kannolo 0.1.6

kANNolo is designed for easy prototyping of ANN Search algorithms while ensuring high effectiveness and efficiency over both dense and sparse vectors.
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

use std::io::Write;
use std::{fmt::Debug, time::Instant};

use clap::Parser;
use half::f16;
use std::fs::File;

use kannolo::{
    hnsw::graph_index::GraphIndex, hnsw_utils::config_hnsw::ConfigHnsw,
    plain_quantizer::PlainQuantizer, read_numpy_f32_flatten_2d, Dataset, DenseDataset,
    DistanceType, IndexSerializer,
};

#[derive(Parser, Debug)]
#[clap(author, version, about, long_about = None)]
struct Args {
    /// The path of the index.
    #[clap(short, long, value_parser)]
    index_file: String,

    /// The query seismic file.
    #[clap(short, long, value_parser)]
    query_file: String,

    /// The output file to write the results.
    #[clap(short, long, value_parser)]
    output_path: Option<String>,

    /// The number of top-k results to retrieve.
    #[clap(short, long, value_parser)]
    #[arg(default_value_t = 10)]
    k: usize,

    /// The list of ef_search values to test separated by commas.
    #[clap(long, value_parser)]
    #[arg(default_value_t = 40)]
    ef_search: usize,

    #[clap(long, value_parser)]
    #[arg(default_value_t = 1)]
    n_run: usize,

    #[clap(long, value_parser)]
    #[arg(default_value_t = false)]
    warmup: bool,
}

fn main() {
    // Parse command line arguments
    let args: Args = Args::parse();

    let query_path = args.query_file;

    let index_path = args.index_file;

    let k = args.k;
    let ef_search = args.ef_search;

    println!("Reading Queries");
    let (queries_vec, d) = read_numpy_f32_flatten_2d(query_path);
    let queries_vec = queries_vec
        .iter()
        .map(|&x| f16::from_f32(x))
        .collect::<Vec<f16>>();
    let queries = DenseDataset::from_vec(
        queries_vec,
        d,
        PlainQuantizer::<half::f16>::new(d, DistanceType::Euclidean),
    );

    let index: GraphIndex<DenseDataset<PlainQuantizer<half::f16>>, PlainQuantizer<half::f16>> =
        IndexSerializer::load_index(&index_path);

    println!("Starting search");
    let num_queries = queries.len();

    let mut config = ConfigHnsw::new().build();
    config.set_ef_search(ef_search);

    println!("N queries {num_queries}");

    if args.warmup {
        // println!("Running warmup!!");

        // let warmup_queries = SparseDataset::<SparsePlainQuantizer<f32>>::read_bin_file_f16(
        //     "/data2/knn_datasets/sparse_datasets/msmarco_v1_passage/cocondenser/data/queries.bin",
        //     None,
        // )
        // .unwrap();
        // for _ in 0..5 {
        //     let useless_results = index.search(&warmup_queries, k, &config, num_threads_search);
        //     println!("{:?}", useless_results[100]);
        // }
    }

    // Search
    let mut total_time_search = 0;
    let mut results = Vec::<(f32, usize)>::with_capacity(num_queries);

    for query in queries.iter() {
        let start_time = Instant::now();
        results.extend(
            index.search::<DenseDataset<PlainQuantizer<f16>>, PlainQuantizer<f16>>(
                query, k, &config,
            ),
        );
        let duration_search = start_time.elapsed();
        total_time_search += duration_search.as_micros();
    }

    let avg_time_search_per_query = total_time_search / (num_queries * args.n_run) as u128;

    println!("[######] Average Query Time: {avg_time_search_per_query}");

    index.print_space_usage_byte();

    let output_path = args.output_path.unwrap();
    let mut output_file = File::create(output_path).unwrap();

    for (query_id, result) in results.chunks_exact(k).enumerate() {
        // Writes results to a file in a parsable format
        for (idx, (score, doc_id)) in result.iter().enumerate() {
            writeln!(
                &mut output_file,
                "{query_id}\t{doc_id}\t{}\t{score}",
                idx + 1,
            )
            .unwrap();
        }
    }
}