faiss 0.1.7 → 0.2.3

data/vendor/faiss/faiss/invlists/InvertedLists.h

@@ -15,9 +15,8 @@
  * the interface.
  */
 
-#include <vector>
 #include <faiss/Index.h>
-
+#include <vector>
 
 namespace faiss {
 
@@ -31,10 +30,10 @@ namespace faiss {
 struct InvertedLists {
     typedef Index::idx_t idx_t;
 
-    size_t nlist;             ///< number of possible key values
-    size_t code_size;         ///< code size per vector in bytes
+    size_t nlist;     ///< number of possible key values
+    size_t code_size; ///< code size per vector in bytes
 
-    InvertedLists (size_t nlist, size_t code_size);
+    InvertedLists(size_t nlist, size_t code_size);
 
     /// used for BlockInvertedLists, where the codes are packed into groups
     /// and the individual code size is meaningless
@@ -51,70 +50,77 @@ struct InvertedLists {
      *
      * @return codes    size list_size * code_size
      */
-    virtual const uint8_t * get_codes (size_t list_no) const = 0;
+    virtual const uint8_t* get_codes(size_t list_no) const = 0;
 
     /** get the ids for an inverted list
      * must be released by release_ids
      *
      * @return ids      size list_size
      */
-    virtual const idx_t * get_ids (size_t list_no) const = 0;
+    virtual const idx_t* get_ids(size_t list_no) const = 0;
 
     /// release codes returned by get_codes (default implementation is nop
-    virtual void release_codes (size_t list_no, const uint8_t *codes) const;
+    virtual void release_codes(size_t list_no, const uint8_t* codes) const;
 
     /// release ids returned by get_ids
-    virtual void release_ids (size_t list_no, const idx_t *ids) const;
+    virtual void release_ids(size_t list_no, const idx_t* ids) const;
 
     /// @return a single id in an inverted list
-    virtual idx_t get_single_id (size_t list_no, size_t offset) const;
+    virtual idx_t get_single_id(size_t list_no, size_t offset) const;
 
     /// @return a single code in an inverted list
     /// (should be deallocated with release_codes)
-    virtual const uint8_t * get_single_code (
-                size_t list_no, size_t offset) const;
+    virtual const uint8_t* get_single_code(size_t list_no, size_t offset) const;
 
     /// prepare the following lists (default does nothing)
     /// a list can be -1 hence the signed long
-    virtual void prefetch_lists (const idx_t *list_nos, int nlist) const;
+    virtual void prefetch_lists(const idx_t* list_nos, int nlist) const;
 
     /*************************
      * writing functions     */
 
     /// add one entry to an inverted list
-    virtual size_t add_entry (size_t list_no, idx_t theid,
-                              const uint8_t *code);
+    virtual size_t add_entry(size_t list_no, idx_t theid, const uint8_t* code);
 
-    virtual size_t add_entries (
-           size_t list_no, size_t n_entry,
-           const idx_t* ids, const uint8_t *code) = 0;
+    virtual size_t add_entries(
+            size_t list_no,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) = 0;
 
-    virtual void update_entry (size_t list_no, size_t offset,
-                               idx_t id, const uint8_t *code);
+    virtual void update_entry(
+            size_t list_no,
+            size_t offset,
+            idx_t id,
+            const uint8_t* code);
 
-    virtual void update_entries (size_t list_no, size_t offset, size_t n_entry,
-                                 const idx_t *ids, const uint8_t *code) = 0;
+    virtual void update_entries(
+            size_t list_no,
+            size_t offset,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) = 0;
 
-    virtual void resize (size_t list_no, size_t new_size) = 0;
+    virtual void resize(size_t list_no, size_t new_size) = 0;
 
-    virtual void reset ();
+    virtual void reset();
 
     /// move all entries from oivf (empty on output)
-    void merge_from (InvertedLists *oivf, size_t add_id);
+    void merge_from(InvertedLists* oivf, size_t add_id);
 
