Skip to content

Commit 69d8143

Browse files
aamijaraamijar
committed
docs fixes
1 parent 6ea3187 commit 69d8143

9 files changed

Lines changed: 37 additions & 26 deletions

File tree

‎cpp/include/cuvs/preprocessing/quantize/binary.hpp‎

Lines changed: 5 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
/*
2-
* SPDX-FileCopyrightText: Copyright (c) 2024-2025, NVIDIA CORPORATION.
2+
* SPDX-FileCopyrightText: Copyright (c) 2024-2026, NVIDIA CORPORATION.
33
* SPDX-License-Identifier: Apache-2.0
44
*/
55

@@ -30,11 +30,10 @@ enum class bit_threshold { zero, mean, sampling_median };
3030
* @brief quantizer parameters.
3131
*/
3232
struct params {
33+
/** Threshold method for binarization. */
3334
bit_threshold threshold = bit_threshold::mean;
3435

35-
/*
36-
* specifies the sampling ratio
37-
*/
36+
/** Specifies the sampling ratio. */
3837
float sampling_ratio = 0.1;
3938
};
4039

@@ -48,8 +47,10 @@ struct params {
4847
*/
4948
template <typename T>
5049
struct quantizer {
50+
/** Threshold vector used for binarization. */
5151
raft::device_vector<T, int64_t> threshold;
5252

53+
/** @brief Construct a quantizer with an empty threshold vector. */
5354
quantizer(raft::resources const& res) : threshold(raft::make_device_vector<T, int64_t>(res, 0)) {}
5455
};
5556

‎cpp/include/cuvs/preprocessing/quantize/pq.hpp‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -84,7 +84,9 @@ struct params {
8484
*/
8585
template <typename T>
8686
struct quantizer {
87+
/** Parameters used to build this quantizer. */
8788
params params_quantizer;
89+
/** VPQ codebooks produced during training. */
8890
cuvs::neighbors::vpq_dataset<T, int64_t> vpq_codebooks;
8991
};
9092

‎cpp/include/cuvs/preprocessing/quantize/scalar.hpp‎

Lines changed: 6 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
/*
2-
* SPDX-FileCopyrightText: Copyright (c) 2024-2025, NVIDIA CORPORATION.
2+
* SPDX-FileCopyrightText: Copyright (c) 2024-2026, NVIDIA CORPORATION.
33
* SPDX-License-Identifier: Apache-2.0
44
*/
55

@@ -24,9 +24,9 @@ namespace cuvs::preprocessing::quantize::scalar {
2424
* @brief quantizer parameters.
2525
*/
2626
struct params {
27-
/*
28-
* specifies how many outliers at top & bottom will be ignored
29-
* needs to be within range of (0, 1]
27+
/**
28+
* Specifies how many outliers at top & bottom will be ignored.
29+
* Needs to be within range of (0, 1].
3030
*/
3131
float quantile = 0.99;
3232
};
@@ -42,7 +42,9 @@ struct params {
4242
*/
4343
template <typename T>
4444
struct quantizer {
45+
/** Minimum value of the quantization range. */
4546
T min_;
47+
/** Maximum value of the quantization range. */
4648
T max_;
4749
};
4850

‎docs/source/cpp_api/cluster_agglomerative.rst‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ namespace *cuvs::cluster::agglomerative*
2121
Agglomerative
2222
-------------
2323

24-
``include <cuvs/cluster/agglomerative.hpp>``
24+
``#include <cuvs/cluster/agglomerative.hpp>``
2525

2626
namespace *cuvs::cluster::agglomerative*
2727

‎docs/source/cpp_api/cluster_kmeans.rst‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ namespace *cuvs::cluster::kmeans*
2121
K-means
2222
-------
2323

24-
``include <cuvs/cluster/kmeans.hpp>``
24+
``#include <cuvs/cluster/kmeans.hpp>``
2525

2626
namespace *cuvs::cluster::kmeans*
2727

@@ -34,7 +34,7 @@ namespace *cuvs::cluster::kmeans*
3434
K-means Helpers
3535
---------------
3636

