rustyml 0.11.0

A high-performance machine learning & deep learning library in pure Rust, offering ML algorithms and neural network support
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
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293

#![cfg(feature = "machine_learning")]

use ndarray::{arr1, arr2};
use rustyml::machine_learning::decision_tree::{
    Algorithm, DecisionTree, DecisionTreeParams, Node, NodeType,
};

#[test]
fn test_decision_tree_new() {
    // Test creating a decision tree with default parameters
    let dt = DecisionTree::new(Algorithm::CART, true, None).unwrap();
    assert!(matches!(dt.get_algorithm(), Algorithm::CART));
    assert!(dt.get_is_classifier());

    // Test creating a decision tree with custom parameters
    let params = DecisionTreeParams {
        max_depth: Some(5),
        min_samples_split: 10,
        min_samples_leaf: 5,
        min_impurity_decrease: 0.1,
        random_state: Some(42),
    };
    let dt = DecisionTree::new(Algorithm::CART, false, Some(params)).unwrap();
    assert!(matches!(dt.get_algorithm(), Algorithm::CART));
    assert!(!dt.get_is_classifier());
    assert_eq!(dt.get_parameters().max_depth, Some(5));
    assert_eq!(dt.get_parameters().min_samples_split, 10);
    assert_eq!(dt.get_parameters().min_samples_leaf, 5);
    assert_eq!(dt.get_parameters().min_impurity_decrease, 0.1);
    assert_eq!(dt.get_parameters().random_state, Some(42));
}

#[test]
fn test_decision_tree_params_default() {
    let params = DecisionTreeParams::default();
    assert_eq!(params.max_depth, None);
    assert_eq!(params.min_samples_split, 2);
    assert_eq!(params.min_samples_leaf, 1);
    assert!(params.min_impurity_decrease < 1e-6);
    assert_eq!(params.random_state, None);
}

#[test]
fn test_fit_predict_classifier() {
    // Create a simple binary classification dataset
    let x = arr2(&[
        [2.0, 2.0],
        [2.0, 3.0],
        [3.0, 2.0],
        [3.0, 3.0],
        [1.0, 1.0],
        [1.0, 2.0],
    ]);
    let y = arr1(&[0.0, 0.0, 0.0, 0.0, 1.0, 1.0]);

    // Create classifier
    let mut dt = DecisionTree::new(Algorithm::CART, true, None).unwrap();

    // Train the model
    dt.fit(&x.view(), &y.view()).unwrap();

    // Ensure the model is trained correctly
    assert_eq!(dt.get_n_features(), 2);
    assert!(dt.get_root().is_some());

    // Predict
    let predictions = dt.predict(&x.view()).unwrap();
    assert_eq!(predictions.len(), 6);

    // Check prediction accuracy
    for i in 0..6 {
        assert_eq!(predictions[i], y[i]);
    }

    // Test single sample prediction
    let sample = &[2.0, 2.0];
    let pred = dt.predict_one(sample).unwrap();
    assert_eq!(pred, 0.0);

    let sample = &[1.0, 1.0];
    let pred = dt.predict_one(sample).unwrap();
    assert_eq!(pred, 1.0);

    // print tree
    println!("{}", dt.generate_tree_structure().unwrap());
}

#[test]
fn test_fit_predict_regressor() {
    // Create a simple regression dataset
    let x = arr2(&[[1.0], [2.0], [3.0], [4.0], [5.0], [6.0]]);
    let y = arr1(&[1.0, 2.0, 3.0, 4.0, 5.0, 6.0]);

    // Create regressor
    let mut dt = DecisionTree::new(Algorithm::CART, false, None).unwrap();

    // Train the model
    dt.fit(&x.view(), &y.view()).unwrap();

    // Ensure the model is trained correctly
    assert_eq!(dt.get_n_features(), 1);
    assert!(dt.get_root().is_some());

    // Predict
    let predictions = dt.predict(&x.view()).unwrap();
    assert_eq!(predictions.len(), 6);

    // Check prediction accuracy (regression might not be exact)
    for i in 0..6 {
        assert!((predictions[i] - y[i]).abs() < 0.5);
    }

    // print tree
    println!("{}", dt.generate_tree_structure().unwrap());
}

#[test]
fn test_predict_proba() {
    // Create a simple multi-class dataset
    let x = arr2(&[
        [1.0, 1.0],
        [1.0, 2.0],
        [2.0, 1.0],
        [2.0, 2.0],
        [3.0, 3.0],
        [4.0, 4.0],
    ]);
    let y = arr1(&[0.0, 0.0, 1.0, 1.0, 2.0, 2.0]);

    // Create classifier
    let mut dt = DecisionTree::new(Algorithm::CART, true, None).unwrap();

    // Train the model
    dt.fit(&x.view(), &y.view()).unwrap();

    // Test class probability prediction
    let prob = dt.predict_proba(&x.view()).unwrap();

    // Check the shape of the probability matrix
    assert_eq!(prob.shape()[0], 6); // 6 samples
    assert_eq!(prob.shape()[1], 3); // 3 classes

