hnswlib 0.6.2 → 0.8.0

data/ext/hnswlib/src/visited_list_pool.h

@@ -5,75 +5,74 @@
 #include <deque>
 
 namespace hnswlib {
-    typedef unsigned short int vl_type;
+typedef unsigned short int vl_type;
 
-    class VisitedList {
-    public:
-        vl_type curV;
-        vl_type *mass;
-        unsigned int numelements;
+class VisitedList {
+ public:
+    vl_type curV;
+    vl_type *mass;
+    unsigned int numelements;
 
-        VisitedList(int numelements1) {
-            curV = -1;
-            numelements = numelements1;
-            mass = new vl_type[numelements];
-        }
+    VisitedList(int numelements1) {
+        curV = -1;
+        numelements = numelements1;
+        mass = new vl_type[numelements];
+    }
 
-        void reset() {
+    void reset() {
+        curV++;
+        if (curV == 0) {
+            memset(mass, 0, sizeof(vl_type) * numelements);
             curV++;
-            if (curV == 0) {
-                memset(mass, 0, sizeof(vl_type) * numelements);
-                curV++;
-            }
-        };
+        }
+    }
 
-        ~VisitedList() { delete[] mass; }
-    };
+    ~VisitedList() { delete[] mass; }
+};
 ///////////////////////////////////////////////////////////
 //
 // Class for multi-threaded pool-management of VisitedLists
 //
 /////////////////////////////////////////////////////////
 
-    class VisitedListPool {
-        std::deque<VisitedList *> pool;
-        std::mutex poolguard;
-        int numelements;
-
-    public:
-        VisitedListPool(int initmaxpools, int numelements1) {
-            numelements = numelements1;
-            for (int i = 0; i < initmaxpools; i++)
-                pool.push_front(new VisitedList(numelements));
-        }
+class VisitedListPool {
+    std::deque<VisitedList *> pool;
+    std::mutex poolguard;
+    int numelements;
 
-        VisitedList *getFreeVisitedList() {
-            VisitedList *rez;
-            {
-                std::unique_lock <std::mutex> lock(poolguard);
-                if (pool.size() > 0) {
-                    rez = pool.front();
-                    pool.pop_front();
-                } else {
-                    rez = new VisitedList(numelements);
-                }
-            }
-            rez->reset();
-            return rez;
-        };
+ public:
+    VisitedListPool(int initmaxpools, int numelements1) {
+        numelements = numelements1;
+        for (int i = 0; i < initmaxpools; i++)
+            pool.push_front(new VisitedList(numelements));
+    }
 
-        void releaseVisitedList(VisitedList *vl) {
+    VisitedList *getFreeVisitedList() {
+        VisitedList *rez;
+        {
             std::unique_lock <std::mutex> lock(poolguard);
-            pool.push_front(vl);
-        };
-
-        ~VisitedListPool() {
-            while (pool.size()) {
-                VisitedList *rez = pool.front();
+            if (pool.size() > 0) {
+                rez = pool.front();
                 pool.pop_front();
-                delete rez;
+            } else {
+                rez = new VisitedList(numelements);
             }
-        };
-    };
-}
+        }
+        rez->reset();
+        return rez;
+    }
 
+    void releaseVisitedList(VisitedList *vl) {
+        std::unique_lock <std::mutex> lock(poolguard);
+        pool.push_front(vl);
+    }
+
+    ~VisitedListPool() {
+        while (pool.size()) {
+            VisitedList *rez = pool.front();
+            pool.pop_front();
+            delete rez;
+        }
+    }
+};
+}  // namespace hnswlib

data/lib/hnswlib/version.rb

@@ -3,8 +3,8 @@
 # Hnswlib.rb provides Ruby bindings for the Hnswlib.
 module Hnswlib
   # The version of Hnswlib.rb you install.
-  VERSION = '0.6.2'
+  VERSION = '0.8.0'
 
