612750, 0.5754861361, 0.5537109852, 0.1077646442],
                 ]));
-            println!("{}", transformed_values);
+            println!("{transformed_values}");
             assert!(transformed_values.approximate_eq(
                 &DenseMatrix::from_2d_array(&[
                     &[-1.1154020653, -0.4031985330, 0.9284605204, -0.4271473866],

src/preprocessing/series_encoder.rs

@@ -206,7 +206,7 @@ mod tests {
     #[test]
     fn from_categories() {
         let fake_categories: Vec<usize> = vec![1, 2, 3, 4, 5, 3, 5, 3, 1, 2, 4];
-        let it = fake_categories.iter().map(|&a| a);
+        let it = fake_categories.iter().copied();
         let enc = CategoryMapper::<usize>::fit_to_iter(it);
         let oh_vec: Vec<f64> = match enc.get_one_hot(&1) {
             None => panic!("Wrong categories"),
@@ -218,8 +218,8 @@ mod tests {
 
     fn build_fake_str_enc<'a>() -> CategoryMapper<&'a str> {
         let fake_category_pos = vec!["background", "dog", "cat"];
-        let enc = CategoryMapper::<&str>::from_positional_category_vec(fake_category_pos);
+
-        enc
+        CategoryMapper::<&str>::from_positional_category_vec(fake_category_pos)
     }
     #[cfg_attr(
         all(target_arch = "wasm32", not(target_os = "wasi")),
@@ -275,7 +275,7 @@ mod tests {
         let lab = enc.invert_one_hot(res).unwrap();
         assert_eq!(lab, "dog");
         if let Err(e) = enc.invert_one_hot(vec![0.0, 0.0, 0.0]) {
-            let pos_entries = format!("Expected a single positive entry, 0 entires found");
+            let pos_entries = "Expected a single positive entry, 0 entires found".to_string();
             assert_eq!(e, Failed::transform(&pos_entries[..]));
         };
     }

src/readers/csv.rs

@@ -83,7 +83,7 @@ where
     Matrix: Array2<T>,
 {
     let csv_text = read_string_from_source(source)?;
-    let rows: Vec<Vec<T>> = extract_row_vectors_from_csv_text::<T, RowVector, Matrix>(
+    let rows: Vec<Vec<T>> = extract_row_vectors_from_csv_text(
         &csv_text,
         &definition,
         detect_row_format(&csv_text, &definition)?,
@@ -103,12 +103,7 @@ where
 
 /// Given a string containing the contents of a csv file, extract its value
 /// into row-vectors.
-fn extract_row_vectors_from_csv_text<
+fn extract_row_vectors_from_csv_text<'a, T: Number + RealNumber + std::str::FromStr>(
-    'a,
-    T: Number + RealNumber + std::str::FromStr,
-    RowVector: Array1<T>,
-    Matrix: Array2<T>,
->(
     csv_text: &'a str,
     definition: &'a CSVDefinition<'_>,
     row_format: CSVRowFormat<'_>,
@@ -167,7 +162,7 @@ where
 }
 
 /// Ensure that a string containing a csv row conforms to a specified row format.
-fn validate_csv_row<'a>(row: &'a str, row_format: &CSVRowFormat<'_>) -> Result<(), ReadingError> {
+fn validate_csv_row(row: &str, row_format: &CSVRowFormat<'_>) -> Result<(), ReadingError> {
     let actual_number_of_fields = row.split(row_format.field_seperator).count();
     if row_format.n_fields == actual_number_of_fields {
         Ok(())
@@ -208,7 +203,7 @@ where
     match value_string.parse::<T>().ok() {
         Some(value) => Ok(value),
         None => Err(ReadingError::InvalidField {
-            msg: format!("Value '{}' could not be read.", value_string,),
+            msg: format!("Value '{value_string}' could not be read.",),
         }),
     }
 }
@@ -305,12 +300,11 @@ mod tests {
     }
     mod extract_row_vectors_from_csv_text {
         use super::super::{extract_row_vectors_from_csv_text, CSVDefinition, CSVRowFormat};
-        use crate::linalg::basic::matrix::DenseMatrix;
 
