v1.10.0 add expected accuracy parameter for search

yahoojapan · Mar 23, 2020 · 5b69650 · 5b69650
1 parent 8a7873d
commit 5b69650
Show file tree

Hide file tree

Showing 15 changed files with 726 additions and 143 deletions.
diff --git a/README-jp.md b/README-jp.md
@@ -41,16 +41,22 @@ Neighborhood Graph and Tree for Indexing High-dimensional Data
 
 - [Releases](https://github.com/yahoojapan/NGT/releases)
 
+### ビルド済み
+
+#### macOS
+
+      $ brew install ngt
+
 ### ビルド
 
 #### Linux
 
       $ unzip NGT-x.x.x.zip
       $ cd NGT-x.x.x
       $ mkdir build
-      $ cd build 
+      $ cd build
       $ cmake ..
-      $ make 
+      $ make
       $ make install
       $ ldconfig /usr/local/lib
 
@@ -59,14 +65,14 @@ Neighborhood Graph and Tree for Indexing High-dimensional Data
       $ /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
       $ brew install cmake
       $ brew install gcc@9
-      $ export CXX=/usr/local/bin/g++
-      $ export CC=/usr/local/bin/gcc
+      $ export CXX=/usr/local/bin/g++-9
+      $ export CC=/usr/local/bin/gcc-9
       $ unzip NGT-x.x.x.zip
       $ cd NGT-x.x.x
       $ mkdir build
-      $ cd build 
+      $ cd build
       $ cmake ..
-      $ make 
+      $ make
       $ make install
 
 #### 共有メモリの利用

diff --git a/README.md b/README.md
@@ -163,5 +163,4 @@ Publications
 ##### [ANNG](bin/ngt/README.md#anng)
 - Iwasaki, M.: Proximity search in metric spaces using approximate k nearest neighbor graph (in Japanese). IPSJ Trans. on Database 3(1) (2010) 18-28. ([pdf](https://s.yimg.jp/i/docs/research_lab/articles/miwasaki-ipsj-tod-2010.pdf))
 
-Copyright &copy; 2015-2019 Yahoo Japan Corporation All Rights Reserved.
 
diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-1.9.1
+1.10.0
diff --git a/lib/NGT/Capi.cpp b/lib/NGT/Capi.cpp
@@ -324,15 +324,18 @@ NGTObjectDistances ngt_create_empty_results(NGTError error) {
   }
 }
 
-static bool ngt_search_index_(NGT::Index* pindex, NGT::Object *ngtquery, size_t size, float epsilon, float radius, NGTObjectDistances results) {
+static bool ngt_search_index_(NGT::Index* pindex, NGT::Object *ngtquery, size_t size, float epsilon, float radius, NGTObjectDistances results, int edge_size = INT_MIN) {
   // set search prameters.
   NGT::SearchContainer sc(*ngtquery);      // search parametera container.
 
   sc.setResults(static_cast<NGT::ObjectDistances*>(results));          // set the result set.
-  sc.setSize(size);             // the number of resultant objects.
-  sc.setRadius(radius);         // search radius.
-  sc.setEpsilon(epsilon);           // set exploration coefficient.
-
+  sc.setSize(size);                        // the number of resultant objects.
+  sc.setRadius(radius);                    // search radius.
+  sc.setEpsilon(epsilon);                  // set exploration coefficient.
+  if (edge_size != INT_MIN) {
+    sc.setEdgeSize(edge_size);// set # of edges for each node
+  }
+
   pindex->search(sc);
 
   // delete the query object.
@@ -402,6 +405,39 @@ bool ngt_search_index_as_float(NGTIndex index, float *query, int32_t query_dim,
   return true;
 }
 
