mirrors
/
faiss
mirror of https://github.com/facebookresearch/faiss.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
faiss/docs/html/structfaiss_1_1SliceInverte...

407 lines
31 KiB
HTML
Raw Normal View History

Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
<meta http-equiv="X-UA-Compatible" content="IE=9"/>
<meta name="generator" content="Doxygen 1.8.5"/>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<title>Faiss: faiss::SliceInvertedLists Struct Reference</title>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<link href="tabs.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="jquery.js"></script>
<script type="text/javascript" src="dynsections.js"></script>
<link href="search/search.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="search/search.js"></script>
<script type="text/javascript">
$(document).ready(function() { searchBox.OnSelectItem(0); });
</script>
<link href="doxygen.css" rel="stylesheet" type="text/css" />
</head>
<body>
<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
<div id="titlearea">
<table cellspacing="0" cellpadding="0">
<tbody>
<tr style="height: 56px;">
<td style="padding-left: 0.5em;">
<div id="projectname">Faiss
</div>
</td>
</tr>
</tbody>
</table>
</div>
<!-- end header part -->
<!-- Generated by Doxygen 1.8.5 -->
<script type="text/javascript">
var searchBox = new SearchBox("searchBox", "search",false,'Search');
</script>
<div id="navrow1" class="tabs">
<ul class="tablist">
<li><a href="index.html"><span>Main&#160;Page</span></a></li>
<li><a href="namespaces.html"><span>Namespaces</span></a></li>
<li class="current"><a href="annotated.html"><span>Classes</span></a></li>
<li><a href="files.html"><span>Files</span></a></li>
<li>
<div id="MSearchBox" class="MSearchBoxInactive">
<span class="left">
<img id="MSearchSelect" src="search/mag_sel.png"
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
alt=""/>
<input type="text" id="MSearchField" value="Search" accesskey="S"
onfocus="searchBox.OnSearchFieldFocus(true)"
onblur="searchBox.OnSearchFieldFocus(false)"
onkeyup="searchBox.OnSearchFieldChange(event)"/>
</span><span class="right">
<a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
</span>
</div>
</li>
</ul>
</div>
<div id="navrow2" class="tabs2">
<ul class="tablist">
<li><a href="annotated.html"><span>Class&#160;List</span></a></li>
<li><a href="classes.html"><span>Class&#160;Index</span></a></li>
<li><a href="hierarchy.html"><span>Class&#160;Hierarchy</span></a></li>
<li><a href="functions.html"><span>Class&#160;Members</span></a></li>
</ul>
</div>
<!-- window showing the filter options -->
<div id="MSearchSelectWindow"
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
onkeydown="return searchBox.OnSearchSelectKey(event)">
<a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(0)"><span class="SelectionMark">&#160;</span>All</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(1)"><span class="SelectionMark">&#160;</span>Classes</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(2)"><span class="SelectionMark">&#160;</span>Namespaces</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(3)"><span class="SelectionMark">&#160;</span>Functions</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(4)"><span class="SelectionMark">&#160;</span>Variables</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(5)"><span class="SelectionMark">&#160;</span>Typedefs</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(6)"><span class="SelectionMark">&#160;</span>Enumerations</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(7)"><span class="SelectionMark">&#160;</span>Enumerator</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(8)"><span class="SelectionMark">&#160;</span>Friends</a></div>
<!-- iframe showing the search results (closed by default) -->
<div id="MSearchResultsWindow">
<iframe src="javascript:void(0)" frameborder="0"
name="MSearchResults" id="MSearchResults">
</iframe>
</div>
<div id="nav-path" class="navpath">
<ul>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<li class="navelem"><a class="el" href="namespacefaiss.html">faiss</a></li><li class="navelem"><a class="el" href="structfaiss_1_1SliceInvertedLists.html">SliceInvertedLists</a></li> </ul>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div>
</div><!-- top -->
<div class="header">
<div class="summary">
<a href="#pub-methods">Public Member Functions</a> &#124;
<a href="#pub-attribs">Public Attributes</a> &#124;
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<a href="structfaiss_1_1SliceInvertedLists-members.html">List of all members</a> </div>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="headertitle">