         #[test]
         fn read_default_csv() {
             assert_eq!(
-                extract_row_vectors_from_csv_text::<f64, Vec<_>, DenseMatrix<_>>(
+                extract_row_vectors_from_csv_text::<f64>(
                     "column 1, column 2, column3\n1.0,2.0,3.0\n4.0,5.0,6.0",
                     &CSVDefinition::default(),
                     CSVRowFormat {

src/svm/svc.rs

@@ -322,19 +322,26 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX> + 'a, Y: Array
         let (n, _) = x.shape();
         let mut y_hat: Vec<TX> = Array1::zeros(n);
 
+        let mut row = Vec::with_capacity(n);
         for i in 0..n {
-            let row_pred: TX =
+            row.clear();
-                self.predict_for_row(Vec::from_iterator(x.get_row(i).iterator(0).copied(), n));
+            row.extend(x.get_row(i).iterator(0).copied());
+            let row_pred: TX = self.predict_for_row(&row);
             y_hat.set(i, row_pred);
         }
 
         Ok(y_hat)
     }
 
-    fn predict_for_row(&self, x: Vec<TX>) -> TX {
+    fn predict_for_row(&self, x: &[TX]) -> TX {
         let mut f = self.b.unwrap();
 
+        let xi: Vec<_> = x.iter().map(|e| e.to_f64().unwrap()).collect();
         for i in 0..self.instances.as_ref().unwrap().len() {
+            let xj: Vec<_> = self.instances.as_ref().unwrap()[i]
+                .iter()
+                .map(|e| e.to_f64().unwrap())
+                .collect();
             f += self.w.as_ref().unwrap()[i]
                 * TX::from(
                     self.parameters
@@ -343,13 +350,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX> + 'a, Y: Array
                         .kernel
                         .as_ref()
                         .unwrap()
-                        .apply(
+                        .apply(&xi, &xj)
-                            &x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                            &self.instances.as_ref().unwrap()[i]
-                                .iter()
-                                .map(|e| e.to_f64().unwrap())
-                                .collect(),
-                        )
                         .unwrap(),
                 )
                 .unwrap();
@@ -472,14 +473,12 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
         let tol = self.parameters.tol;
         let good_enough = TX::from_i32(1000).unwrap();
 
+        let mut x = Vec::with_capacity(n);
         for _ in 0..self.parameters.epoch {
             for i in self.permutate(n) {
-                self.process(
+                x.clear();
-                    i,
+                x.extend(self.x.get_row(i).iterator(0).take(n).copied());
-                    Vec::from_iterator(self.x.get_row(i).iterator(0).copied(), n),
+                self.process(i, &x, *self.y.get(i), &mut cache);
-                    *self.y.get(i),
-                    &mut cache,
-                );
                 loop {
                     self.reprocess(tol, &mut cache);
                     self.find_min_max_gradient();
@@ -511,24 +510,17 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
         let mut cp = 0;
         let mut cn = 0;
 
+        let mut x = Vec::with_capacity(n);
         for i in self.permutate(n) {
+            x.clear();
+            x.extend(self.x.get_row(i).iterator(0).take(n).copied());
             if *self.y.get(i) == TY::one() && cp < few {
-                if self.process(
+                if self.process(i, &x, *self.y.get(i), cache) {
-                    i,
-                    Vec::from_iterator(self.x.get_row(i).iterator(0).copied(), n),
-                    *self.y.get(i),
-                    cache,
-                ) {
                     cp += 1;
                 }
             } else if *self.y.get(i) == TY::from(-1).unwrap()
                 && cn < few
-                && self.process(
+                && self.process(i, &x, *self.y.get(i), cache)
-                    i,
-                    Vec::from_iterator(self.x.get_row(i).iterator(0).copied(), n),
-                    *self.y.get(i),
-                    cache,
-                )
             {
                 cn += 1;
             }
@@ -539,7 +531,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
         }
     }
 
-    fn process(&mut self, i: usize, x: Vec<TX>, y: TY, cache: &mut Cache<TX, TY, X, Y>) -> bool {
+    fn process(&mut self, i: usize, x: &[TX], y: TY, cache: &mut Cache<TX, TY, X, Y>) -> bool {
         for j in 0..self.sv.len() {
             if self.sv[j].index == i {
                 return true;
@@ -551,15 +543,14 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
         let mut cache_values: Vec<((usize, usize), TX)> = Vec::new();
 
         for v in self.sv.iter() {
+            let xi: Vec<_> = v.x.iter().map(|e| e.to_f64().unwrap()).collect();
+            let xj: Vec<_> = x.iter().map(|e| e.to_f64().unwrap()).collect();
             let k = self
                 .parameters
                 .kernel
                 .as_ref()
                 .unwrap()
-                .apply(
+                .apply(&xi, &xj)
-                    &v.x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                    &x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                )
                 .unwrap();
             cache_values.push(((i, v.index), TX::from(k).unwrap()));
             g -= v.alpha * k;
@@ -578,7 +569,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
             cache.insert(v.0, v.1.to_f64().unwrap());
         }
 
