modify codes suggested by cargo clippy

src/booster/adaboost/adaboost_algorithm.rs

@@ -38,18 +38,18 @@ use std::ops::ControlFlow;
 /// 
 /// # Related information
 /// - As some papers proved, 
-/// `AdaBoost` **approximately maximizes the hard margin.**
+///   `AdaBoost` **approximately maximizes the hard margin.**
 /// 
 /// - [`AdaBoostV`](crate::booster::AdaBoostV), 
-/// a successor of AdaBoost, maximizes the hard margin.
+///   a successor of AdaBoost, maximizes the hard margin.
 /// 
 /// 
 /// ```no_run
 /// use miniboosts::prelude::*;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -259,7 +259,7 @@ impl<F> Booster<F> for AdaBoost<'_, F>
         let quit = if let Some(it) = self.force_quit_at {
             format!("At round {it}")
         } else {
-            format!("-")
+            "-".to_string()
         };
         let info = Vec::from([
             ("# of examples", format!("{}", n_sample)),

src/booster/adaboostv/adaboostv_algorithm.rs

@@ -69,7 +69,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -291,7 +291,7 @@ impl<F> Booster<F> for AdaBoostV<'_, F>
         let quit = if let Some(it) = self.force_quit_at {
             format!("At round {it}")
         } else {
-            format!("-")
+            "-".to_string()
         };
         let info = Vec::from([
             ("# of examples", format!("{}", n_sample)),

src/booster/cerlpboost/cerlpboost_algorithm.rs

@@ -72,7 +72,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")

src/booster/erlpboost/erlpboost_algorithm.rs

@@ -64,7 +64,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")

src/booster/erlpboost/qp_model.rs

@@ -170,7 +170,7 @@ impl QPModel {
 
     /// Returns `true` if `dist[i] > 0` holds for all `i = 1, 2, ..., m.` 
     pub(self) fn all_positive(&self, dist: &[f64]) -> bool {
-        dist.into_iter()
+        dist.iter()
             .copied()
             .all(|d| d > 0f64)
     }
@@ -179,7 +179,7 @@ impl QPModel {
     pub(self) fn build_linear_part_objective(&self, dist: &[f64]) -> Vec<f64> {
         let mut linear = Vec::with_capacity(1 + self.n_examples);
         linear.push(1f64);
-        let iter = dist.into_iter()
+        let iter = dist.iter()
             .copied()
             .map(|di| (1f64 / self.eta) * di.ln());
         linear.extend(iter);

src/booster/gradient_boost/gbm.rs

@@ -44,7 +44,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -149,7 +149,7 @@ impl<'a, F, L> GBM<'a, F, L>
 }
 
 
-impl<'a, F, L> GBM<'a, F, L> {
+impl<F, L> GBM<'_, F, L> {
     /// Returns the maximum iteration
     /// of the `GBM` to find a combined hypothesis
     /// that has error at most `tolerance`.
@@ -194,7 +194,7 @@ impl<F, L> Booster<F> for GBM<'_, F, L>
             ("# of examples", format!("{n_sample}")),
             ("# of features", format!("{n_feature}")),
             ("Tolerance", format!("{}", self.tolerance)),
-            ("Loss", format!("{}", self.loss.name())),
+            ("Loss", self.loss.name().to_string()),
             ("Max iteration", format!("{}", self.max_iter)),
         ]);
         Some(info)
@@ -236,7 +236,7 @@ impl<F, L> Booster<F> for GBM<'_, F, L>
 
         let predictions = h.predict_all(self.sample);
         let coef = self.loss.best_coefficient(
-            &self.sample.target(), &predictions[..]
+            self.sample.target(), &predictions[..]
         );
 
         // If the best coefficient is zero,

src/booster/graph_separation_boosting/graph_separation_algorithm.rs

@@ -45,7 +45,7 @@ use std::collections::HashSet;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -113,7 +113,7 @@ impl<'a, F> GraphSepBoost<'a, F> {
     }
 }
 