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<div class="title">faiss::SliceInvertedLists Struct Reference</div> </div>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div><!--header-->
<div class="contents">
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>vertical slice of indexes in another <a class="el" href="structfaiss_1_1InvertedLists.html">InvertedLists</a>
<a href="structfaiss_1_1SliceInvertedLists.html#details">More...</a></p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<p><code>#include &lt;<a class="el" href="InvertedLists_8h_source.html">InvertedLists.h</a>&gt;</code></p>
<div class="dynheader">
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
Inheritance diagram for faiss::SliceInvertedLists:</div>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="dyncontent">
<div class="center">
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<img src="structfaiss_1_1SliceInvertedLists.png" usemap="#faiss::SliceInvertedLists_map" alt=""/>
<map id="faiss::SliceInvertedLists_map" name="faiss::SliceInvertedLists_map">
<area href="structfaiss_1_1ReadOnlyInvertedLists.html" alt="faiss::ReadOnlyInvertedLists" shape="rect" coords="0,56,171,80"/>
<area href="structfaiss_1_1InvertedLists.html" alt="faiss::InvertedLists" shape="rect" coords="0,0,171,24"/>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</map>
</div></div>
<table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-methods"></a>
Public Member Functions</h2></td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="memitem:a09154b5731a6567121c86ce9ef0ed367"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a09154b5731a6567121c86ce9ef0ed367"></a>
&#160;</td><td class="memItemRight" valign="bottom"><b>SliceInvertedLists</b> (const <a class="el" href="structfaiss_1_1InvertedLists.html">InvertedLists</a> *il, idx_t i0, idx_t i1)</td></tr>
<tr class="separator:a09154b5731a6567121c86ce9ef0ed367"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ab58bf7449476fc4638bc0a004f7ecb07"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="ab58bf7449476fc4638bc0a004f7ecb07"></a>
size_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#ab58bf7449476fc4638bc0a004f7ecb07">list_size</a> (size_t list_no) const override</td></tr>
<tr class="memdesc:ab58bf7449476fc4638bc0a004f7ecb07"><td class="mdescLeft">&#160;</td><td class="mdescRight">get the size of a list <br/></td></tr>
<tr class="separator:ab58bf7449476fc4638bc0a004f7ecb07"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a95e1c091bea8695c6df29ed87b3cf0d9"><td class="memItemLeft" align="right" valign="top">const uint8_t *&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#a95e1c091bea8695c6df29ed87b3cf0d9">get_codes</a> (size_t list_no) const override</td></tr>
<tr class="separator:a95e1c091bea8695c6df29ed87b3cf0d9"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ac5eb6df92e3bc9d49d7e88f0ff8d3a88"><td class="memItemLeft" align="right" valign="top">const idx_t *&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#ac5eb6df92e3bc9d49d7e88f0ff8d3a88">get_ids</a> (size_t list_no) const override</td></tr>
<tr class="separator:ac5eb6df92e3bc9d49d7e88f0ff8d3a88"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a362456fac3002d1a5625a13dbfc0752b"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a362456fac3002d1a5625a13dbfc0752b"></a>
void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#a362456fac3002d1a5625a13dbfc0752b">release_codes</a> (size_t list_no, const uint8_t *codes) const override</td></tr>
<tr class="memdesc:a362456fac3002d1a5625a13dbfc0752b"><td class="mdescLeft">&#160;</td><td class="mdescRight">release codes returned by get_codes (default implementation is nop <br/></td></tr>
<tr class="separator:a362456fac3002d1a5625a13dbfc0752b"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ad3001048c22bcff5a1a8fb3c13d6a9ca"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="ad3001048c22bcff5a1a8fb3c13d6a9ca"></a>
void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#ad3001048c22bcff5a1a8fb3c13d6a9ca">release_ids</a> (size_t list_no, const idx_t *ids) const override</td></tr>
<tr class="memdesc:ad3001048c22bcff5a1a8fb3c13d6a9ca"><td class="mdescLeft">&#160;</td><td class="mdescRight">release ids returned by get_ids <br/></td></tr>
<tr class="separator:ad3001048c22bcff5a1a8fb3c13d6a9ca"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a714d282706a7301772fcd145854a976e"><td class="memItemLeft" align="right" valign="top">idx_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#a714d282706a7301772fcd145854a976e">get_single_id</a> (size_t list_no, size_t offset) const override</td></tr>
