faiss 0.6.2 → 0.6.3

data/vendor/faiss/faiss/IndexRefine.cpp

@@ -156,7 +156,17 @@ void IndexRefine::range_search(
     SearchParameters* base_index_params =
             (params != nullptr) ? params->base_index_params : nullptr;
 
-    base_index->range_search(n, x, radius, result, base_index_params);
+    const float kf = (params != nullptr) ? params->k_factor : this->k_factor;
+
+    const bool is_similarity = is_similarity_metric(metric_type);
+
+    // Scale the base_index search radius by k_factor. Results are filtered to
+    // the exact radius below, so this only affects recall, not correctness; it
+    // is up to the user to pick a k_factor that widens the base search for
+    // their metric.
+    const float base_radius = radius * kf;
+
+    base_index->range_search(n, x, base_radius, result, base_index_params);
 
 #pragma omp parallel if (n > 1)
     {
@@ -177,6 +187,25 @@ void IndexRefine::range_search(
             }
         }
     }
+
+    // Exact filtering done here, as base index's approximations
+    // can fall outside the specified exact radius.
+    const std::vector<size_t> prev_lims(result->lims, result->lims + n + 1);
+    size_t wp = 0;
+    for (idx_t i = 0; i < n; i++) {
+        for (size_t j = prev_lims[i]; j < prev_lims[i + 1]; j++) {
+            const float dis = result->distances[j];
+            const bool within = is_similarity
+                    ? CMin<float, idx_t>::cmp(radius, dis)
+                    : CMax<float, idx_t>::cmp(radius, dis);
+            if (within) {
+                result->labels[wp] = result->labels[j];
+                result->distances[wp] = result->distances[j];
+                wp++;
+            }
+        }
+        result->lims[i + 1] = wp;
+    }
 }
 
 void IndexRefine::reconstruct(idx_t key, float* recons) const {

data/vendor/faiss/faiss/IndexReplicas.cpp

@@ -127,8 +127,7 @@ void IndexReplicasTemplate<IndexT>::search(
         distance_t* distances,
         idx_t* labels,
         const SearchParameters* params) const {
-    FAISS_THROW_IF_NOT_MSG(
-            !params, "search params not supported for this index");
+    FAISS_THROW_IF_MSG(params, "search params not supported for this index");
     FAISS_THROW_IF_NOT(k > 0);
     FAISS_THROW_IF_NOT_MSG(this->count() > 0, "no replicas in index");
 

data/vendor/faiss/faiss/IndexShards.cpp

@@ -143,8 +143,8 @@ void IndexShardsTemplate<IndexT>::add_with_ids(
             "request them to be shifted");
 
     if (successive_ids) {
-        FAISS_THROW_IF_NOT_MSG(
-                !xids,
+        FAISS_THROW_IF_MSG(
+                xids,
                 "It makes no sense to pass in ids and "
                 "request them to be shifted");
         FAISS_THROW_IF_NOT_MSG(

data/vendor/faiss/faiss/IndexShardsIVF.cpp

@@ -106,8 +106,8 @@ void IndexShardsIVF::add_with_ids(
             "request them to be shifted");
 
     if (successive_ids) {
-        FAISS_THROW_IF_NOT_MSG(
-                !xids,
+        FAISS_THROW_IF_MSG(
+                xids,
                 "It makes no sense to pass in ids and "
                 "request them to be shifted");
         FAISS_THROW_IF_NOT_MSG(

data/vendor/faiss/faiss/MetaIndexes.cpp

@@ -63,8 +63,7 @@ void IndexSplitVectors::search(
         float* distances,
         idx_t* labels,
         const SearchParameters* params) const {
-    FAISS_THROW_IF_NOT_MSG(
-            !params, "search params not supported for this index");
+    FAISS_THROW_IF_MSG(params, "search params not supported for this index");
     FAISS_THROW_IF_NOT_MSG(k == 1, "search implemented only for k=1");
     FAISS_THROW_IF_NOT_MSG(
             sum_d == d, "not enough indexes compared to # dimensions");
@@ -187,8 +186,7 @@ void IndexRandom::search(
         float* distances,
         idx_t* labels,
         const SearchParameters* params) const {
-    FAISS_THROW_IF_NOT_MSG(
-            !params, "search params not supported for this index");
+    FAISS_THROW_IF_MSG(params, "search params not supported for this index");
     FAISS_THROW_IF_NOT(k <= ntotal);
 #pragma omp parallel for if (n > 1000)
     for (idx_t i = 0; i < n; i++) {