torsh_cluster/initialization/
random_partition.rs

1//! Random partition initialization strategy
2
3use super::InitializationStrategy;
4use crate::error::{ClusterError, ClusterResult};
5use scirs2_core::random::Random;
6use torsh_tensor::Tensor;
7
8/// Random partition initialization: randomly assign points to clusters, then compute centroids
9#[derive(Debug, Default)]
10pub struct RandomPartition;
11
12impl InitializationStrategy for RandomPartition {
13    fn initialize(
14        &self,
15        data: &Tensor,
16        n_clusters: usize,
17        seed: Option<u64>,
18    ) -> ClusterResult<Tensor> {
19        let n_samples = data.shape().dims()[0];
20        let n_features = data.shape().dims()[1];
21
22        if n_clusters > n_samples {
23            return Err(ClusterError::InvalidClusters(n_clusters));
24        }
25
26        let mut rng = Random::seed(seed.unwrap_or_else(|| {
27            use std::time::{SystemTime, UNIX_EPOCH};
28            SystemTime::now()
29                .duration_since(UNIX_EPOCH)
30                .expect("system time should be after UNIX_EPOCH")
31                .as_secs()
32        }));
33
34        let data_vec = data.to_vec().map_err(ClusterError::TensorError)?;
35
36        // Randomly assign points to clusters
37        let mut cluster_assignments = Vec::new();
38        for _ in 0..n_samples {
39            cluster_assignments.push(rng.gen_range(0..n_clusters));
40        }
41
42        // Compute centroid for each cluster
43        let mut centroids_data = vec![0.0; n_clusters * n_features];
44        let mut cluster_counts = vec![0; n_clusters];
45
46        for i in 0..n_samples {
47            let cluster = cluster_assignments[i];
48            cluster_counts[cluster] += 1;
49            for j in 0..n_features {
50                centroids_data[cluster * n_features + j] += data_vec[i * n_features + j];
51            }
52        }
53
54        // Average to get centroids
55        for k in 0..n_clusters {
56            if cluster_counts[k] > 0 {
57                for j in 0..n_features {
58                    centroids_data[k * n_features + j] /= cluster_counts[k] as f32;
59                }
60            }
61        }
62
63        Tensor::from_vec(centroids_data, &[n_clusters, n_features])
64            .map_err(ClusterError::TensorError)
65    }
66
67    fn name(&self) -> &str {
68        "Random Partition"
69    }
70}
torsh_cluster/initialization/random_partition.rs

torsh_cluster/initialization/
random_partition.rs
