| |
| |
| <!DOCTYPE html> |
| <!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]--> |
| <!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]--> |
| <head> |
| <meta charset="utf-8"> |
| |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> |
| |
| <title>apache_beam.transforms.stats — Apache Beam documentation</title> |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" /> |
| |
| |
| |
| |
| |
| <link rel="index" title="Index" |
| href="../../../genindex.html"/> |
| <link rel="search" title="Search" href="../../../search.html"/> |
| <link rel="top" title="Apache Beam documentation" href="../../../index.html"/> |
| <link rel="up" title="Module code" href="../../index.html"/> |
| |
| |
| <script src="../../../_static/js/modernizr.min.js"></script> |
| |
| </head> |
| |
| <body class="wy-body-for-nav" role="document"> |
| |
| |
| <div class="wy-grid-for-nav"> |
| |
| |
| <nav data-toggle="wy-nav-shift" class="wy-nav-side"> |
| <div class="wy-side-scroll"> |
| <div class="wy-side-nav-search"> |
| |
| |
| |
| <a href="../../../index.html" class="icon icon-home"> Apache Beam |
| |
| |
| |
| </a> |
| |
| |
| |
| |
| |
| |
| |
| <div role="search"> |
| <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get"> |
| <input type="text" name="q" placeholder="Search docs" /> |
| <input type="hidden" name="check_keywords" value="yes" /> |
| <input type="hidden" name="area" value="default" /> |
| </form> |
| </div> |
| |
| |
| </div> |
| |
| <div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation"> |
| |
| |
| |
| |
| |
| |
| <ul> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.coders.html">apache_beam.coders package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.internal.html">apache_beam.internal package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.io.html">apache_beam.io package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.metrics.html">apache_beam.metrics package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.options.html">apache_beam.options package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.portability.html">apache_beam.portability package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.runners.html">apache_beam.runners package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.testing.html">apache_beam.testing package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.tools.html">apache_beam.tools package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.transforms.html">apache_beam.transforms package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.typehints.html">apache_beam.typehints package</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.utils.html">apache_beam.utils package</a></li> |
| </ul> |
| <ul> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.error.html">apache_beam.error module</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.pipeline.html">apache_beam.pipeline module</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.pvalue.html">apache_beam.pvalue module</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../../apache_beam.version.html">apache_beam.version module</a></li> |
| </ul> |
| |
| |
| |
| </div> |
| </div> |
| </nav> |
| |
| <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"> |
| |
| |
| <nav class="wy-nav-top" role="navigation" aria-label="top navigation"> |
| |
| <i data-toggle="wy-nav-top" class="fa fa-bars"></i> |
| <a href="../../../index.html">Apache Beam</a> |
| |
| </nav> |
| |
| |
| |
| <div class="wy-nav-content"> |
| <div class="rst-content"> |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| <div role="navigation" aria-label="breadcrumbs navigation"> |
| |
| <ul class="wy-breadcrumbs"> |
| |
| <li><a href="../../../index.html">Docs</a> »</li> |
| |
| <li><a href="../../index.html">Module code</a> »</li> |
| |
| <li>apache_beam.transforms.stats</li> |
| |
| |
| <li class="wy-breadcrumbs-aside"> |
| |
| |
| |
| </li> |
| |
| </ul> |
| |
| |
| <hr/> |
| </div> |
| <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article"> |
| <div itemprop="articleBody"> |
| |
| <h1>Source code for apache_beam.transforms.stats</h1><div class="highlight"><pre> |
| <span></span><span class="c1">#</span> |
| <span class="c1"># Licensed to the Apache Software Foundation (ASF) under one or more</span> |
| <span class="c1"># contributor license agreements. See the NOTICE file distributed with</span> |
| <span class="c1"># this work for additional information regarding copyright ownership.</span> |
| <span class="c1"># The ASF licenses this file to You under the Apache License, Version 2.0</span> |
| <span class="c1"># (the "License"); you may not use this file except in compliance with</span> |
| <span class="c1"># the License. You may obtain a copy of the License at</span> |
| <span class="c1">#</span> |
| <span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span> |
| <span class="c1">#</span> |
| <span class="c1"># Unless required by applicable law or agreed to in writing, software</span> |
| <span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span> |