-        let x_f64 = x.iter().map(|e| e.to_f64().unwrap()).collect();
+        let x_f64: Vec<_> = x.iter().map(|e| e.to_f64().unwrap()).collect();
         let k_v = self
             .parameters
             .kernel
@@ -701,8 +692,10 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                 let km = sv1.k;
                 let gm = sv1.grad;
                 let mut best = 0f64;
+                let xi: Vec<_> = sv1.x.iter().map(|e| e.to_f64().unwrap()).collect();
                 for i in 0..self.sv.len() {
                     let v = &self.sv[i];
+                    let xj: Vec<_> = v.x.iter().map(|e| e.to_f64().unwrap()).collect();
                     let z = v.grad - gm;
                     let k = cache.get(
                         sv1,
@@ -711,10 +704,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                             .kernel
                             .as_ref()
                             .unwrap()
-                            .apply(
+                            .apply(&xi, &xj)
-                                &sv1.x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                                &v.x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                            )
                             .unwrap(),
                     );
                     let mut curv = km + v.k - 2f64 * k;
@@ -732,6 +722,12 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                     }
                 }
 
+                let xi: Vec<_> = self.sv[idx_1]
+                    .x
+                    .iter()
+                    .map(|e| e.to_f64().unwrap())
+                    .collect::<Vec<_>>();
+
                 idx_2.map(|idx_2| {
                     (
                         idx_1,
@@ -742,16 +738,12 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                                 .as_ref()
                                 .unwrap()
                                 .apply(
-                                    &self.sv[idx_1]
+                                    &xi,
-                                        .x
-                                        .iter()
-                                        .map(|e| e.to_f64().unwrap())
-                                        .collect(),
                                     &self.sv[idx_2]
                                         .x
                                         .iter()
                                         .map(|e| e.to_f64().unwrap())
-                                        .collect(),
+                                        .collect::<Vec<_>>(),
                                 )
                                 .unwrap()
                         }),
@@ -765,8 +757,11 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                 let km = sv2.k;
                 let gm = sv2.grad;
                 let mut best = 0f64;
+
+                let xi: Vec<_> = sv2.x.iter().map(|e| e.to_f64().unwrap()).collect();
                 for i in 0..self.sv.len() {
                     let v = &self.sv[i];
+                    let xj: Vec<_> = v.x.iter().map(|e| e.to_f64().unwrap()).collect();
                     let z = gm - v.grad;
                     let k = cache.get(
                         sv2,
@@ -775,10 +770,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                             .kernel
                             .as_ref()
                             .unwrap()
-                            .apply(
+                            .apply(&xi, &xj)
-                                &sv2.x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                                &v.x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                            )
                             .unwrap(),
                     );
                     let mut curv = km + v.k - 2f64 * k;
@@ -797,6 +789,12 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                     }
                 }
 
+                let xj: Vec<_> = self.sv[idx_2]
+                    .x
+                    .iter()
+                    .map(|e| e.to_f64().unwrap())
+                    .collect();
+
                 idx_1.map(|idx_1| {
                     (
                         idx_1,
@@ -811,12 +809,8 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                                         .x
                                         .iter()
                                         .map(|e| e.to_f64().unwrap())
-                                        .collect(),
+                                        .collect::<Vec<_>>(),
-                                    &self.sv[idx_2]
+                                    &xj,
-                                        .x
-                                        .iter()
-                                        .map(|e| e.to_f64().unwrap())
-                                        .collect(),
                                 )
                                 .unwrap()
                         }),
@@ -835,12 +829,12 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                             .x
                             .iter()
                             .map(|e| e.to_f64().unwrap())
-                            .collect(),
+                            .collect::<Vec<_>>(),
                         &self.sv[idx_2]
                             .x
                             .iter()
                             .map(|e| e.to_f64().unwrap())
-                            .collect(),
+                            .collect::<Vec<_>>(),
                     )
                     .unwrap(),
             )),
@@ -895,7 +889,10 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
         self.sv[v1].alpha -= step.to_f64().unwrap();
         self.sv[v2].alpha += step.to_f64().unwrap();
 