<tr class="separator:a714d282706a7301772fcd145854a976e"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:abc95a98970bb7b4b044c5a61520c20d3"><td class="memItemLeft" align="right" valign="top">const uint8_t *&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#abc95a98970bb7b4b044c5a61520c20d3">get_single_code</a> (size_t list_no, size_t offset) const override</td></tr>
<tr class="separator:abc95a98970bb7b4b044c5a61520c20d3"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a40f74e21271fa5492fb9b78bb04a514f"><td class="memItemLeft" align="right" valign="top">void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1SliceInvertedLists.html#a40f74e21271fa5492fb9b78bb04a514f">prefetch_lists</a> (const idx_t *list_nos, int <a class="el" href="structfaiss_1_1InvertedLists.html#afaf8b7aa3e846aa4553f6aae67827ec8">nlist</a>) const override</td></tr>
<tr class="separator:a40f74e21271fa5492fb9b78bb04a514f"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="inherit_header pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_structfaiss_1_1ReadOnlyInvertedLists')"><img src="closed.png" alt="-"/>&#160;Public Member Functions inherited from <a class="el" href="structfaiss_1_1ReadOnlyInvertedLists.html">faiss::ReadOnlyInvertedLists</a></td></tr>
<tr class="memitem:a38b8814659bc9dd48a96191aa3f1b01c inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a38b8814659bc9dd48a96191aa3f1b01c"></a>
&#160;</td><td class="memItemRight" valign="bottom"><b>ReadOnlyInvertedLists</b> (size_t <a class="el" href="structfaiss_1_1InvertedLists.html#afaf8b7aa3e846aa4553f6aae67827ec8">nlist</a>, size_t <a class="el" href="structfaiss_1_1InvertedLists.html#ae06049e4259edb47d443ef87bdedc744">code_size</a>)</td></tr>
<tr class="separator:a38b8814659bc9dd48a96191aa3f1b01c inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a0734538244d8c5969e3e23e38faf1c0c inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a0734538244d8c5969e3e23e38faf1c0c"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
size_t&#160;</td><td class="memItemRight" valign="bottom"><b>add_entries</b> (size_t list_no, size_t n_entry, const idx_t *ids, const uint8_t *code) override</td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="separator:a0734538244d8c5969e3e23e38faf1c0c inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a33c3a5174bcf6709dcefec8bdc4b9a52 inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a33c3a5174bcf6709dcefec8bdc4b9a52"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
void&#160;</td><td class="memItemRight" valign="bottom"><b>update_entries</b> (size_t list_no, size_t offset, size_t n_entry, const idx_t *ids, const uint8_t *code) override</td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="separator:a33c3a5174bcf6709dcefec8bdc4b9a52 inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a7b65c76388ba2fbc17a7060f6cf87039 inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a7b65c76388ba2fbc17a7060f6cf87039"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
void&#160;</td><td class="memItemRight" valign="bottom"><b>resize</b> (size_t list_no, size_t new_size) override</td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="separator:a7b65c76388ba2fbc17a7060f6cf87039 inherit pub_methods_structfaiss_1_1ReadOnlyInvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<tr class="inherit_header pub_methods_structfaiss_1_1InvertedLists"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_structfaiss_1_1InvertedLists')"><img src="closed.png" alt="-"/>&#160;Public Member Functions inherited from <a class="el" href="structfaiss_1_1InvertedLists.html">faiss::InvertedLists</a></td></tr>
<tr class="memitem:a83d053d03720d938dc4fc4e2f674619c inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a83d053d03720d938dc4fc4e2f674619c"></a>
&#160;</td><td class="memItemRight" valign="bottom"><b>InvertedLists</b> (size_t <a class="el" href="structfaiss_1_1InvertedLists.html#afaf8b7aa3e846aa4553f6aae67827ec8">nlist</a>, size_t <a class="el" href="structfaiss_1_1InvertedLists.html#ae06049e4259edb47d443ef87bdedc744">code_size</a>)</td></tr>
<tr class="separator:a83d053d03720d938dc4fc4e2f674619c inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:acd3e36cc06fc49e02773e403a84dba95 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="acd3e36cc06fc49e02773e403a84dba95"></a>