-    virtual ~InvertedLists ();
+    virtual ~InvertedLists();
 
     /*************************
      * statistics            */
 
     /// 1= perfectly balanced, >1: imbalanced
-    double imbalance_factor () const;
+    double imbalance_factor() const;
 
     /// display some stats about the inverted lists
-    void print_stats () const;
+    void print_stats() const;
 
     /// sum up list sizes
-    size_t compute_ntotal () const;
+    size_t compute_ntotal() const;
 
     /**************************************
      * Scoped inverted lists (for automatic deallocation)
@@ -138,71 +144,76 @@ struct InvertedLists {
      */
 
     struct ScopedIds {
-        const InvertedLists *il;
-        const idx_t *ids;
+        const InvertedLists* il;
+        const idx_t* ids;
         size_t list_no;
 
-        ScopedIds (const InvertedLists *il, size_t list_no):
-        il (il), ids (il->get_ids (list_no)), list_no (list_no)
-        {}
+        ScopedIds(const InvertedLists* il, size_t list_no)
+                : il(il), ids(il->get_ids(list_no)), list_no(list_no) {}
 
-        const idx_t *get() {return ids; }
+        const idx_t* get() {
+            return ids;
+        }
 
-        idx_t operator [] (size_t i) const {
+        idx_t operator[](size_t i) const {
             return ids[i];
         }
 
-        ~ScopedIds () {
-            il->release_ids (list_no, ids);
+        ~ScopedIds() {
+            il->release_ids(list_no, ids);
         }
     };
 
     struct ScopedCodes {
-        const InvertedLists *il;
-        const uint8_t *codes;
+        const InvertedLists* il;
+        const uint8_t* codes;
         size_t list_no;
 
-        ScopedCodes (const InvertedLists *il, size_t list_no):
-            il (il), codes (il->get_codes (list_no)), list_no (list_no)
-        {}
+        ScopedCodes(const InvertedLists* il, size_t list_no)
+                : il(il), codes(il->get_codes(list_no)), list_no(list_no) {}
 
-        ScopedCodes (const InvertedLists *il, size_t list_no, size_t offset):
-            il (il), codes (il->get_single_code (list_no, offset)),
-            list_no (list_no)
-        {}
+        ScopedCodes(const InvertedLists* il, size_t list_no, size_t offset)
+                : il(il),
+                  codes(il->get_single_code(list_no, offset)),
+                  list_no(list_no) {}
 
-        const uint8_t *get() {return codes; }
+        const uint8_t* get() {
+            return codes;
+        }
 
-        ~ScopedCodes () {
-            il->release_codes (list_no, codes);
+        ~ScopedCodes() {
+            il->release_codes(list_no, codes);
         }
     };
-
-
 };
 
-
 /// simple (default) implementation as an array of inverted lists
-struct ArrayInvertedLists: InvertedLists {
-    std::vector < std::vector<uint8_t> > codes; // binary codes, size nlist
-    std::vector < std::vector<idx_t> > ids;  ///< Inverted lists for indexes
+struct ArrayInvertedLists : InvertedLists {
+    std::vector<std::vector<uint8_t>> codes; // binary codes, size nlist
+    std::vector<std::vector<idx_t>> ids;     ///< Inverted lists for indexes
 
-    ArrayInvertedLists (size_t nlist, size_t code_size);
+    ArrayInvertedLists(size_t nlist, size_t code_size);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    size_t add_entries (
-           size_t list_no, size_t n_entry,
-           const idx_t* ids, const uint8_t *code) override;
+    size_t add_entries(
+            size_t list_no,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) override;
 
-    void update_entries (size_t list_no, size_t offset, size_t n_entry,
-                         const idx_t *ids, const uint8_t *code) override;
+    void update_entries(
+            size_t list_no,
+            size_t offset,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) override;
 
-    void resize (size_t list_no, size_t new_size) override;
+    void resize(size_t list_no, size_t new_size) override;
 