+        let xi_v1: Vec<_> = self.sv[v1].x.iter().map(|e| e.to_f64().unwrap()).collect();
+        let xi_v2: Vec<_> = self.sv[v2].x.iter().map(|e| e.to_f64().unwrap()).collect();
         for i in 0..self.sv.len() {
+            let xj: Vec<_> = self.sv[i].x.iter().map(|e| e.to_f64().unwrap()).collect();
             let k2 = cache.get(
                 &self.sv[v2],
                 &self.sv[i],
@@ -903,10 +900,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                     .kernel
                     .as_ref()
                     .unwrap()
-                    .apply(
+                    .apply(&xi_v2, &xj)
-                        &self.sv[v2].x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                        &self.sv[i].x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                    )
                     .unwrap(),
             );
             let k1 = cache.get(
@@ -916,10 +910,7 @@ impl<'a, TX: Number + RealNumber, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>
                     .kernel
                     .as_ref()
                     .unwrap()
-                    .apply(
+                    .apply(&xi_v1, &xj)
-                        &self.sv[v1].x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                        &self.sv[i].x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                    )
                     .unwrap(),
             );
             self.sv[i].grad -= step.to_f64().unwrap() * (k2 - k1);
@@ -983,11 +974,7 @@ mod tests {
             .unwrap();
         let acc = accuracy(&y, &(y_hat.iter().map(|e| e.to_i32().unwrap()).collect()));
 
-        assert!(
+        assert!(acc >= 0.9, "accuracy ({acc}) is not larger or equal to 0.9");
-            acc >= 0.9,
-            "accuracy ({}) is not larger or equal to 0.9",
-            acc
-        );
     }
 
     #[cfg_attr(
@@ -1076,11 +1063,7 @@ mod tests {
 
         let acc = accuracy(&y, &(y_hat.iter().map(|e| e.to_i32().unwrap()).collect()));
 
-        assert!(
+        assert!(acc >= 0.9, "accuracy ({acc}) is not larger or equal to 0.9");
-            acc >= 0.9,
-            "accuracy ({}) is not larger or equal to 0.9",
-            acc
-        );
     }
 
     #[cfg_attr(

src/svm/svr.rs

@@ -248,19 +248,20 @@ impl<'a, T: Number + FloatNumber + PartialOrd, X: Array2<T>, Y: Array1<T>> SVR<'
 
         let mut y_hat: Vec<T> = Vec::<T>::zeros(n);
 
+        let mut x_i = Vec::with_capacity(n);
         for i in 0..n {
-            y_hat.set(
+            x_i.clear();
-                i,
+            x_i.extend(x.get_row(i).iterator(0).copied());
-                self.predict_for_row(Vec::from_iterator(x.get_row(i).iterator(0).copied(), n)),
+            y_hat.set(i, self.predict_for_row(&x_i));
-            );
         }
 
         Ok(y_hat)
     }
 
-    pub(crate) fn predict_for_row(&self, x: Vec<T>) -> T {
+    pub(crate) fn predict_for_row(&self, x: &[T]) -> T {
         let mut f = self.b;
 
+        let xi: Vec<_> = x.iter().map(|e| e.to_f64().unwrap()).collect();
         for i in 0..self.instances.as_ref().unwrap().len() {
             f += self.w.as_ref().unwrap()[i]
                 * T::from(
@@ -270,10 +271,7 @@ impl<'a, T: Number + FloatNumber + PartialOrd, X: Array2<T>, Y: Array1<T>> SVR<'
                         .kernel
                         .as_ref()
                         .unwrap()
-                        .apply(
+                        .apply(&xi, &self.instances.as_ref().unwrap()[i])
-                            &x.iter().map(|e| e.to_f64().unwrap()).collect(),
-                            &self.instances.as_ref().unwrap()[i],
-                        )
                         .unwrap(),
                 )
                 .unwrap()
@@ -662,7 +660,7 @@ mod tests {
         .unwrap();
 
         let t = mean_squared_error(&y_hat, &y);
-        println!("{:?}", t);
+        println!("{t:?}");
         assert!(t < 2.5);
     }
 

src/tree/decision_tree_classifier.rs

@@ -137,16 +137,17 @@ impl<TX: Number + PartialOrd, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>>
         self.classes.as_ref()
     }
     /// Get depth of tree
-    fn depth(&self) -> u16 {
+    pub fn depth(&self) -> u16 {
         self.depth
     }
 }
 