virtual size_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#acd3e36cc06fc49e02773e403a84dba95">add_entry</a> (size_t list_no, idx_t theid, const uint8_t *code)</td></tr>
<tr class="memdesc:acd3e36cc06fc49e02773e403a84dba95 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">add one entry to an inverted list <br/></td></tr>
<tr class="separator:acd3e36cc06fc49e02773e403a84dba95 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a57e1382949b91f77acc46790b6431056 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a57e1382949b91f77acc46790b6431056"></a>
virtual void&#160;</td><td class="memItemRight" valign="bottom"><b>update_entry</b> (size_t list_no, size_t offset, idx_t id, const uint8_t *code)</td></tr>
<tr class="separator:a57e1382949b91f77acc46790b6431056 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a72e736253a669dad5f6e5188168e5c60 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a72e736253a669dad5f6e5188168e5c60"></a>
virtual void&#160;</td><td class="memItemRight" valign="bottom"><b>reset</b> ()</td></tr>
<tr class="separator:a72e736253a669dad5f6e5188168e5c60 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ac2a0686121ef5bdd6eb70f7eb3cf9b3a inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="ac2a0686121ef5bdd6eb70f7eb3cf9b3a"></a>
void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#ac2a0686121ef5bdd6eb70f7eb3cf9b3a">merge_from</a> (<a class="el" href="structfaiss_1_1InvertedLists.html">InvertedLists</a> *oivf, size_t add_id)</td></tr>
<tr class="memdesc:ac2a0686121ef5bdd6eb70f7eb3cf9b3a inherit pub_methods_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">move all entries from oivf (empty on output) <br/></td></tr>
<tr class="separator:ac2a0686121ef5bdd6eb70f7eb3cf9b3a inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="memitem:aacf34f6cf392771fe2e68210cd4e009b inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="aacf34f6cf392771fe2e68210cd4e009b"></a>
double&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#aacf34f6cf392771fe2e68210cd4e009b">imbalance_factor</a> () const </td></tr>
<tr class="memdesc:aacf34f6cf392771fe2e68210cd4e009b inherit pub_methods_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">1= perfectly balanced, &gt;1: imbalanced <br/></td></tr>
<tr class="separator:aacf34f6cf392771fe2e68210cd4e009b inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a156de8be4b21a0dfa9a48a3918a0b5c3 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a156de8be4b21a0dfa9a48a3918a0b5c3"></a>
void&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#a156de8be4b21a0dfa9a48a3918a0b5c3">print_stats</a> () const </td></tr>
<tr class="memdesc:a156de8be4b21a0dfa9a48a3918a0b5c3 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">display some stats about the inverted lists <br/></td></tr>
<tr class="separator:a156de8be4b21a0dfa9a48a3918a0b5c3 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:a722eef5a0e5bf5dcf78228a903e16af9 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a722eef5a0e5bf5dcf78228a903e16af9"></a>
size_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#a722eef5a0e5bf5dcf78228a903e16af9">compute_ntotal</a> () const </td></tr>
<tr class="memdesc:a722eef5a0e5bf5dcf78228a903e16af9 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">sum up list sizes <br/></td></tr>
<tr class="separator:a722eef5a0e5bf5dcf78228a903e16af9 inherit pub_methods_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-attribs"></a>
Public Attributes</h2></td></tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<tr class="memitem:a580a773b1857eb771604f2e95ceb28ea"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a580a773b1857eb771604f2e95ceb28ea"></a>
const <a class="el" href="structfaiss_1_1InvertedLists.html">InvertedLists</a> *&#160;</td><td class="memItemRight" valign="bottom"><b>il</b></td></tr>
<tr class="separator:a580a773b1857eb771604f2e95ceb28ea"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ae8443276822f77a2c744a0148bb5fe7f"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="ae8443276822f77a2c744a0148bb5fe7f"></a>
idx_t&#160;</td><td class="memItemRight" valign="bottom"><b>i0</b></td></tr>
<tr class="separator:ae8443276822f77a2c744a0148bb5fe7f"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:abbde49f50e02a506e4f5df42d1afea91"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="abbde49f50e02a506e4f5df42d1afea91"></a>
idx_t&#160;</td><td class="memItemRight" valign="bottom"><b>i1</b></td></tr>