-     ~ArrayInvertedLists () override;
+    ~ArrayInvertedLists() override;
 };
 
 /*****************************************************************
@@ -213,154 +224,143 @@ struct ArrayInvertedLists: InvertedLists {
  *****************************************************************/
 
 /// invlists that fail for all write functions
-struct ReadOnlyInvertedLists: InvertedLists {
-
-    ReadOnlyInvertedLists (size_t nlist, size_t code_size):
-    InvertedLists (nlist, code_size) {}
-
-    size_t add_entries (
-           size_t list_no, size_t n_entry,
-           const idx_t* ids, const uint8_t *code) override;
-
-    void update_entries (size_t list_no, size_t offset, size_t n_entry,
-                         const idx_t *ids, const uint8_t *code) override;
-
-    void resize (size_t list_no, size_t new_size) override;
-
+struct ReadOnlyInvertedLists : InvertedLists {
+    ReadOnlyInvertedLists(size_t nlist, size_t code_size)
+            : InvertedLists(nlist, code_size) {}
+
+    size_t add_entries(
+            size_t list_no,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) override;
+
+    void update_entries(
+            size_t list_no,
+            size_t offset,
+            size_t n_entry,
+            const idx_t* ids,
+            const uint8_t* code) override;
+
+    void resize(size_t list_no, size_t new_size) override;
 };
 
-
 /// Horizontal stack of inverted lists
-struct HStackInvertedLists: ReadOnlyInvertedLists {
-
-    std::vector<const InvertedLists *>ils;
+struct HStackInvertedLists : ReadOnlyInvertedLists {
+    std::vector<const InvertedLists*> ils;
 
     /// build InvertedLists by concatenating nil of them
-    HStackInvertedLists (int nil, const InvertedLists **ils);
+    HStackInvertedLists(int nil, const InvertedLists** ils);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
-
-    void prefetch_lists (const idx_t *list_nos, int nlist) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    void release_codes (size_t list_no, const uint8_t *codes) const override;
-    void release_ids (size_t list_no, const idx_t *ids) const override;
+    void prefetch_lists(const idx_t* list_nos, int nlist) const override;
 
-    idx_t get_single_id (size_t list_no, size_t offset) const override;
+    void release_codes(size_t list_no, const uint8_t* codes) const override;
+    void release_ids(size_t list_no, const idx_t* ids) const override;
 
-    const uint8_t * get_single_code (
-           size_t list_no, size_t offset) const override;
+    idx_t get_single_id(size_t list_no, size_t offset) const override;
 
+    const uint8_t* get_single_code(size_t list_no, size_t offset)
+            const override;
 };
 
 using ConcatenatedInvertedLists = HStackInvertedLists;
 
-
 /// vertical slice of indexes in another InvertedLists
-struct SliceInvertedLists: ReadOnlyInvertedLists {
-    const InvertedLists *il;
+struct SliceInvertedLists : ReadOnlyInvertedLists {
+    const InvertedLists* il;
     idx_t i0, i1;
 
-    SliceInvertedLists(const InvertedLists *il, idx_t i0, idx_t i1);
+    SliceInvertedLists(const InvertedLists* il, idx_t i0, idx_t i1);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    void release_codes (size_t list_no, const uint8_t *codes) const override;
-    void release_ids (size_t list_no, const idx_t *ids) const override;
+    void release_codes(size_t list_no, const uint8_t* codes) const override;
+    void release_ids(size_t list_no, const idx_t* ids) const override;
 
-    idx_t get_single_id (size_t list_no, size_t offset) const override;
+    idx_t get_single_id(size_t list_no, size_t offset) const override;
 
-    const uint8_t * get_single_code (
-           size_t list_no, size_t offset) const override;
+    const uint8_t* get_single_code(size_t list_no, size_t offset)
+            const override;
 
-    void prefetch_lists (const idx_t *list_nos, int nlist) const override;
+    void prefetch_lists(const idx_t* list_nos, int nlist) const override;
 };
 