 /// The function to measure the quality of a split.
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Default)]
 pub enum SplitCriterion {
     /// [Gini index](../decision_tree_classifier/index.html)
+    #[default]
     Gini,
     /// [Entropy](../decision_tree_classifier/index.html)
     Entropy,
@@ -154,12 +155,6 @@ pub enum SplitCriterion {
     ClassificationError,
 }
 
-impl Default for SplitCriterion {
-    fn default() -> Self {
-        SplitCriterion::Gini
-    }
-}
-
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 struct Node {
@@ -543,6 +538,10 @@ impl<TX: Number + PartialOrd, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>>
         parameters: DecisionTreeClassifierParameters,
     ) -> Result<DecisionTreeClassifier<TX, TY, X, Y>, Failed> {
         let (x_nrows, num_attributes) = x.shape();
+        if x_nrows != y.shape() {
+            return Err(Failed::fit("Size of x should equal size of y"));
+        }
+
         let samples = vec![1; x_nrows];
         DecisionTreeClassifier::fit_weak_learner(x, y, samples, num_attributes, parameters)
     }
@@ -560,8 +559,7 @@ impl<TX: Number + PartialOrd, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>>
         let k = classes.len();
         if k < 2 {
             return Err(Failed::fit(&format!(
-                "Incorrect number of classes: {}. Should be >= 2.",
+                "Incorrect number of classes: {k}. Should be >= 2."
-                k
             )));
         }
 
@@ -901,15 +899,13 @@ mod tests {
     )]
     #[test]
     fn gini_impurity() {
+        assert!((impurity(&SplitCriterion::Gini, &[7, 3], 10) - 0.42).abs() < std::f64::EPSILON);
         assert!(
-            (impurity(&SplitCriterion::Gini, &vec![7, 3], 10) - 0.42).abs() < std::f64::EPSILON
+            (impurity(&SplitCriterion::Entropy, &[7, 3], 10) - 0.8812908992306927).abs()
-        );
-        assert!(
-            (impurity(&SplitCriterion::Entropy, &vec![7, 3], 10) - 0.8812908992306927).abs()
                 < std::f64::EPSILON
         );
         assert!(
-            (impurity(&SplitCriterion::ClassificationError, &vec![7, 3], 10) - 0.3).abs()
+            (impurity(&SplitCriterion::ClassificationError, &[7, 3], 10) - 0.3).abs()
                 < std::f64::EPSILON
         );
     }
@@ -971,6 +967,17 @@ mod tests {
         );
     }
 
+    #[test]
+    fn test_random_matrix_with_wrong_rownum() {
+        let x_rand: DenseMatrix<f64> = DenseMatrix::<f64>::rand(21, 200);
+
+        let y: Vec<u32> = vec![0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1];
+
+        let fail = DecisionTreeClassifier::fit(&x_rand, &y, Default::default());
+
+        assert!(fail.is_err());
+    }
+
     #[cfg_attr(
         all(target_arch = "wasm32", not(target_os = "wasi")),
         wasm_bindgen_test::wasm_bindgen_test

src/tree/decision_tree_regressor.rs

@@ -18,7 +18,6 @@
 //! Example:
 //!
 //! ```
-//! use rand::thread_rng;
 //! use smartcore::linalg::basic::matrix::DenseMatrix;
 //! use smartcore::tree::decision_tree_regressor::*;
 //!
@@ -422,6 +421,10 @@ impl<TX: Number + PartialOrd, TY: Number, X: Array2<TX>, Y: Array1<TY>>
         parameters: DecisionTreeRegressorParameters,
     ) -> Result<DecisionTreeRegressor<TX, TY, X, Y>, Failed> {
         let (x_nrows, num_attributes) = x.shape();
+        if x_nrows != y.shape() {
+            return Err(Failed::fit("Size of x should equal size of y"));
+        }
+
         let samples = vec![1; x_nrows];
         DecisionTreeRegressor::fit_weak_learner(x, y, samples, num_attributes, parameters)
     }