-impl<'a, F> GraphSepBoost<'a, F>
+impl<F> GraphSepBoost<'_, F>
     where F: Classifier
 {
     /// Returns a weight on the new hypothesis.
@@ -236,7 +236,7 @@ impl<F> Booster<F> for GraphSepBoost<'_, F>
         where W: WeakLearner<Hypothesis = F>
     {
         let hypotheses = std::mem::take(&mut self.hypotheses);
-        NaiveAggregation::new(hypotheses, &self.sample)
+        NaiveAggregation::new(hypotheses, self.sample)
     }
 }
 
@@ -246,6 +246,6 @@ impl<H> Research for GraphSepBoost<'_, H>
 {
     type Output = NaiveAggregation<H>;
     fn current_hypothesis(&self) -> Self::Output {
-        NaiveAggregation::from_slice(&self.hypotheses, &self.sample)
+        NaiveAggregation::from_slice(&self.hypotheses, self.sample)
     }
 }

src/booster/lpboost/lpboost_algorithm.rs

@@ -58,8 +58,8 @@ use std::ops::ControlFlow;
 /// # Related information
 /// - Currently (2023), `LPBoost` has no convergence guarantee.
 /// - [`ERLPBoost`](crate::booster::ERLPBoost), 
-/// A stabilized version of `LPBoost` is 
-/// proposed by Warmuth et al. (2008).
+///   A stabilized version of `LPBoost` is 
+///   proposed by Warmuth et al. (2008).
 /// 
 /// # Example
 /// The following code shows a small example for running [`LPBoost`].  
@@ -70,7 +70,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -262,7 +262,7 @@ impl<F> Booster<F> for LPBoost<'_, F>
             ("# of examples", format!("{n_sample}")),
             ("# of features", format!("{n_feature}")),
             ("Tolerance", format!("{}", self.tolerance)),
-            ("Max iteration", format!("-")),
+            ("Max iteration", "-".to_string()),
             ("Capping (outliers)", format!("{nu} ({ratio: >7.3} %)"))
         ]);
         Some(info)

src/booster/madaboost/madaboost_algorithm.rs

@@ -44,7 +44,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -274,7 +274,7 @@ impl<F> Booster<F> for MadaBoost<'_, F>
         let quit = if let Some(it) = self.force_quit_at {
             format!("At round {it}")
         } else {
-            format!("-")
+            "-".to_string()
         };
         let info = Vec::from([
             ("# of examples", format!("{}", n_sample)),

src/booster/mlpboost/mlpboost_algorithm.rs

@@ -92,7 +92,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")
@@ -388,7 +388,7 @@ impl<F> Booster<F> for MLPBoost<'_, F>
             ("Max iteration", format!("{}", self.max_iter)),
             ("Capping (outliers)", format!("{nu} ({ratio: >7.3} %)")),
             ("Primary", format!("{}", self.primary.current_type())),
-            ("Secondary", format!("LPBoost"))
+            ("Secondary", "LPBoost".to_string())
         ]);
         Some(info)
     }

src/booster/mlpboost/perturbed_lp_model.rs

@@ -3,7 +3,6 @@ use clarabel::{
     solver::*,
 };
 
-use rand;
 use rand::rngs::StdRng;
 use rand::prelude::*;
 

src/booster/smoothboost/smoothboost_algorithm.rs

@@ -49,7 +49,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")

src/booster/softboost/qp_model.rs

@@ -154,21 +154,21 @@ impl QPModel {
             }
             old_objval = objval;
         }
-        return Some(())
+        Some(())
     }
 
 
     /// Returns `true` if `dist[i] > 0` holds for all `i = 1, 2, ..., m.` 
     pub(self) fn all_positive(&self, dist: &[f64]) -> bool {
-        dist.into_iter()
+        dist.iter()
             .copied()
             .all(|d| d > 0f64)
     }
 
 
     pub(self) fn build_linear_part_objective(&self, dist: &[f64]) -> Vec<f64> {
         let mut linear = Vec::with_capacity(self.n_examples);
-        let iter = dist.into_iter()
+        let iter = dist.iter()
             .copied()
             .map(|di| di.ln());
         linear.extend(iter);

src/booster/softboost/softboost_algorithm.rs

@@ -64,7 +64,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")