-
-struct VStackInvertedLists: ReadOnlyInvertedLists {
-    std::vector<const InvertedLists *>ils;
+struct VStackInvertedLists : ReadOnlyInvertedLists {
+    std::vector<const InvertedLists*> ils;
     std::vector<idx_t> cumsz;
 
     /// build InvertedLists by concatenating nil of them
-    VStackInvertedLists (int nil, const InvertedLists **ils);
+    VStackInvertedLists(int nil, const InvertedLists** ils);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    void release_codes (size_t list_no, const uint8_t *codes) const override;
-    void release_ids (size_t list_no, const idx_t *ids) const override;
+    void release_codes(size_t list_no, const uint8_t* codes) const override;
+    void release_ids(size_t list_no, const idx_t* ids) const override;
 
-    idx_t get_single_id (size_t list_no, size_t offset) const override;
+    idx_t get_single_id(size_t list_no, size_t offset) const override;
 
-    const uint8_t * get_single_code (
-           size_t list_no, size_t offset) const override;
-
-    void prefetch_lists (const idx_t *list_nos, int nlist) const override;
+    const uint8_t* get_single_code(size_t list_no, size_t offset)
+            const override;
 
+    void prefetch_lists(const idx_t* list_nos, int nlist) const override;
 };
 
-
 /** use the first inverted lists if they are non-empty otherwise use the second
  *
  * This is useful if il1 has a few inverted lists that are too long,
  * and that il0 has replacement lists for those, with empty lists for
  * the others. */
-struct MaskedInvertedLists: ReadOnlyInvertedLists {
-
-    const InvertedLists *il0;
-    const InvertedLists *il1;
+struct MaskedInvertedLists : ReadOnlyInvertedLists {
+    const InvertedLists* il0;
+    const InvertedLists* il1;
 
-    MaskedInvertedLists (const InvertedLists *il0,
-                         const InvertedLists *il1);
+    MaskedInvertedLists(const InvertedLists* il0, const InvertedLists* il1);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    void release_codes (size_t list_no, const uint8_t *codes) const override;
-    void release_ids (size_t list_no, const idx_t *ids) const override;
+    void release_codes(size_t list_no, const uint8_t* codes) const override;
+    void release_ids(size_t list_no, const idx_t* ids) const override;
 
-    idx_t get_single_id (size_t list_no, size_t offset) const override;
+    idx_t get_single_id(size_t list_no, size_t offset) const override;
 
-    const uint8_t * get_single_code (
-           size_t list_no, size_t offset) const override;
-
-    void prefetch_lists (const idx_t *list_nos, int nlist) const override;
+    const uint8_t* get_single_code(size_t list_no, size_t offset)
+            const override;
 
+    void prefetch_lists(const idx_t* list_nos, int nlist) const override;
 };
 
-
 /** if the inverted list in il is smaller than maxsize then return it,
  *  otherwise return an empty invlist */
-struct StopWordsInvertedLists: ReadOnlyInvertedLists {
-
-    const InvertedLists *il0;
+struct StopWordsInvertedLists : ReadOnlyInvertedLists {
+    const InvertedLists* il0;
     size_t maxsize;
 
-    StopWordsInvertedLists (const InvertedLists *il, size_t maxsize);
+    StopWordsInvertedLists(const InvertedLists* il, size_t maxsize);
 
     size_t list_size(size_t list_no) const override;
-    const uint8_t * get_codes (size_t list_no) const override;
-    const idx_t * get_ids (size_t list_no) const override;
+    const uint8_t* get_codes(size_t list_no) const override;
+    const idx_t* get_ids(size_t list_no) const override;
 
-    void release_codes (size_t list_no, const uint8_t *codes) const override;
-    void release_ids (size_t list_no, const idx_t *ids) const override;
+    void release_codes(size_t list_no, const uint8_t* codes) const override;
+    void release_ids(size_t list_no, const idx_t* ids) const override;
 