   # The version of Hnswlib included with gem.
-  HSWLIB_VERSION = '0.6.2'
+  HSWLIB_VERSION = '0.7.0'
 end

data/lib/hnswlib.rb

@@ -18,6 +18,7 @@ module Hnswlib
   #
   #   index.get_nns_by_item(0, 100)
   #
+  # @deprecated This class was prepared as a class with an interface similar to Annoy, but it is not very useful and will be deleted in the next version.
   class HnswIndex
     # Returns the metric of index.
     # @return [String]
@@ -27,29 +28,28 @@ module Hnswlib
     #
     # @param n_features [Integer] The number of features (dimensions) of stored vector.
     # @param max_item [Integer] The maximum number of items.
-    # @param metric [String] The distance metric between vectors ('l2' or 'dot').
+    # @param metric [String] The distance metric between vectors ('l2', 'dot', or 'cosine').
     # @param m [Integer] The maximum number of outgoing connections in the graph
     # @param ef_construction [Integer] The size of the dynamic list for the nearest neighbors. It controls the index time/accuracy trade-off.
     # @param random_seed [Integer] The seed value using to initialize the random generator.
-    def initialize(n_features:, max_item:, metric: 'l2', m: 16, ef_construction: 200, random_seed: 100)
+    # @param allow_replace_removed [Boolean] The flag to replace removed element when adding new element.
+    def initialize(n_features:, max_item:, metric: 'l2', m: 16, ef_construction: 200,
+                   random_seed: 100, allow_replace_removed: false)
       @metric = metric
-      space = if @metric == 'dot'
-                Hnswlib::InnerProductSpace.new(n_features)
-              else
-                Hnswlib::L2Space.new(n_features)
-              end
-      @index = Hnswlib::HierarchicalNSW.new(
-        space: space, max_elements: max_item, m: m, ef_construction: ef_construction, random_seed: random_seed
-      )
+      space = @metric == 'dot' ? 'ip' : 'l2'
+      @index = Hnswlib::HierarchicalNSW.new(space: space, dim: n_features)
+      @index.init_index(max_elements: max_item, m: m, ef_construction: ef_construction,
+                        random_seed: random_seed, allow_replace_deleted: allow_replace_removed)
     end
 
     # Add item to be indexed.
     #
     # @param i [Integer] The ID of item.
     # @param v [Array] The vector of item.
+    # @param replace_removed [Boolean] The flag to replace a removed element.
     # @return [Boolean]
-    def add_item(i, v)
-      @index.add_point(v, i)
+    def add_item(i, v, replace_removed: false)
+      @index.add_point(v, i, replace_deleted: replace_removed)
     end
 
     # Return the item vector.
@@ -72,10 +72,11 @@ module Hnswlib
     # @param i [Integer] The ID of query item.
     # @param n [Integer] The number of nearest neighbors.
     # @param include_distances [Boolean] The flag indicating whether to returns all corresponding distances.
+    # @param filter [Proc] The function that filters elements by its labels.
     # @return [Array<Integer> or Array<Array<Integer>, Array<Float>>]
-    def get_nns_by_item(i, n, include_distances: false)
+    def get_nns_by_item(i, n, include_distances: false, filter: nil)
       v = @index.get_point(i)
-      ids, dists = @index.search_knn(v, n)
+      ids, dists = @index.search_knn(v, n, filter: filter)
       include_distances ? [ids, dists] : ids
     end
 
@@ -84,9 +85,10 @@ module Hnswlib
     # @param v [Array] The vector of query item.
     # @param n [Integer] The number of nearest neighbors.
     # @param include_distances [Boolean] The flag indicating whether to returns all corresponding distances.
+    # @param filter [Proc] The function that filters elements by its labels.
     # @return [Array<Integer> or Array<Array<Integer>, Array<Float>>]
-    def get_nns_by_vector(v, n, include_distances: false)
-      ids, dists = @index.search_knn(v, n)
+    def get_nns_by_vector(v, n, include_distances: false, filter: nil)
+      ids, dists = @index.search_knn(v, n, filter: filter)
       include_distances ? [ids, dists] : ids
     end
 
@@ -114,8 +116,9 @@ module Hnswlib
     # Load a search index from disk.
     #
     # @param filename [String] The filename of search index.
-    def load(filename)
-      @index.load_index(filename)
+    # @param allow_replace_removed [Boolean] The flag to replace removed element when adding new element.
+    def load(filename, allow_replace_removed: false)
+      @index.load_index(filename, allow_replace_deleted: allow_replace_removed)
     end
 
