oxits 0.1.0

Time series classification and transformation library for Rust
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204

use crate::core::config::BinStrategy;
use crate::preprocessing::quantile_transform::norm_ppf;

#[derive(Debug, Clone)]
pub struct KBinsDiscretizerConfig {
    pub n_bins: usize,
    pub strategy: BinStrategy,
}

impl KBinsDiscretizerConfig {
    pub fn new(n_bins: usize) -> Self {
        Self {
            n_bins,
            strategy: BinStrategy::Quantile,
        }
    }
}

/// Discretize time series into bin indices (0-based).
pub struct KBinsDiscretizer;

impl KBinsDiscretizer {
    pub fn transform(config: &KBinsDiscretizerConfig, x: &[Vec<f64>]) -> Vec<Vec<usize>> {
        assert!(!x.is_empty(), "Input must have at least one sample");
        assert!(
            config.n_bins >= 2,
            "n_bins must be at least 2, got {}",
            config.n_bins
        );

        match config.strategy {
            BinStrategy::Normal => {
                let edges = normal_bin_edges(config.n_bins);

                #[cfg(feature = "parallel")]
                {
                    use rayon::prelude::*;
                    return x
                        .par_iter()
                        .map(|sample| digitize_1d(sample, &edges))
                        .collect();
                }

                #[cfg(not(feature = "parallel"))]
                x.iter().map(|sample| digitize_1d(sample, &edges)).collect()
            }
            BinStrategy::Uniform => {
                let n_bins = config.n_bins;
                let discretize = |sample: &Vec<f64>| {
                    let edges = uniform_bin_edges(sample, n_bins);
                    digitize_1d(sample, &edges)
                };

                #[cfg(feature = "parallel")]
                {
                    use rayon::prelude::*;
                    return x.par_iter().map(discretize).collect();
                }

                #[cfg(not(feature = "parallel"))]
                x.iter().map(discretize).collect()
            }
            BinStrategy::Quantile => {
                let n_bins = config.n_bins;
                let discretize = |sample: &Vec<f64>| {
                    let edges = quantile_bin_edges(sample, n_bins);
                    digitize_1d(sample, &edges)
                };

                #[cfg(feature = "parallel")]
                {
                    use rayon::prelude::*;
                    return x.par_iter().map(discretize).collect();
                }

                #[cfg(not(feature = "parallel"))]
                x.iter().map(discretize).collect()
            }
        }
    }
}

/// Compute bin edges for normal strategy: standard normal quantiles.
fn normal_bin_edges(n_bins: usize) -> Vec<f64> {
    (1..n_bins)
        .map(|i| {
            let p = i as f64 / n_bins as f64;
            norm_ppf(p)
        })
        .collect()
}

/// Compute bin edges for uniform strategy (per-sample).
fn uniform_bin_edges(x: &[f64], n_bins: usize) -> Vec<f64> {
    let x_min = x.iter().copied().fold(f64::INFINITY, f64::min);
    let x_max = x.iter().copied().fold(f64::NEG_INFINITY, f64::max);
    (1..n_bins)
        .map(|i| x_min + (x_max - x_min) * i as f64 / n_bins as f64)
        .collect()
}

/// Compute bin edges for quantile strategy (per-sample).
/// Sorts once and computes all percentiles from the sorted array.
fn quantile_bin_edges(x: &[f64], n_bins: usize) -> Vec<f64> {
    let mut sorted: Vec<f64> = x.to_vec();
    sorted.sort_by(|a, b| a.partial_cmp(b).unwrap());
    let n = sorted.len();

    let mut edges: Vec<f64> = (1..n_bins)
        .map(|i| {
            let p = i as f64 / n_bins as f64;
            let idx = p * (n - 1) as f64;
            let lo = idx.floor() as usize;
            let hi = lo + 1;
            let frac = idx - lo as f64;
            if hi >= n {
                sorted[n - 1]
            } else {
                sorted[lo] + frac * (sorted[hi] - sorted[lo])
            }
        })
        .collect();

    // Deduplicate adjacent edges (within tolerance)
    edges.dedup_by(|a, b| (*a - *b).abs() < 1e-8);
    edges
}

/// Digitize: for each value, find the bin index via searchsorted (left).
/// Returns 0-based bin indices in [0, n_edges].
fn digitize_1d(x: &[f64], edges: &[f64]) -> Vec<usize> {
    x.iter()
        .map(|&v| edges.partition_point(|&e| e <= v))
        .collect()
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_uniform_basic() {
        let config = KBinsDiscretizerConfig {
            n_bins: 4,
            strategy: BinStrategy::Uniform,
        };
        let x = vec![vec![0.0, 2.5, 5.0, 7.5, 10.0]];
        let result = KBinsDiscretizer::transform(&config, &x);
        assert_eq!(result[0].len(), 5);
        assert_eq!(result[0][0], 0); // 0.0 -> bin 0
                                     // Edges at 2.5, 5.0, 7.5 — 10.0 is past all edges -> bin 3
        assert_eq!(result[0][4], 3);
    }

    #[test]
    fn test_normal_strategy() {
        let config = KBinsDiscretizerConfig {
            n_bins: 3,
            strategy: BinStrategy::Normal,
        };
        let x = vec![vec![-2.0, -0.5, 0.0, 0.5, 2.0]];
        let result = KBinsDiscretizer::transform(&config, &x);
        // Normal bins: edges at norm_ppf(1/3) ≈ -0.43 and norm_ppf(2/3) ≈ 0.43
        assert_eq!(result[0].len(), 5);
        assert_eq!(result[0][0], 0); // -2.0 -> bin 0
        assert_eq!(result[0][4], 2); // 2.0 -> bin 2
    }

    #[test]
    fn test_quantile_strategy() {
        let config = KBinsDiscretizerConfig {
            n_bins: 2,
            strategy: BinStrategy::Quantile,
        };
        let x = vec![vec![1.0, 2.0, 3.0, 4.0]];
        let result = KBinsDiscretizer::transform(&config, &x);
        assert_eq!(result[0].len(), 4);
        // 1 edge at 50th percentile = 2.5
        // Values < 2.5 -> bin 0, values >= 2.5 -> bin 1
    }

    #[test]
    fn test_multiple_samples() {
        let config = KBinsDiscretizerConfig::new(3);
        let x = vec![vec![1.0, 2.0, 3.0, 4.0], vec![10.0, 20.0, 30.0, 40.0]];
        let result = KBinsDiscretizer::transform(&config, &x);
        assert_eq!(result.len(), 2);
    }

    #[test]
    #[should_panic(expected = "n_bins must be at least 2")]
    fn test_invalid_n_bins() {
        let config = KBinsDiscretizerConfig::new(1);
        KBinsDiscretizer::transform(&config, &[vec![1.0, 2.0]]);
    }

    #[test]
    fn test_normal_bin_edges() {
        let edges = normal_bin_edges(3);
        assert_eq!(edges.len(), 2);
        // Should be symmetric around 0
        assert!((edges[0] + edges[1]).abs() < 1e-10);
    }
}