tenflowers-dataset 0.1.1

Data pipeline and dataset utilities for TenfloweRS
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

//! Dataset visualization module
//!
//! This module provides utilities for visualizing and analyzing datasets
//! including sample previews, distribution analysis, histograms, and
//! augmentation effect analysis.
//!
//! The module is organized into several submodules:
//! - `types`: Core data structures for visualization information
//! - `visualizer`: Main DatasetVisualizer implementation with analysis methods
//! - `formatting`: Display and formatting implementations for all types
//! - `extensions`: Convenient trait extensions for datasets

pub mod extensions;
pub mod formatting;
pub mod types;
pub mod visualizer;

// Re-export all public types and traits for convenience
pub use extensions::DatasetVisualizationExt;
pub use types::*;
pub use visualizer::DatasetVisualizer;

#[cfg(test)]
mod tests {
    use super::*;
    use crate::TensorDataset;
    use tenflowers_core::Tensor;

    #[test]
    fn test_sample_preview() {
        let features = Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0, 5.0, 6.0], &[3, 2])
            .expect("test: tensor creation should succeed");
        let labels = Tensor::<f32>::from_vec(vec![0.0, 1.0, 2.0], &[3])
            .expect("test: tensor creation should succeed");

        let dataset = TensorDataset::new(features, labels);
        let preview = dataset
            .sample_preview(2)
            .expect("test: sample preview should succeed");

        assert_eq!(preview.total_samples, 3);
        assert!(preview.samples_shown <= 2);
        assert!(!preview.samples.is_empty());
    }

    #[test]
    fn test_feature_distribution() {
        let features = Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0, 5.0, 6.0], &[3, 2])
            .expect("test: tensor creation should succeed");
        let labels = Tensor::<f32>::from_vec(vec![0.0, 1.0, 2.0], &[3])
            .expect("test: tensor creation should succeed");

        let dataset = TensorDataset::new(features, labels);
        let distribution = dataset
            .feature_distribution(None)
            .expect("test: feature distribution should succeed");

        assert_eq!(distribution.samples_analyzed, 3);
        assert_eq!(distribution.feature_stats.len(), 2); // 2 feature dimensions
        assert_eq!(distribution.label_stats.len(), 1); // 1 label dimension
    }

    #[test]
    fn test_class_distribution() {
        let features = Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0], &[2, 2])
            .expect("test: tensor creation should succeed");
        let labels = Tensor::<f32>::from_vec(vec![0.0, 1.0], &[2])
            .expect("test: tensor creation should succeed");

        let dataset = TensorDataset::new(features, labels);
        let class_dist = dataset
            .class_distribution()
            .expect("test: class distribution should succeed");

        assert_eq!(class_dist.total_samples, 2);
        assert!(!class_dist.class_counts.is_empty());
    }

    #[test]
    fn test_feature_histogram() {
        let features =
            Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0], &[4, 2])
                .expect("test: tensor creation should succeed");
        let labels = Tensor::<f32>::from_vec(vec![0.0, 1.0, 2.0, 3.0], &[4])
            .expect("test: tensor creation should succeed");

        let dataset = TensorDataset::new(features, labels);
        let histogram = dataset
            .feature_histogram(0, 3)
            .expect("test: feature histogram should succeed");

        assert_eq!(histogram.feature_index, 0);
        assert_eq!(histogram.bin_counts.len(), 3);
        assert!(histogram.max_value >= histogram.min_value);
    }

    #[test]
    fn test_display_methods() {
        let features = Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0], &[2, 2])
            .expect("test: tensor creation should succeed");
        let labels = Tensor::<f32>::from_vec(vec![0.0, 1.0], &[2])
            .expect("test: tensor creation should succeed");

        let dataset = TensorDataset::new(features, labels);

        // Test display methods don't panic
        let preview = dataset
            .sample_preview(1)
            .expect("test: sample preview should succeed");
        let preview_text = preview.display();
        assert!(preview_text.contains("Dataset Sample Preview"));

        let distribution = dataset
            .feature_distribution(None)
            .expect("test: feature distribution should succeed");
        let dist_text = distribution.display();
        assert!(dist_text.contains("Dataset Distribution Analysis"));

        let class_dist = dataset
            .class_distribution()
            .expect("test: class distribution should succeed");
        let class_text = class_dist.display();
        assert!(class_text.contains("Class Distribution"));

        let histogram = dataset
            .feature_histogram(0, 2)
            .expect("test: feature histogram should succeed");
        let hist_text = histogram.display(10);
        assert!(hist_text.contains("Histogram"));
    }

    #[test]
    fn test_augmentation_effects_analysis() {
        use crate::transforms::Transform;