    // Check that the sum of probabilities for each sample is 1
    for i in 0..6 {
        let row_sum: f64 = prob.row(i).sum();
        assert!((row_sum - 1.0).abs() < 1e-5);
    }

    // Test single sample probability prediction
    let sample = &[1.0, 1.0];
    let probs = dt.predict_proba_one(sample).unwrap();
    assert_eq!(probs.len(), 3);
    assert!((probs.iter().sum::<f64>() - 1.0).abs() < 1e-5);
}

#[test]
fn test_different_algorithms() {
    let x = arr2(&[[1.0, 1.0], [1.0, 2.0], [2.0, 1.0], [2.0, 2.0]]);
    let y = arr1(&[0.0, 0.0, 1.0, 1.0]);

    // Test ID3 algorithm
    let mut dt_id3 = DecisionTree::new(Algorithm::ID3, true, None).unwrap();
    dt_id3.fit(&x.view(), &y.view()).unwrap();
    let pred_id3 = dt_id3.predict(&x.view()).unwrap();

    // Test C4.5 algorithm
    let mut dt_c45 = DecisionTree::new(Algorithm::C45, true, None).unwrap();
    dt_c45.fit(&x.view(), &y.view()).unwrap();
    let pred_c45 = dt_c45.predict(&x.view()).unwrap();

    // Test CART algorithm
    let mut dt_cart = DecisionTree::new(Algorithm::CART, true, None).unwrap();
    dt_cart.fit(&x.view(), &y.view()).unwrap();
    let pred_cart = dt_cart.predict(&x.view()).unwrap();

    // Verify that all algorithms correctly predict the training data
    for i in 0..4 {
        assert_eq!(pred_id3[i], y[i]);
        assert_eq!(pred_c45[i], y[i]);
        assert_eq!(pred_cart[i], y[i]);
    }
}

#[test]
fn test_max_depth_parameter() {
    let x = arr2(&[
        [1.0, 1.0],
        [1.0, 2.0],
        [2.0, 1.0],
        [2.0, 2.0],
        [3.0, 3.0],
        [3.0, 4.0],
        [4.0, 3.0],
        [4.0, 4.0],
    ]);
    let y = arr1(&[0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0]);

    let params_depth1 = DecisionTreeParams {
        max_depth: Some(1),
        ..DecisionTreeParams::default()
    };

    let mut dt_limited = DecisionTree::new(Algorithm::CART, true, Some(params_depth1)).unwrap();
    dt_limited.fit(&x.view(), &y.view()).unwrap();

    // Check if tree depth is limited
    // We can't directly check the depth, but we can infer that if the model
    // has only one level of depth, its prediction ability will be limited

    let _predictions = dt_limited.predict(&x.view()).unwrap();

    // Create an unlimited depth tree for comparison
    let mut dt_unlimited = DecisionTree::new(Algorithm::CART, true, None).unwrap();
    dt_unlimited.fit(&x.view(), &y.view()).unwrap();

    let predictions_unlimited = dt_unlimited.predict(&x.view()).unwrap();

    // The unlimited depth tree should perfectly fit the training data
    for i in 0..8 {
        assert_eq!(predictions_unlimited[i], y[i]);
    }
}

#[test]
fn test_error_handling() {
    // Create an untrained tree
    let dt = DecisionTree::new(Algorithm::CART, true, None).unwrap();

    // Trying to get the root node should fail
    assert!(dt.get_root().is_none());

    // Trying to get the number of classes should also fail
    assert!(dt.get_n_classes().is_none());

    // Trying to predict should fail
    let x = arr2(&[[1.0, 2.0]]);
    assert!(dt.predict(&x.view()).is_err());

    // Trying to get a single prediction should fail
    let sample = &[1.0, 2.0];
    assert!(dt.predict_one(sample).is_err());

    // Trying to get class probabilities should fail
    assert!(dt.predict_proba(&x.view()).is_err());
    assert!(dt.predict_proba_one(sample).is_err());
}

#[test]
fn test_node_creation() {
    // Test leaf node creation
    let leaf = Node::new_leaf(1.5, Some(0), Some(vec![0.8, 0.2]));
    match leaf.node_type {
        NodeType::Leaf {
            value,
            class,
            probabilities,
        } => {
            assert_eq!(value, 1.5);
            assert_eq!(class, Some(0));
            assert_eq!(probabilities, Some(vec![0.8, 0.2]));
        }
        _ => panic!("Expected a leaf node"),
    }

    // Test internal node creation
    let internal = Node::new_internal(1, 0.5);
    match internal.node_type {
        NodeType::Internal {
            feature_index,
            threshold,
            categories,
        } => {
            assert_eq!(feature_index, 1);
            assert_eq!(threshold, 0.5);
            assert_eq!(categories, None);
        }
        _ => panic!("Expected an internal node"),
    }

    // Test categorical node creation
    let categorical = Node::new_categorical(2, vec!["A".to_string(), "B".to_string()]);
    match categorical.node_type {
        NodeType::Internal {
            feature_index,
            threshold: _,
            categories,
        } => {
            assert_eq!(feature_index, 2);
            assert_eq!(categories, Some(vec!["A".to_string(), "B".to_string()]));
        }
        _ => panic!("Expected an internal categorical node"),
    }
}