+bool ngt_search_index_with_query(NGTIndex index,  NGTQuery query, NGTObjectDistances results, NGTError error) {
+  if(index == NULL || query.query == NULL || results == NULL){
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : parametor error: index = " << index << " query = " << query.query << " results = " << results;
+    operate_error_string_(ss, error);
+    return false;
+  }
+
+  NGT::Index* pindex = static_cast<NGT::Index*>(index);   
+  int32_t dim = pindex->getObjectSpace().getDimension();
+
+  NGT::Object *ngtquery = NULL;
+
+  if(query.radius < 0.0){
+    query.radius = FLT_MAX;
+  }
+
+  try{
+    std::vector<float> vquery(&query.query[0], &query.query[dim]);
+    ngtquery = pindex->allocateObject(vquery);
+    ngt_search_index_(pindex, ngtquery, query.size, query.epsilon, query.radius, results, query.edge_size);
+  }catch(std::exception &err) {
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : Error: " << err.what();
+    operate_error_string_(ss, error);
+    if(ngtquery != NULL){
+      pindex->deleteObject(ngtquery);
+    }
+    return false;
+  }
+  return true;
+}
+
 
 // * deprecated *
 int32_t ngt_get_size(NGTObjectDistances results, NGTError error) {
@@ -768,10 +804,11 @@ bool ngt_optimizer_execute(NGTOptimizer optimizer, const char *inIndex, const ch
   return true;
 }
 
+// obsolute because of a lack of a parameter
 bool ngt_optimizer_set(NGTOptimizer optimizer, int outgoing, int incoming, int nofqs, 
 		       float baseAccuracyFrom, float baseAccuracyTo,
 		       float rateAccuracyFrom, float rateAccuracyTo,
-		       double qte, double m, NGTError error) {
+		       double gte, double m, NGTError error) {
   if(optimizer == NULL){
     std::stringstream ss;
     ss << "Capi : " << __FUNCTION__ << "() : parametor error: optimizer = " << optimizer;
@@ -780,7 +817,48 @@ bool ngt_optimizer_set(NGTOptimizer optimizer, int outgoing, int incoming, int n
   }
   try{
     (static_cast<NGT::GraphOptimizer*>(optimizer))->set(outgoing, incoming, nofqs, baseAccuracyFrom, baseAccuracyTo,
-							rateAccuracyFrom, rateAccuracyTo, qte, m);
+    							rateAccuracyFrom, rateAccuracyTo, gte, m);
+  }catch(std::exception &err) {
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : Error: " << err.what();
+    operate_error_string_(ss, error);                  
+    return false;
+  }
+  return true;
+}
+
+bool ngt_optimizer_set_minimum(NGTOptimizer optimizer, int outgoing, int incoming,
+			       int nofqs, int nofrs, NGTError error) {
+  if(optimizer == NULL){
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : parametor error: optimizer = " << optimizer;
+    operate_error_string_(ss, error);      
+    return false;
+  }
+  try{
+    (static_cast<NGT::GraphOptimizer*>(optimizer))->set(outgoing, incoming, nofqs, nofrs);
+  }catch(std::exception &err) {
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : Error: " << err.what();
+    operate_error_string_(ss, error);                  
+    return false;
+  }
+  return true;
+}
+
+bool ngt_optimizer_set_extension(NGTOptimizer optimizer,
+				 float baseAccuracyFrom, float baseAccuracyTo,
+				 float rateAccuracyFrom, float rateAccuracyTo,
+				 double gte, double m, NGTError error) {
+  if(optimizer == NULL){
+    std::stringstream ss;
+    ss << "Capi : " << __FUNCTION__ << "() : parametor error: optimizer = " << optimizer;
+    operate_error_string_(ss, error);      
+    return false;
+  }
+  try{
+    (static_cast<NGT::GraphOptimizer*>(optimizer))->setExtension(baseAccuracyFrom, baseAccuracyTo,
+								  rateAccuracyFrom, rateAccuracyTo, gte, m);
   }catch(std::exception &err) {
     std::stringstream ss;
     ss << "Capi : " << __FUNCTION__ << "() : Error: " << err.what();
@@ -796,3 +874,16 @@ void ngt_destroy_optimizer(NGTOptimizer optimizer)
     delete(static_cast<NGT::GraphOptimizer*>(optimizer));
 }
 