        // Create a simple test transform that adds 1.0 to all features
        struct AddOneTransform;

        impl Transform<f32> for AddOneTransform {
            fn apply(
                &self,
                sample: (tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>),
            ) -> tenflowers_core::Result<(tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>)>
            {
                let (features, labels) = sample;
                if let Some(data) = features.as_slice() {
                    let transformed_data: Vec<f32> = data.iter().map(|x| x + 1.0).collect();
                    let transformed_features = tenflowers_core::Tensor::from_vec(
                        transformed_data,
                        features.shape().dims(),
                    )?;
                    Ok((transformed_features, labels))
                } else {
                    Ok((features, labels))
                }
            }
        }

        let features =
            tenflowers_core::Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0, 5.0, 6.0], &[3, 2])
                .expect("test: operation should succeed");
        let labels = tenflowers_core::Tensor::<f32>::from_vec(vec![0.0, 1.0, 2.0], &[3])
            .expect("test: tensor creation should succeed");

        let dataset = crate::TensorDataset::new(features, labels);
        let transform = AddOneTransform;

        let analysis = dataset
            .analyze_augmentation_effects(&transform, 2)
            .expect("test: augmentation analysis should succeed");

        assert_eq!(analysis.samples_analyzed, 2);
        assert!(analysis.transform_success_rate > 0.0);
        assert_eq!(analysis.feature_changes.feature_count, 2);
        assert!(analysis.feature_changes.average_change > 0.0);

        // The transform adds 1.0, so mean should increase by 1.0
        let expected_mean_change = 1.0;
        let actual_mean_change = analysis.distribution_changes.mean_change;
        assert!((actual_mean_change - expected_mean_change).abs() < 0.001);
    }

    #[test]
    fn test_sample_comparison() {
        use crate::transforms::Transform;

        // Create a transform that multiplies by 2
        struct MultiplyByTwoTransform;

        impl Transform<f32> for MultiplyByTwoTransform {
            fn apply(
                &self,
                sample: (tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>),
            ) -> tenflowers_core::Result<(tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>)>
            {
                let (features, labels) = sample;
                if let Some(data) = features.as_slice() {
                    let transformed_data: Vec<f32> = data.iter().map(|x| x * 2.0).collect();
                    let transformed_features = tenflowers_core::Tensor::from_vec(
                        transformed_data,
                        features.shape().dims(),
                    )?;
                    Ok((transformed_features, labels))
                } else {
                    Ok((features, labels))
                }
            }
        }

        let sample1 = (
            tenflowers_core::Tensor::<f32>::from_vec(vec![1.0, 2.0], &[2])
                .expect("test: tensor creation should succeed"),
            tenflowers_core::Tensor::<f32>::from_vec(vec![0.0], &[1])
                .expect("test: tensor creation should succeed"),
        );
        let sample2 = (
            tenflowers_core::Tensor::<f32>::from_vec(vec![3.0, 4.0], &[2])
                .expect("test: tensor creation should succeed"),
            tenflowers_core::Tensor::<f32>::from_vec(vec![1.0], &[1])
                .expect("test: tensor creation should succeed"),
        );

        let samples = vec![sample1, sample2];
        let transform = MultiplyByTwoTransform;

        let comparisons = DatasetVisualizer::compare_samples(&samples, &transform, 2)
            .expect("test: compare samples should succeed");

        assert_eq!(comparisons.len(), 2);

        // Check that the change magnitude is non-zero
        for comparison in &comparisons {
            assert!(comparison.change_magnitude > 0.0);
            assert!(comparison.transformed_stats.mean > comparison.original_stats.mean);
        }
    }

    #[test]
    fn test_augmentation_effects_display() {
        use crate::transforms::Transform;

        // Identity transform for testing display
        struct IdentityTransform;

        impl Transform<f32> for IdentityTransform {
            fn apply(
                &self,
                sample: (tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>),
            ) -> tenflowers_core::Result<(tenflowers_core::Tensor<f32>, tenflowers_core::Tensor<f32>)>
            {
                Ok(sample)
            }
        }

        let features = tenflowers_core::Tensor::<f32>::from_vec(vec![1.0, 2.0, 3.0, 4.0], &[2, 2])
            .expect("test: tensor creation should succeed");
        let labels = tenflowers_core::Tensor::<f32>::from_vec(vec![0.0, 1.0], &[2])
            .expect("test: tensor creation should succeed");

        let dataset = crate::TensorDataset::new(features, labels);
        let transform = IdentityTransform;

        let analysis = dataset
            .analyze_augmentation_effects(&transform, 2)
            .expect("test: augmentation analysis should succeed");
        let display_text = analysis.display();

        assert!(display_text.contains("Augmentation Effects Analysis"));
        assert!(display_text.contains("Samples analyzed"));
        assert!(display_text.contains("Transform success rate"));
        assert!(display_text.contains("Feature Change Analysis"));
        assert!(display_text.contains("Distribution Change Analysis"));
    }
}