1berry/MariaDB-server
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

mhnsw: make the search less greedy

Browse Source 
introduced a generosity factor that makes the search less greedy.
it dramatically improves the recall by making the search a bit slower
(for the same recall one can use half the M and smaller ef).

had to add Queue::safe_push() method that removes one of the
furthest elements (not necessarily the furthest) in the queue
to keep it from overflowing.
This commit is contained in:
Sergei Golubchik 2024-07-24 22:31:36 +02:00
parent 885eb19823
commit 26e599cd32
2 changed files with 15 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
sql/sql_queue.h
View File

@ -44,6 +44,11 @@ public:
Element *top() const { return (Element*)queue_top(&m_queue); } Element *top() const { return (Element*)queue_top(&m_queue); }
void push(const Element *element) { queue_insert(&m_queue, (uchar*)element); } void push(const Element *element) { queue_insert(&m_queue, (uchar*)element); }
void safe_push(const Element *element)
{
if (is_full()) m_queue.elements--; // remove one of the furthest elements
queue_insert(&m_queue, (uchar*)element);
}
Element *pop() { return (Element *)queue_remove_top(&m_queue); } Element *pop() { return (Element *)queue_remove_top(&m_queue); }
void clear() { queue_remove_all(&m_queue); } void clear() { queue_remove_all(&m_queue); }
void propagate_top() { queue_replace_top(&m_queue); } void propagate_top() { queue_replace_top(&m_queue); }

13
sql/vector_mhnsw.cc
View File

@ -27,6 +27,7 @@ ulonglong mhnsw_cache_size;
// Algorithm parameters // Algorithm parameters
static constexpr float alpha = 1.1f; static constexpr float alpha = 1.1f;
static constexpr float generosity = 1.1f;
static constexpr uint ef_construction= 10; static constexpr uint ef_construction= 10;
enum Graph_table_fields { enum Graph_table_fields {
@ -928,7 +929,8 @@ static int search_layer(MHNSW_Context *ctx, TABLE *graph, const FVector *target,
best.push(v); best.push(v);
} }
float furthest_best= FLT_MAX; float furthest_best= best.is_empty() ? FLT_MAX
: best.top()->distance_to_target * generosity;
while (candidates.elements()) while (candidates.elements())
{ {
const Visited &cur= *candidates.pop(); const Visited &cur= *candidates.pop();
@ -958,15 +960,18 @@ static int search_layer(MHNSW_Context *ctx, TABLE *graph, const FVector *target,
if (skip_deleted && v->node->deleted) if (skip_deleted && v->node->deleted)
continue; continue;
best.push(v); best.push(v);
furthest_best= best.top()->distance_to_target; furthest_best= best.top()->distance_to_target * generosity;
} }
else if (v->distance_to_target < furthest_best) else if (v->distance_to_target < furthest_best)
{ {
candidates.push(v); candidates.safe_push(v);
if (skip_deleted && v->node->deleted) if (skip_deleted && v->node->deleted)
continue; continue;
if (v->distance_to_target < best.top()->distance_to_target)
{
best.replace_top(v); best.replace_top(v);
furthest_best= best.top()->distance_to_target; furthest_best= best.top()->distance_to_target * generosity;
}
} }
} }
} }