Skip to content

web search done right

hyperloglog off by one

Mikkel Denkerb01ef0ada801
diff --git a/core/src/hyperloglog.rs b/core/src/hyperloglog.rs
index 668496b2..7f1f0b61 100644
--- a/core/src/hyperloglog.rs
+++ b/core/src/hyperloglog.rs
@@ -4087,7 +4087,7 @@ impl<const N: usize> HyperLogLog<N> {
pub fn estimate_bias(&self, e: f64, b: usize) -> f64 {
// binary search first nearest neighbor
- let lookup_array = RAW_ESTIMATE_DATA_VEC[b - RAW_ESTIMATE_DATA_OFFSET];
+ let lookup_array = RAW_ESTIMATE_DATA_VEC[b - 1 - RAW_ESTIMATE_DATA_OFFSET];
let mut idx_left = match lookup_array.binary_search_by(|v| v.partial_cmp(&e).unwrap()) {
Ok(i) => Some(i), // exact match
Err(i) => Some(i), // no match, i points to left neighbor
@@ -4105,10 +4105,10 @@ impl<const N: usize> HyperLogLog<N> {
};
// collect k nearest neighbors
- let k = 6;
- assert!(lookup_array.len() >= k);
- let mut neighbors = vec![];
- for _ in 0..k {
+ const K: usize = 6;
+ debug_assert!(lookup_array.len() >= K);
+ let mut neighbors = Vec::with_capacity(K);
+ for _ in 0..K {
let (right_instead_left, idx) = match (idx_left, idx_right) {
(Some(i_left), Some(i_right)) => {
// 2 candidates, find better one
@@ -4144,7 +4144,7 @@ impl<const N: usize> HyperLogLog<N> {
// calculate mean of neighbors
let bias_data = BIAS_DATA_VEC[b - BIAS_DATA_OFFSET];
- neighbors.iter().map(|&i| bias_data[i]).sum::<f64>() / (k as f64)
+ neighbors.iter().map(|&i| bias_data[i]).sum::<f64>() / (K as f64)
}
fn linear_counting(&self, v: usize) -> f64 {