<tr class="separator:abbde49f50e02a506e4f5df42d1afea91"><td class="memSeparator" colspan="2">&#160;</td></tr>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<tr class="inherit_header pub_attribs_structfaiss_1_1InvertedLists"><td colspan="2" onclick="javascript:toggleInherit('pub_attribs_structfaiss_1_1InvertedLists')"><img src="closed.png" alt="-"/>&#160;Public Attributes inherited from <a class="el" href="structfaiss_1_1InvertedLists.html">faiss::InvertedLists</a></td></tr>
<tr class="memitem:afaf8b7aa3e846aa4553f6aae67827ec8 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="afaf8b7aa3e846aa4553f6aae67827ec8"></a>
size_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#afaf8b7aa3e846aa4553f6aae67827ec8">nlist</a></td></tr>
<tr class="memdesc:afaf8b7aa3e846aa4553f6aae67827ec8 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">number of possible key values <br/></td></tr>
<tr class="separator:afaf8b7aa3e846aa4553f6aae67827ec8 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
<tr class="memitem:ae06049e4259edb47d443ef87bdedc744 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="ae06049e4259edb47d443ef87bdedc744"></a>
size_t&#160;</td><td class="memItemRight" valign="bottom"><a class="el" href="structfaiss_1_1InvertedLists.html#ae06049e4259edb47d443ef87bdedc744">code_size</a></td></tr>
<tr class="memdesc:ae06049e4259edb47d443ef87bdedc744 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="mdescLeft">&#160;</td><td class="mdescRight">code size per vector in bytes <br/></td></tr>
<tr class="separator:ae06049e4259edb47d443ef87bdedc744 inherit pub_attribs_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="inherited"></a>
Additional Inherited Members</h2></td></tr>
<tr class="inherit_header pub_types_structfaiss_1_1InvertedLists"><td colspan="2" onclick="javascript:toggleInherit('pub_types_structfaiss_1_1InvertedLists')"><img src="closed.png" alt="-"/>&#160;Public Types inherited from <a class="el" href="structfaiss_1_1InvertedLists.html">faiss::InvertedLists</a></td></tr>
<tr class="memitem:a2bb406f72dcad45cf03974ff14bfa473 inherit pub_types_structfaiss_1_1InvertedLists"><td class="memItemLeft" align="right" valign="top"><a class="anchor" id="a2bb406f72dcad45cf03974ff14bfa473"></a>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
typedef <a class="el" href="structfaiss_1_1Index.html#ad3f0d3071f987baabbdd3da4500c87ea">Index::idx_t</a>&#160;</td><td class="memItemRight" valign="bottom"><b>idx_t</b></td></tr>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<tr class="separator:a2bb406f72dcad45cf03974ff14bfa473 inherit pub_types_structfaiss_1_1InvertedLists"><td class="memSeparator" colspan="2">&#160;</td></tr>
</table>
<a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<div class="textblock"><p>vertical slice of indexes in another <a class="el" href="structfaiss_1_1InvertedLists.html">InvertedLists</a> </p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>Definition at line <a class="el" href="InvertedLists_8h_source.html#l00256">256</a> of file <a class="el" href="InvertedLists_8h_source.html">InvertedLists.h</a>.</p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div><h2 class="groupheader">Member Function Documentation</h2>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<a class="anchor" id="a95e1c091bea8695c6df29ed87b3cf0d9"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<td class="memname">const uint8_t * faiss::SliceInvertedLists::get_codes </td>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<td>(</td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>list_no</em></td><td>)</td>
<td> const</td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<p>get the codes for an inverted list must be released by release_codes</p>
<dl class="section return"><dt>Returns</dt><dd>codes size list_size * code_size </dd></dl>
<p>Implements <a class="el" href="structfaiss_1_1InvertedLists.html#a60b12f9efe940b617a8380adceef4032">faiss::InvertedLists</a>.</p>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>Definition at line <a class="el" href="InvertedLists_8cpp_source.html#l00362">362</a> of file <a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a>.</p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div>
</div>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<a class="anchor" id="ac5eb6df92e3bc9d49d7e88f0ff8d3a88"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<td class="memname">const <a class="el" href="structfaiss_1_1Index.html#ad3f0d3071f987baabbdd3da4500c87ea">Index::idx_t</a> * faiss::SliceInvertedLists::get_ids </td>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<td>(</td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>list_no</em></td><td>)</td>