-    idx_t get_single_id (size_t list_no, size_t offset) const override;
+    idx_t get_single_id(size_t list_no, size_t offset) const override;
 
-    const uint8_t * get_single_code (
-           size_t list_no, size_t offset) const override;
-
-    void prefetch_lists (const idx_t *list_nos, int nlist) const override;
+    const uint8_t* get_single_code(size_t list_no, size_t offset)
+            const override;
 
+    void prefetch_lists(const idx_t* list_nos, int nlist) const override;
 };
 
-
 } // namespace faiss
 
-
 #endif

data/vendor/faiss/faiss/invlists/InvertedListsIOHook.cpp

@@ -7,9 +7,9 @@
 
 #include <faiss/invlists/InvertedListsIOHook.h>
 
+#include <faiss/impl/FaissAssert.h>
 #include <faiss/impl/io.h>
 #include <faiss/impl/io_macros.h>
-#include <faiss/impl/FaissAssert.h>
 
 #include <faiss/invlists/BlockInvertedLists.h>
 
@@ -17,24 +17,21 @@
 #include <faiss/invlists/OnDiskInvertedLists.h>
 #endif // !_MSC_VER
 
-
 namespace faiss {
 
-
 /**********************************************************
  * InvertedListIOHook's
  **********************************************************/
 
 InvertedListsIOHook::InvertedListsIOHook(
-        const std::string & key, const std::string & classname):
-    key(key), classname(classname)
-{}
+        const std::string& key,
+        const std::string& classname)
+        : key(key), classname(classname) {}
 
 namespace {
 
 /// std::vector that deletes its contents
-struct IOHookTable: std::vector<InvertedListsIOHook*> {
-
+struct IOHookTable : std::vector<InvertedListsIOHook*> {
     IOHookTable() {
 #ifndef _MSC_VER
         push_back(new OnDiskInvertedListsIOHook());
@@ -43,7 +40,7 @@ struct IOHookTable: std::vector<InvertedListsIOHook*> {
     }
 
     ~IOHookTable() {
-        for (auto x: *this) {
+        for (auto x : *this) {
             delete x;
         }
     }
@@ -51,44 +48,41 @@ struct IOHookTable: std::vector<InvertedListsIOHook*> {
 
 static IOHookTable InvertedListsIOHook_table;
 
-} // anonymous namepsace
+} // namespace
 
-InvertedListsIOHook* InvertedListsIOHook::lookup(int h)
-{
-    for(const auto & callback: InvertedListsIOHook_table) {
+InvertedListsIOHook* InvertedListsIOHook::lookup(int h) {
+    for (const auto& callback : InvertedListsIOHook_table) {
         if (h == fourcc(callback->key)) {
             return callback;
         }
     }
-    FAISS_THROW_FMT (
-        "read_InvertedLists: could not load ArrayInvertedLists as "
-        "%08x (\"%s\")", h, fourcc_inv_printable(h).c_str()
-    );
+    FAISS_THROW_FMT(
+            "read_InvertedLists: could not load ArrayInvertedLists as "
+            "%08x (\"%s\")",
+            h,
+            fourcc_inv_printable(h).c_str());
 }
 
-InvertedListsIOHook* InvertedListsIOHook::lookup_classname(const std::string & classname)
-{
-    for(const auto & callback: InvertedListsIOHook_table) {
+InvertedListsIOHook* InvertedListsIOHook::lookup_classname(
+        const std::string& classname) {
+    for (const auto& callback : InvertedListsIOHook_table) {
         if (callback->classname == classname) {
             return callback;
         }
     }
-    FAISS_THROW_FMT (
+    FAISS_THROW_FMT(
             "read_InvertedLists: could not find classname %s",
-            classname.c_str()
-    );
+            classname.c_str());
 }
 
-void InvertedListsIOHook::add_callback(InvertedListsIOHook *cb)
-{
+void InvertedListsIOHook::add_callback(InvertedListsIOHook* cb) {
     InvertedListsIOHook_table.push_back(cb);
 }
 
