faiss 0.3.0 → 0.3.1

data/vendor/faiss/faiss/utils/utils.h

@@ -17,7 +17,9 @@
 #define FAISS_utils_h
 
 #include <stdint.h>
+#include <set>
 #include <string>
+#include <vector>
 
 #include <faiss/impl/platform_macros.h>
 #include <faiss/utils/Heap.h>
@@ -47,25 +49,6 @@ uint64_t get_cycles();
  * Misc  matrix and vector manipulation functions
  ***************************************************************************/
 
-/** compute c := a + bf * b for a, b and c tables
- *
- * @param n   size of the tables
- * @param a   size n
- * @param b   size n
- * @param c   restult table, size n
- */
-void fvec_madd(size_t n, const float* a, float bf, const float* b, float* c);
-
-/** same as fvec_madd, also return index of the min of the result table
- * @return    index of the min of table c
- */
-int fvec_madd_and_argmin(
-        size_t n,
-        const float* a,
-        float bf,
-        const float* b,
-        float* c);
-
 /* perform a reflection (not an efficient implementation, just for test ) */
 void reflection(const float* u, float* x, size_t n, size_t d, size_t nu);
 
@@ -121,7 +104,19 @@ int ivec_hist(size_t n, const int* v, int vmax, int* hist);
 void bincode_hist(size_t n, size_t nbits, const uint8_t* codes, int* hist);
 
 /// compute a checksum on a table.
-size_t ivec_checksum(size_t n, const int32_t* a);
+uint64_t ivec_checksum(size_t n, const int32_t* a);
+
+/// compute a checksum on a table.
+uint64_t bvec_checksum(size_t n, const uint8_t* a);
+
+/** compute checksums for the rows of a matrix
+ *
+ * @param n   number of rows
+ * @param d   size per row
+ * @param a   matrix to handle, size n * d
+ * @param cs  output checksums, size n
+ */
+void bvecs_checksum(size_t n, size_t d, const uint8_t* a, uint64_t* cs);
 
 /** random subsamples a set of vectors if there are too many of them
  *
@@ -163,6 +158,48 @@ uint64_t hash_bytes(const uint8_t* bytes, int64_t n);
 /** Whether OpenMP annotations were respected. */
 bool check_openmp();
 
+/** This class is used to combine range and knn search results
+ * in contrib.exhaustive_search.range_search_gpu */
+
+template <typename T>
+struct CombinerRangeKNN {
+    int64_t nq;    /// nb of queries
+    size_t k;      /// number of neighbors for the knn search part
+    T r2;          /// range search radius
+    bool keep_max; /// whether to keep max values instead of min.
+
+    CombinerRangeKNN(int64_t nq, size_t k, T r2, bool keep_max)
+            : nq(nq), k(k), r2(r2), keep_max(keep_max) {}
+
+    /// Knn search results
+    const int64_t* I = nullptr; /// size nq * k
+    const T* D = nullptr;       /// size nq * k
+
+    /// optional: range search results (ignored if mask is NULL)
+    const bool* mask =
+            nullptr; /// mask for where knn results are valid, size nq
+    // range search results for remaining entries nrange = sum(mask)
+    const int64_t* lim_remain = nullptr; /// size nrange + 1
+    const T* D_remain = nullptr;         /// size lim_remain[nrange]
+    const int64_t* I_remain = nullptr;   /// size lim_remain[nrange]
+
+    const int64_t* L_res = nullptr; /// size nq + 1
+    // Phase 1: compute sizes into limits array (of size nq + 1)
+    void compute_sizes(int64_t* L_res);
+
+    /// Phase 2: caller allocates D_res and I_res (size L_res[nq])
+    /// Phase 3: fill in D_res and I_res
+    void write_result(T* D_res, int64_t* I_res);
+};
+
+struct CodeSet {
+    size_t d;
+    std::set<std::vector<uint8_t>> s;
+
+    explicit CodeSet(size_t d) : d(d) {}
+    void insert(size_t n, const uint8_t* codes, bool* inserted);
+};
+
 } // namespace faiss
 
 #endif /* FAISS_utils_h */

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: faiss
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.3.1
 platform: ruby
 authors:
 - Andrew Kane
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-05-12 00:00:00.000000000 Z
+date: 2024-03-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rice
@@ -107,6 +107,8 @@ files:
 - vendor/faiss/faiss/IndexIVFFastScan.h
 - vendor/faiss/faiss/IndexIVFFlat.cpp
 - vendor/faiss/faiss/IndexIVFFlat.h
+- vendor/faiss/faiss/IndexIVFIndependentQuantizer.cpp
+- vendor/faiss/faiss/IndexIVFIndependentQuantizer.h
 - vendor/faiss/faiss/IndexIVFPQ.cpp
 - vendor/faiss/faiss/IndexIVFPQ.h
 - vendor/faiss/faiss/IndexIVFPQFastScan.cpp
@@ -202,6 +204,7 @@ files:
 - vendor/faiss/faiss/gpu/test/TestUtils.h
 - vendor/faiss/faiss/gpu/test/demo_ivfpq_indexing_gpu.cpp
 - vendor/faiss/faiss/gpu/utils/DeviceUtils.h
+- vendor/faiss/faiss/gpu/utils/RaftUtils.h
 - vendor/faiss/faiss/gpu/utils/StackDeviceMemory.cpp
 - vendor/faiss/faiss/gpu/utils/StackDeviceMemory.h
 - vendor/faiss/faiss/gpu/utils/StaticUtils.h
@@ -261,6 +264,8 @@ files:
 - vendor/faiss/faiss/impl/pq4_fast_scan.h
 - vendor/faiss/faiss/impl/pq4_fast_scan_search_1.cpp
 - vendor/faiss/faiss/impl/pq4_fast_scan_search_qbs.cpp
+- vendor/faiss/faiss/impl/residual_quantizer_encode_steps.cpp
+- vendor/faiss/faiss/impl/residual_quantizer_encode_steps.h
 - vendor/faiss/faiss/impl/simd_result_handlers.h
 - vendor/faiss/faiss/index_factory.cpp
 - vendor/faiss/faiss/index_factory.h
@@ -299,6 +304,7 @@ files:
 - vendor/faiss/faiss/utils/extra_distances-inl.h
 - vendor/faiss/faiss/utils/extra_distances.cpp
 - vendor/faiss/faiss/utils/extra_distances.h
+- vendor/faiss/faiss/utils/fp16-arm.h
 - vendor/faiss/faiss/utils/fp16-fp16c.h
 - vendor/faiss/faiss/utils/fp16-inl.h
 - vendor/faiss/faiss/utils/fp16.h
@@ -313,6 +319,7 @@ files:
 - vendor/faiss/faiss/utils/ordered_key_value.h
 - vendor/faiss/faiss/utils/partitioning.cpp
 - vendor/faiss/faiss/utils/partitioning.h
+- vendor/faiss/faiss/utils/prefetch.h
 - vendor/faiss/faiss/utils/quantize_lut.cpp
 - vendor/faiss/faiss/utils/quantize_lut.h
 - vendor/faiss/faiss/utils/random.cpp
@@ -345,7 +352,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.10
+rubygems_version: 3.5.3
 signing_key:
 specification_version: 4
 summary: Efficient similarity search and clustering for Ruby