+bool ngt_refine_anng(NGTIndex index, float epsilon, float accuracy, int edgeSize, size_t batchSize, NGTError error)
+{
+    NGT::Index* pindex = static_cast<NGT::Index*>(index);    
+    try {
+      NGT::GraphReconstructor::refineANNG(*pindex, epsilon, accuracy, edgeSize, batchSize);
+    } catch(std::exception &err) {
+      std::stringstream ss;
+      ss << "Capi : " << __FUNCTION__ << "() : Error: " << err.what();
+      operate_error_string_(ss, error);      
+      return false;
+    }
+    return true;
+}
diff --git a/lib/NGT/Capi.h b/lib/NGT/Capi.h
@@ -37,6 +37,15 @@ typedef struct {
   float distance;
 } NGTObjectDistance;
 
+typedef struct {
+  float		*query;
+  size_t	size;		// # of returned objects
+  float		epsilon;
+  float		accuracy;	// expected accuracy
+  float		radius;
+  size_t	edge_size;	// # of edges to explore for each node
+} NGTQuery;
+
 NGTIndex ngt_open_index(const char *, NGTError);
 
 NGTIndex ngt_create_graph_and_tree(const char *, NGTProperty, NGTError);
@@ -88,7 +97,9 @@ NGTObjectDistances ngt_create_empty_results(NGTError);
 bool ngt_search_index(NGTIndex, double*, int32_t, size_t, float, float, NGTObjectDistances, NGTError);
 
 bool ngt_search_index_as_float(NGTIndex, float*, int32_t, size_t, float, float, NGTObjectDistances, NGTError);
-
+
+bool ngt_search_index_with_query(NGTIndex, NGTQuery, NGTObjectDistances, NGTError);
+
 int32_t ngt_get_size(NGTObjectDistances, NGTError); // deprecated
 
 uint32_t ngt_get_result_size(NGTObjectDistances, NGTError); 
@@ -146,10 +157,24 @@ bool ngt_optimizer_execute(NGTOptimizer, const char *, const char *, NGTError);
 bool ngt_optimizer_set(NGTOptimizer optimizer, int outgoing, int incoming, int nofqs, 
 		       float baseAccuracyFrom, float baseAccuracyTo,
 		       float rateAccuracyFrom, float rateAccuracyTo,
-		       double qte, double m, NGTError error);
+		       double gte, double m, NGTError error);
+
+bool ngt_optimizer_set_minimum(NGTOptimizer optimizer, int outgoing, int incoming, 
+			       int nofqs, int nofrs, NGTError error);
+
+bool ngt_optimizer_set_extension(NGTOptimizer optimizer,
+				 float baseAccuracyFrom, float baseAccuracyTo,
+				 float rateAccuracyFrom, float rateAccuracyTo,
+				 double gte, double m, NGTError error);
 
 void ngt_destroy_optimizer(NGTOptimizer);
 
+// refine the specified index by searching each node.
+// epsilon, exepectedAccuracy and edgeSize are the same as the prameters for search
+// batchSize is the degree of parallelizm.
+bool ngt_refine_anng(NGTIndex index, float epsilon, float expectedAccuracy, 
+		     int edgeSize, size_t batchSize, NGTError error);
+
 #ifdef __cplusplus
 }
 #endif
Original file line number	Diff line number	Diff line change
Expand Up		@@ -163,5 +163,4 @@ Publications
		##### [ANNG](bin/ngt/README.md#anng)
		- Iwasaki, M.: Proximity search in metric spaces using approximate k nearest neighbor graph (in Japanese). IPSJ Trans. on Database 3(1) (2010) 18-28. ([pdf](https://s.yimg.jp/i/docs/research_lab/articles/miwasaki-ipsj-tod-2010.pdf))

		Copyright © 2015-2019 Yahoo Japan Corporation All Rights Reserved.