aprender-core 0.29.1

Next-generation machine learning library in pure Rust
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


    #[test]
    fn test_curriculum_trainer_reset() {
        let scheduler = LinearCurriculum::new(5);
        let mut trainer = CurriculumTrainer::new(scheduler);

        let features = vec![1.0, 2.0, 3.0, 4.0];
        let targets = vec![1.0, 2.0];
        let scorer = FeatureNormScorer::new();

        trainer
            .add_samples(&features, &targets, 2, &scorer)
            .unwrap();

        trainer.advance();
        assert!(trainer.stage() > 0.0);

        trainer.reset();
        assert_eq!(trainer.stage(), 0.0);
    }

    #[test]
    fn test_curriculum_trainer_scheduler_access() {
        let scheduler = LinearCurriculum::new(5);
        let trainer = CurriculumTrainer::new(scheduler);

        let sched = trainer.scheduler();
        assert_eq!(sched.stage(), 0.0);
    }

    #[test]
    fn test_curriculum_trainer_is_complete() {
        let scheduler = LinearCurriculum::new(2);
        let mut trainer = CurriculumTrainer::new(scheduler);

        assert!(!trainer.is_complete());

        // Advance past completion
        for _ in 0..5 {
            trainer.advance();
        }
        assert!(trainer.is_complete());
    }

    #[test]
    fn test_scored_sample_debug_clone() {
        let sample = ScoredSample::new(vec![1.0, 2.0], 3.0, 0.5);
        let debug_str = format!("{:?}", sample);
        assert!(debug_str.contains("ScoredSample"));

        let cloned = sample.clone();
        assert_eq!(cloned.features, vec![1.0, 2.0]);
        assert_eq!(cloned.target, 3.0);
        assert_eq!(cloned.difficulty, 0.5);
    }

    #[test]
    fn test_linear_curriculum_debug_clone() {
        let curriculum = LinearCurriculum::new(5);
        let debug_str = format!("{:?}", curriculum);
        assert!(debug_str.contains("LinearCurriculum"));

        let cloned = curriculum.clone();
        assert_eq!(cloned.stage(), 0.0);
    }

    #[test]
    fn test_exponential_curriculum_debug_clone() {
        let curriculum = ExponentialCurriculum::new(0.3);
        let debug_str = format!("{:?}", curriculum);
        assert!(debug_str.contains("ExponentialCurriculum"));

        let cloned = curriculum.clone();
        assert_eq!(cloned.stage(), 0.0);
    }

    #[test]
    fn test_self_paced_curriculum_debug_clone() {
        let curriculum = SelfPacedCurriculum::new(0.5, 1.5);
        let debug_str = format!("{:?}", curriculum);
        assert!(debug_str.contains("SelfPacedCurriculum"));

        let cloned = curriculum.clone();
        assert_eq!(cloned.n_total(), 0);
    }

    #[test]
    fn test_loss_difficulty_scorer_debug_clone() {
        let scorer = LossDifficultyScorer::with_mean(3.0);
        let debug_str = format!("{:?}", scorer);
        assert!(debug_str.contains("LossDifficultyScorer"));

        let cloned = scorer.clone();
        // Cloned should produce same result
        let d1 = scorer.score(&[0.0], 5.0);
        let d2 = cloned.score(&[0.0], 5.0);
        assert!((d1 - d2).abs() < 1e-10);
    }

    #[test]
    fn test_self_paced_advance_resets_batch_idx() {
        let mut curriculum = SelfPacedCurriculum::new(1.0, 1.5);

        let samples = vec![
            ScoredSample::new(vec![1.0], 1.0, 0.1),
            ScoredSample::new(vec![2.0], 2.0, 0.2),
        ];
        curriculum.add_samples(samples);

        // Consume some batch
        let _ = curriculum.next_batch(1);

        // Advance resets batch_idx
        curriculum.advance();

        // Should start from beginning again
        let batch = curriculum.next_batch(10);
        assert_eq!(batch.len(), 2);
    }

    #[test]
    fn test_linear_curriculum_full_cycle() {
        let mut curriculum = LinearCurriculum::new(4).with_difficulty_range(0.0, 1.0);

        // Stage 0: threshold = 0.0
        assert!((curriculum.current_threshold() - 0.0).abs() < 0.01);

        // Advance through all stages
        curriculum.advance(); // 0.25
        assert!((curriculum.current_threshold() - 0.25).abs() < 0.01);

        curriculum.advance(); // 0.5
        assert!((curriculum.current_threshold() - 0.5).abs() < 0.01);

        curriculum.advance(); // 0.75
        assert!((curriculum.current_threshold() - 0.75).abs() < 0.01);

        curriculum.advance(); // 1.0
        assert!((curriculum.current_threshold() - 1.0).abs() < 0.01);
        assert!(curriculum.is_complete());
    }

    #[test]
    fn test_curriculum_trainer_eligible_empty_samples() {
        let scheduler = LinearCurriculum::new(5);
        let trainer = CurriculumTrainer::<LinearCurriculum>::new(scheduler);

        // No samples added - eligible should be empty
        let eligible = trainer.eligible_samples();
        assert!(eligible.is_empty());
        assert_eq!(trainer.n_eligible(), 0);
    }