Adding ivf-flat description to tuning guide

conda/environments/bench_ann_cuda-118_arch-x86_64.yaml

@@ -7,6 +7,7 @@ channels:
 - conda-forge
 - nvidia
 dependencies:
+- benchmark>=1.8.2
 - c-compiler
 - clang-tools=16.0.1
 - clang=16.0.1

dependencies.yaml

@@ -168,6 +168,7 @@ dependencies:
           - glog>=0.6.0
           - h5py>=3.8.0
           - libfaiss>=1.7.1
+          - benchmark>=1.8.2
           - faiss-proc=*=cuda
           - matplotlib
           - pyyaml

docs/source/ann_benchmarks_param_tuning.md

@@ -7,6 +7,10 @@ This guide outlines the various parameter settings that can be specified in [RAF
 
 ### IVF-Flat
 
+IVF-flat uses an inverted-file index, which partitions the vectors into a series of clusters, or lists, storing them in an interleaved format which is optimized for fast distance computation. The searching of an IVF-flat index reduces the total vectors in the index to those within some user-specified nearest clusters called probes.
+
+IVF-flat is a simple algorithm which won't save any space, but it provides competitive search times even at higher levels of recall.
+
 | Parameter | Type           | Data Type             | Description                                                                                                                                                                       |
 |-----------|----------------|-----------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
 | `nlists`  | `build_param`  | Positive Integer `>0` | Number of clusters to partition the vectors into. Larger values will put less points into each cluster but this will impact index build time as more clusters need to be trained. |

docs/source/raft_ann_benchmarks.md

@@ -158,6 +158,8 @@ options:
   --algorithms ALGORITHMS
                         run only comma separated list of named algorithms (default: None)
   --indices INDICES     run only comma separated list of named indices. parameter `algorithms` is ignored (default: None)
+  -k, --count           number of nearest neighbors to return
+  --batch-size          number of query vectors to pass into search
   -f, --force           re-run algorithms even if their results already exist (default: False)
 ```
 `configuration` and `dataset` : `configuration` is a path to a configuration file for a given dataset.