Fix typos in documentation, tutorials, and C API

Summary:
*   `for OpenMP support support` → `for OpenMP support`
*   `neccesary` → `necessary`
*   `journal= {arXiv:1702.08734},,` → `journal = {arXiv:1702.08734},`
*   `verinsion` → `version`
*   `consistency_check--config` → `consistency_check --config`
*   `etracted` → `extracted` (2 occurrences)
*   `sb` → `xb` (variable reference)
*   `statists` → `statistics`
*   `elemnts` → `elements`

Reviewed By: alibeklfc

Differential Revision: D86474214

Author: Subhadeep Karan

INSTALL.md

@@ -84,7 +84,7 @@ found to run on other platforms as well, see
 [other platforms](https://github.com/facebookresearch/faiss/wiki/Related-projects#bindings-to-other-languages-and-porting-to-other-platforms).
 
 The basic requirements are:
-- a C++17 compiler (with support for OpenMP support version 2 or higher),
+- a C++17 compiler (with OpenMP support version 2 or higher),
 - a BLAS implementation (on Intel machines we strongly recommend using Intel MKL for best
 performance).
 
@@ -197,7 +197,7 @@ For AVX512 features available since Intel(R) Sapphire Rapids.
 $ make -C build -j faiss_avx512_spr
 ```
 
-This will ensure the creation of neccesary files when building and installing the python package.
+This will ensure the creation of necessary files when building and installing the python package.
 
 ## Step 3: Building the python bindings (optional)
 

benchs/README.md

@@ -18,7 +18,7 @@ and
 ```
 @inproceedings{JDJ17,
    Author = {Jeff Johnson and Matthijs Douze and Herv{\'e} J{\'e}gou},
-   journal= {arXiv:1702.08734},,
+   journal = {arXiv:1702.08734},
    Title = {Billion-scale similarity search with GPUs},
    Year = {2017},
 }
@@ -150,10 +150,10 @@ The experiments in the appendix are only in the ArXiv version of the paper (tabl
 ```
 python bench_polysemous_1bn.py SIFT1000M OPQ8_64,IMI2x13,PQ8 nprobe={1,2,4,8,16,32,64,128},ht={20,24,26,28,30}
 
-               	R@1    R@10   R@100     time    %pass
-nprobe=1,ht=20 	0.0351 0.0616 0.0751    0.158   19.01
+                        R@1    R@10   R@100     time    %pass
+nprobe=1,ht=20          0.0351 0.0616 0.0751    0.158   19.01
 ...
-nprobe=32,ht=28 	0.1256 0.3563 0.5026    0.561   52.61
+nprobe=32,ht=28         0.1256 0.3563 0.5026    0.561   52.61
 ...
 ```
 Here again the runs are not exactly the same but the original result was obtained from nprobe=32,ht=28.

benchs/link_and_code/README.md

@@ -22,4 +22,4 @@ graph to improve the reconstruction. It is described in
 ArXiV [here](https://arxiv.org/abs/1804.09996)
 
 The necessary code for this paper was removed from Faiss in version 1.8.0.
-For a functioning verinsion, use Faiss 1.7.4.
+For a functioning version, use Faiss 1.7.4.

c_api/IndexFlat_c.h

@@ -35,7 +35,7 @@ int faiss_IndexFlat_new_with(
  *
  * @param index   opaque pointer to index object
  * @param p_xb    output, the pointer to the beginning of `xb`.
- * @param p_size  output, the current size of `sb` in number of float values.
+ * @param p_size  output, the current size of `xb` in number of float values.
  */
 void faiss_IndexFlat_xb(FaissIndexFlat* index, float** p_xb, size_t* p_size);
 

c_api/IndexIVF_c.h

@@ -174,7 +174,7 @@ inline void faiss_IndexIVFStats_init(FaissIndexIVFStats* stats) {
     faiss_IndexIVFStats_reset(stats);
 }
 
-/// global var that collects all statists
+/// global var that collects all statistics
 FaissIndexIVFStats* faiss_get_indexIVF_stats();
 
 #ifdef __cplusplus

c_api/impl/AuxIndexStructures_c.h

@@ -136,7 +136,7 @@ int faiss_BufferList_new(FaissBufferList** p_bl, size_t buffer_size);
 
 int faiss_BufferList_add(FaissBufferList* bl, idx_t id, float dis);
 
-/// copy elemnts ofs:ofs+n-1 seen as linear data in the buffers to
+/// copy elements ofs:ofs+n-1 seen as linear data in the buffers to
 /// tables dest_ids, dest_dis
 int faiss_BufferList_copy_range(
         FaissBufferList* bl,

tutorial/python/7-PQFastScan.py

@@ -16,7 +16,7 @@
 xq[:, 0] += np.arange(nq) / 1000.
 
 m = 8   # 8 specifies that the number of sub-vector is 8
-k = 4   # number of dimension in etracted vector
+k = 4  # number of dimension in extracted vector
 n_bit = 4   # 4 specifies that each sub-vector is encoded as 4 bits
 bbs = 32    # build block size ( bbs % 32 == 0 ) for PQ
 index = faiss.IndexPQFastScan(d, m, n_bit, faiss.METRIC_L2, bbs)

tutorial/python/8-PQFastScanRefine.py

@@ -16,7 +16,7 @@
 xq[:, 0] += np.arange(nq) / 1000.
 
 m = 8  # 8 specifies that the number of sub-vector is 8
-k = 4  # number of dimension in etracted vector
+k = 4  # number of dimension in extracted vector
 n_bit = 4  # 4 specifies that each sub-vector is encoded as 4 bits
 bbs = 32  # build block size ( bbs % 32 == 0 ) for PQ