     # Calculate the distances between items.

data/sig/hnswlib.rbs

@@ -5,8 +5,8 @@ module Hnswlib
   class HnswIndex
     attr_reader metric: String
 
-    def initialize: (n_features: Integer n_features, max_item: Integer max_item, ?metric: ::String metric, ?m: ::Integer m, ?ef_construction: ::Integer ef_construction, ?random_seed: ::Integer random_seed) -> void
-    def add_item: (Integer i, Array[Float] v) -> bool
+    def initialize: (n_features: Integer n_features, max_item: Integer max_item, ?metric: ::String metric, ?m: ::Integer m, ?ef_construction: ::Integer ef_construction, ?random_seed: ::Integer random_seed, ?allow_replace_removed: (true | false) allow_replace_removed) -> void
+    def add_item: (Integer i, Array[Float] v, ?replace_removed: (true | false) replace_removed) -> bool
     def get_item: (Integer i) -> Array[Float]
     def remove_item: (Integer i) -> void
     def get_nns_by_item: (Integer i, Integer n, ?include_distances: (true | false) include_distances) -> ([Array[Integer], Array[Float]] | Array[Integer])
@@ -14,7 +14,7 @@ module Hnswlib
     def resize_index: (Integer new_max_item) -> void
     def set_ef: (Integer ef) -> void
     def save: (String filename) -> void
-    def load: (String filename) -> void
+    def load: (String filename, ?allow_replace_removed: (true | false) allow_replace_removed) -> void
     def get_distance: (Integer i, Integer j) -> Float
     def n_items: () -> Integer
     def n_features: () -> Integer
@@ -40,7 +40,8 @@ module Hnswlib
   class BruteforceSearch
     attr_accessor space: (::Hnswlib::L2Space | ::Hnswlib::InnerProductSpace)
 
-    def initialize: (space: (::Hnswlib::L2Space | ::Hnswlib::InnerProductSpace) space, max_elements: Integer max_elements) -> void
+    def initialize: (space: String space, dim: Integer dim) -> void
+    def init_index: (max_elements: Integer max_elements) -> void
     def add_point: (Array[Float] arr, Integer idx) -> bool
     def current_count: () -> Integer
     def load_index: (String filename) -> void
@@ -53,12 +54,13 @@ module Hnswlib
   class HierarchicalNSW
     attr_accessor space: (::Hnswlib::L2Space | ::Hnswlib::InnerProductSpace)
 
-    def initialize: (space: (::Hnswlib::L2Space | ::Hnswlib::InnerProductSpace) space, max_elements: Integer max_elements, ?m: Integer m, ?ef_construction: Integer ef_construction, ?random_seed: Integer random_seed) -> void
-    def add_point: (Array[Float] arr, Integer idx) -> bool
+    def initialize: (space: String space, dim: Integer dim) -> void
+    def init_index: (max_elements: Integer max_elements, ?m: Integer m, ?ef_construction: Integer ef_construction, ?random_seed: Integer random_seed, ?allow_replace_deleted: (true | false) allow_replace_deleted) -> void
+    def add_point: (Array[Float] arr, Integer idx, ?replace_deleted: (true | false) replace_deleted) -> bool
     def current_count: () -> Integer
     def get_ids: () -> Array[Integer]
     def get_point: (Integer idx) -> Array[Float]
-    def load_index: (String filename) -> void
+    def load_index: (String filename, ?allow_replace_deleted: (true | false) allow_replace_deleted) -> void
     def mark_deleted: (Integer idx) -> void
     def max_elements: () -> Integer
     def resize_index: (Integer new_max_elements) -> void

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: hnswlib
 version: !ruby/object:Gem::Version
-  version: 0.6.2
+  version: 0.8.0
 platform: ruby
 authors:
 - yoshoku
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2022-06-25 00:00:00.000000000 Z
+date: 2023-03-14 00:00:00.000000000 Z
 dependencies: []
 description: Hnswlib.rb provides Ruby bindings for the Hnswlib.
 email:
@@ -57,7 +57,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.33
+rubygems_version: 3.3.26
 signing_key:
 specification_version: 4
 summary: Ruby bindings for the Hnswlib.