src/booster/totalboost/totalboost_algorithm.rs

@@ -66,7 +66,7 @@ use std::ops::ControlFlow;
 /// 
 /// // Read the training sample from the CSV file.
 /// // We use the column named `class` as the label.
-/// let sample = SampleReader::new()
+/// let sample = SampleReader::default()
 ///     .file(path_to_file)
 ///     .has_header(true)
 ///     .target_feature("class")

src/common/loss_functions.rs

@@ -130,7 +130,7 @@ impl LossFunction for GBMLoss {
     fn hessian(&self, predictions: &[f64], target: &[f64]) -> Vec<f64>
     {
         let n_sample = predictions.len();
-        assert_eq!(n_sample as usize, target.len());
+        assert_eq!(n_sample, target.len());
 
         match self {
             Self::L1 => {

src/common/utils.rs

@@ -321,7 +321,7 @@ pub(crate) fn total_weight_for_label(
     weight: &[f64],
 ) -> f64
 {
-    target.into_iter()
+    target.iter()
         .copied()
         .zip(weight)
         .filter_map(|(t, w)| if t == y { Some(w) } else { None })

src/research/cross_validation.rs

@@ -152,7 +152,7 @@ impl<'a> CrossValidation<'a> {
 }
 
 
-impl<'a> Iterator for CrossValidation<'a> {
+impl Iterator for CrossValidation<'_> {
     type Item = (Sample, Sample);
     fn next(&mut self) -> Option<Self::Item> {
         if self.current_fold >= self.n_folds { return None; }

src/research/logger.rs

@@ -62,7 +62,7 @@ impl<'a, B, W, F, G> Logger<'a, B, W, F, G> {
     }
 }
 
-impl<'a, H, B, W, F, G, O> Logger<'a, B, W, F, G>
+impl<H, B, W, F, G, O> Logger<'_, B, W, F, G>
     where B: Booster<H, Output=O> + Research<Output=O>,
           O: Classifier,
           W: WeakLearner<Hypothesis = H>,
@@ -106,17 +106,15 @@ impl<'a, H, B, W, F, G, O> Logger<'a, B, W, F, G>
     #[inline(always)]
     fn print_log_header(&self) {
         println!(
-            "{} {:>WIDTH$}\t\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}",
-            "     ",
+            "      {:>WIDTH$}\t\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}",
             "".bold().red(),
             "OBJ.".bold().blue(),
             "TRAIN".bold().green(),
             "TEST".bold().yellow(),
             "ACC.".bold().cyan(),
         );
         println!(
-            "{} {:>WIDTH$}\t\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\n",
-            "     ",
+            "      {:>WIDTH$}\t\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\t{:>WIDTH$}\n",
             "ROUND".bold().red(),
             "VALUE".bold().blue(),
             "ERROR".bold().green(),
@@ -184,15 +182,14 @@ impl<'a, H, B, W, F, G, O> Logger<'a, B, W, F, G>
             "\
             + {:<STAT_WIDTH$}\t{:>STAT_WIDTH$}\n\
             + {:<STAT_WIDTH$}\t{:>STAT_WIDTH$}\n\
-            {:=^FULL_WIDTH$}\
+            {:=^FULL_WIDTH$}\n\
             ",
             "Objective".bold(),
             self.objective_func.name().bold().green(),
             "Time Limit".bold(),
             limit.bold().green(),
             "".bold(),
         );
-        println!("");
     }
 
 

src/sample/feature_struct.rs

@@ -422,7 +422,7 @@ impl SparseFeature {
     fn distinct_value_count(&self) -> usize {
 
         let values = self.sample[..]
-            .into_iter()
+            .iter()
             .map(|(_, v)| *v)
             .collect::<Vec<_>>();
         let mut uniq_value_count = inner_distinct_value_count(values);
@@ -526,7 +526,7 @@ impl Index<usize> for Feature {
 /// Count the number of items in `src` that has the same value.
 /// The given vector `src` is assumed to be sorted in ascending order.
 fn inner_distinct_value_count(mut src: Vec<f64>) -> usize {
-    src.sort_by(|a, b| a.partial_cmp(&b).unwrap());
+    src.sort_by(|a, b| a.partial_cmp(b).unwrap());
     let mut iter = src.into_iter();
     let mut value = match iter.next() {
         Some(v) => v,