-void InvertedListsIOHook::print_callbacks()
-{
+void InvertedListsIOHook::print_callbacks() {
     printf("registered %zd InvertedListsIOHooks:\n",
-          InvertedListsIOHook_table.size());
-    for(const auto & cb: InvertedListsIOHook_table) {
+           InvertedListsIOHook_table.size());
+    for (const auto& cb : InvertedListsIOHook_table) {
         printf("%08x %s %s\n",
                fourcc(cb->key.c_str()),
                cb->key.c_str(),
@@ -96,11 +90,12 @@ void InvertedListsIOHook::print_callbacks()
     }
 }
 
-InvertedLists * InvertedListsIOHook::read_ArrayInvertedLists(
-        IOReader *, int ,
-        size_t , size_t ,
-        const std::vector<size_t> &) const
-{
+InvertedLists* InvertedListsIOHook::read_ArrayInvertedLists(
+        IOReader*,
+        int,
+        size_t,
+        size_t,
+        const std::vector<size_t>&) const {
     FAISS_THROW_FMT("read to array not implemented for %s", classname.c_str());
 }
 

data/vendor/faiss/faiss/invlists/InvertedListsIOHook.h

@@ -7,14 +7,12 @@
 
 #pragma once
 
-#include <string>
-#include <faiss/invlists/InvertedLists.h>
 #include <faiss/impl/io.h>
-
+#include <faiss/invlists/InvertedLists.h>
+#include <string>
 
 namespace faiss {
 
-
 /** Callbacks to handle other types of InvertedList objects.
  *
  * The callbacks should be registered with add_callback before calling
@@ -26,38 +24,39 @@ namespace faiss {
  * - the class name (as given by typeid.name) at write time
  */
 struct InvertedListsIOHook {
-    const std::string key; ///< string version of the fourcc
+    const std::string key;       ///< string version of the fourcc
     const std::string classname; ///< typeid.name
 
-    InvertedListsIOHook(const std::string & key, const std::string & classname);
+    InvertedListsIOHook(const std::string& key, const std::string& classname);
 
     /// write the index to the IOWriter (including the fourcc)
-    virtual void write(const InvertedLists *ils, IOWriter *f) const = 0;
+    virtual void write(const InvertedLists* ils, IOWriter* f) const = 0;
 
     /// called when the fourcc matches this class's fourcc
-    virtual InvertedLists * read(IOReader *f, int io_flags) const = 0;
+    virtual InvertedLists* read(IOReader* f, int io_flags) const = 0;
 
     /** read from a ArrayInvertedLists into this invertedlist type.
-     * For this to work, the callback has to be enabled and the io_flag has to be set to
-     * IO_FLAG_SKIP_IVF_DATA | (16 upper bits of the fourcc)
+     * For this to work, the callback has to be enabled and the io_flag has to
+     * be set to IO_FLAG_SKIP_IVF_DATA | (16 upper bits of the fourcc)
      *
      * (default implementation fails)
      */
-    virtual InvertedLists * read_ArrayInvertedLists(
-            IOReader *f, int io_flags,
-            size_t nlist, size_t code_size,
-            const std::vector<size_t> &sizes) const;
+    virtual InvertedLists* read_ArrayInvertedLists(
+            IOReader* f,
+            int io_flags,
+            size_t nlist,
+            size_t code_size,
+            const std::vector<size_t>& sizes) const;
 
     virtual ~InvertedListsIOHook() {}
 
     /**************************** Manage the set of callbacks ******/
 
     // transfers ownership
-    static void add_callback(InvertedListsIOHook *);
+    static void add_callback(InvertedListsIOHook*);
     static void print_callbacks();
     static InvertedListsIOHook* lookup(int h);
-    static InvertedListsIOHook* lookup_classname(const std::string & classname);
-
+    static InvertedListsIOHook* lookup_classname(const std::string& classname);
 };
 
-} // namespace faiss
+} // namespace faiss