<td> const</td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<p>get the ids for an inverted list must be released by release_ids</p>
<dl class="section return"><dt>Returns</dt><dd>ids size list_size </dd></dl>
<p>Implements <a class="el" href="structfaiss_1_1InvertedLists.html#a3ab6b389d8d85c5fbf23889064df7e95">faiss::InvertedLists</a>.</p>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>Definition at line <a class="el" href="InvertedLists_8cpp_source.html#l00379">379</a> of file <a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a>.</p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div>
</div>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<a class="anchor" id="abc95a98970bb7b4b044c5a61520c20d3"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<td class="memname">const uint8_t * faiss::SliceInvertedLists::get_single_code </td>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<td>(</td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>list_no</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>offset</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td> const</td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<dl class="section return"><dt>Returns</dt><dd>a single code in an inverted list (should be deallocated with release_codes) </dd></dl>
<p>Reimplemented from <a class="el" href="structfaiss_1_1InvertedLists.html#afc2760bed7c210f2fad9b03109f561ee">faiss::InvertedLists</a>.</p>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>Definition at line <a class="el" href="InvertedLists_8cpp_source.html#l00367">367</a> of file <a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a>.</p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div>
</div>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<a class="anchor" id="a714d282706a7301772fcd145854a976e"></a>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<td class="memname"><a class="el" href="structfaiss_1_1Index.html#ad3f0d3071f987baabbdd3da4500c87ea">Index::idx_t</a> faiss::SliceInvertedLists::get_single_id </td>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
<td>(</td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>list_no</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">size_t&#160;</td>
<td class="paramname"><em>offset</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td> const</td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<dl class="section return"><dt>Returns</dt><dd>a single id in an inverted list </dd></dl>
<p>Reimplemented from <a class="el" href="structfaiss_1_1InvertedLists.html#a21b52cdb0601cd225bd7924932310876">faiss::InvertedLists</a>.</p>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<p>Definition at line <a class="el" href="InvertedLists_8cpp_source.html#l00384">384</a> of file <a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a>.</p>
</div>
</div>
<a class="anchor" id="a40f74e21271fa5492fb9b78bb04a514f"></a>
<div class="memitem">
<div class="memproto">
<table class="mlabels">
<tr>
<td class="mlabels-left">
<table class="memname">
<tr>
<td class="memname">void faiss::SliceInvertedLists::prefetch_lists </td>
<td>(</td>
<td class="paramtype">const idx_t *&#160;</td>
<td class="paramname"><em>list_nos</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">int&#160;</td>
<td class="paramname"><em>nlist</em>&#160;</td>
</tr>
<tr>
<td></td>
<td>)</td>
<td></td><td> const</td>
</tr>
</table>
</td>
<td class="mlabels-right">
<span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
<p>prepare the following lists (default does nothing) a list can be -1 hence the signed long </p>
<p>Reimplemented from <a class="el" href="structfaiss_1_1InvertedLists.html#ac537aa14d2731c0ff907169211389b76">faiss::InvertedLists</a>.</p>
<p>Definition at line <a class="el" href="InvertedLists_8cpp_source.html#l00395">395</a> of file <a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a>.</p>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</div>
</div>
<hr/>The documentation for this struct was generated from the following files:<ul>
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
<li>/data/users/hoss/faiss/<a class="el" href="InvertedLists_8h_source.html">InvertedLists.h</a></li>
<li>/data/users/hoss/faiss/<a class="el" href="InvertedLists_8cpp_source.html">InvertedLists.cpp</a></li>
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
</ul>
</div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
Generated by &#160;<a href="http://www.doxygen.org/index.html">
<img class="footer" src="doxygen.png" alt="doxygen"/>
</a> 1.8.5
</small></address>
</body>
</html>