37-
``include <cuvs/cluster/kmeans.hpp>``
37+
``#include <cuvs/cluster/kmeans.hpp>``
3838

3939
namespace *cuvs::cluster::kmeans::helpers*
4040

‎docs/source/cpp_api/distance.rst‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,7 @@ namespace *cuvs::distance*
2222
Pairwise Distances
2323
------------------
2424

25-
``include <cuvs/distance/distance.hpp>``
25+
``#include <cuvs/distance/distance.hpp>``
2626

2727
namespace *cuvs::distance*
2828

‎docs/source/cpp_api/preprocessing_quantize.rst‎

Lines changed: 18 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -8,25 +8,17 @@ This page provides C++ class references for the publicly-exposed elements of the
88
:language: c++
99
:class: highlight
1010

11-
Scalar
12-
------
13-
14-
``#include <cuvs/preprocessing/quantize/scalar.hpp>``
15-
16-
namespace *cuvs::preprocessing::quantize::scalar*
17-
18-
.. doxygengroup:: scalar
19-
:project: cuvs
20-
21-
Binary
22-
------
11+
Binary Quantizer
12+
----------------
2313

2414
``#include <cuvs/preprocessing/quantize/binary.hpp>``
2515

2616
namespace *cuvs::preprocessing::quantize::binary*
2717

2818
.. doxygengroup:: binary
2919
:project: cuvs
20+
:members:
21+
:content-only:
3022

3123
Product Quantizer
3224
-----------------
@@ -37,3 +29,17 @@ namespace *cuvs::preprocessing::quantize::pq*
3729

3830
.. doxygengroup:: pq
3931
:project: cuvs
32+
:members:
33+
:content-only:
34+
35+
Scalar Quantizer
36+
----------------
37+
38+
``#include <cuvs/preprocessing/quantize/scalar.hpp>``
39+
40+
namespace *cuvs::preprocessing::quantize::scalar*
41+
42+
.. doxygengroup:: scalar
43+
:project: cuvs
44+
:members:
45+
:content-only:

‎docs/source/getting_started.rst‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -45,7 +45,7 @@ New to vector search?
4545

4646
If you are unfamiliar with the basics of vector search or how vector search differs from vector databases, then :doc:`this primer on vector search guide <choosing_and_configuring_indexes>` should provide some good insight. Another good resource for the uninitiated is our :doc:`vector databases vs vector search <vector_databases_vs_vector_search>` guide. As outlined in the primer, vector search as used in vector databases is often closer to machine learning than to traditional databases. This means that while traditional databases can often be slow without any performance tuning, they will usually still yield the correct results. Unfortunately, vector search indexes, like other machine learning models, can yield garbage results of not tuned correctly.
4747

48-
Fortunately, this opens up the whole world of hyperparamer optimization to improve vector search performance and quality. Please see our :doc:`index tuning guide <tuning_guide>` for more information.
48+
Fortunately, this opens up the whole world of hyperparameter optimization to improve vector search performance and quality. Please see our :doc:`index tuning guide <tuning_guide>` for more information.
4949

5050
When comparing the performance of vector search indexes, it is important that considerations are made with respect to three main dimensions:
5151

‎docs/source/index.rst‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -63,7 +63,7 @@ There are several benefits to using cuVS and GPUs for vector search, including
6363
6. Multiple language support
6464
7. Building blocks for composing new or accelerating existing algorithms
6565

66-
In addition to the items above, cuVS shoulders the responsibility of keeping non-trivial accelerated code up to date as new NVIDIA architectures and CUDA versions are released. This provides a deslightful development experimence, guaranteeing that any libraries, databases, or applications built on top of it will always be receiving the best performance and scale.
66+
In addition to the items above, cuVS shoulders the responsibility of keeping non-trivial accelerated code up to date as new NVIDIA architectures and CUDA versions are released. This provides a delightful development experience, guaranteeing that any libraries, databases, or applications built on top of it will always be receiving the best performance and scale.
6767

6868
cuVS Technology Stack
6969
#####################

0 commit comments

Comments
 (0)