| <span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span> |
| <span class="c1"># See the License for the specific language governing permissions and</span> |
| <span class="c1"># limitations under the License.</span> |
| <span class="c1">#</span> |
| |
| <span class="sd">"""This module has all statistic related transforms."""</span> |
| |
| <span class="c1"># pytype: skip-file</span> |
| |
| <span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">absolute_import</span> |
| <span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">division</span> |
| |
| <span class="kn">import</span> <span class="nn">heapq</span> |
| <span class="kn">import</span> <span class="nn">itertools</span> |
| <span class="kn">import</span> <span class="nn">math</span> |
| <span class="kn">import</span> <span class="nn">sys</span> |
| <span class="kn">import</span> <span class="nn">typing</span> |
| <span class="kn">from</span> <span class="nn">builtins</span> <span class="kn">import</span> <span class="nb">round</span> |
| |
| <span class="kn">from</span> <span class="nn">apache_beam</span> <span class="kn">import</span> <span class="n">coders</span> |
| <span class="kn">from</span> <span class="nn">apache_beam</span> <span class="kn">import</span> <span class="n">typehints</span> |
| <span class="kn">from</span> <span class="nn">apache_beam.transforms.core</span> <span class="kn">import</span> <span class="o">*</span> |
| <span class="kn">from</span> <span class="nn">apache_beam.transforms.display</span> <span class="kn">import</span> <span class="n">DisplayDataItem</span> |
| <span class="kn">from</span> <span class="nn">apache_beam.transforms.ptransform</span> <span class="kn">import</span> <span class="n">PTransform</span> |
| |
| <span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span> |
| <span class="s1">'ApproximateQuantiles'</span><span class="p">,</span> |
| <span class="s1">'ApproximateUnique'</span><span class="p">,</span> |
| <span class="p">]</span> |
| |
| <span class="c1"># Type variables</span> |
| <span class="n">T</span> <span class="o">=</span> <span class="n">typing</span><span class="o">.</span><span class="n">TypeVar</span><span class="p">(</span><span class="s1">'T'</span><span class="p">)</span> |
| <span class="n">K</span> <span class="o">=</span> <span class="n">typing</span><span class="o">.</span><span class="n">TypeVar</span><span class="p">(</span><span class="s1">'K'</span><span class="p">)</span> |
| <span class="n">V</span> <span class="o">=</span> <span class="n">typing</span><span class="o">.</span><span class="n">TypeVar</span><span class="p">(</span><span class="s1">'V'</span><span class="p">)</span> |
| |
| |
| <div class="viewcode-block" id="ApproximateUnique"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique">[docs]</a><span class="k">class</span> <span class="nc">ApproximateUnique</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Hashes input elements and uses those to extrapolate the size of the entire</span> |
| <span class="sd"> set of hash values by assuming the rest of the hash values are as densely</span> |
| <span class="sd"> distributed as the sample space.</span> |
| <span class="sd"> """</span> |
| |
| <span class="n">_NO_VALUE_ERR_MSG</span> <span class="o">=</span> <span class="s1">'Either size or error should be set. Received </span><span class="si">{}</span><span class="s1">.'</span> |
| <span class="n">_MULTI_VALUE_ERR_MSG</span> <span class="o">=</span> <span class="s1">'Either size or error should be set. '</span> \ |
| <span class="s1">'Received {size = </span><span class="si">%s</span><span class="s1">, error = </span><span class="si">%s</span><span class="s1">}.'</span> |
| <span class="n">_INPUT_SIZE_ERR_MSG</span> <span class="o">=</span> <span class="s1">'ApproximateUnique needs a size >= 16 for an error '</span> \ |
| <span class="s1">'<= 0.50. In general, the estimation error is about '</span> \ |
| <span class="s1">'2 / sqrt(sample_size). Received {size = </span><span class="si">%s</span><span class="s1">}.'</span> |
| <span class="n">_INPUT_ERROR_ERR_MSG</span> <span class="o">=</span> <span class="s1">'ApproximateUnique needs an estimation error '</span> \ |
| <span class="s1">'between 0.01 and 0.50. Received {error = </span><span class="si">%s</span><span class="s1">}.'</span> |
| |
| <div class="viewcode-block" id="ApproximateUnique.parse_input_params"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique.parse_input_params">[docs]</a> <span class="nd">@staticmethod</span> |
| <span class="k">def</span> <span class="nf">parse_input_params</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">error</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Check if input params are valid and return sample size.</span> |
| |
| <span class="sd"> :param size: an int not smaller than 16, which we would use to estimate</span> |
| <span class="sd"> number of unique values.</span> |
| <span class="sd"> :param error: max estimation error, which is a float between 0.01 and 0.50.</span> |
| <span class="sd"> If error is given, sample size will be calculated from error with</span> |
| <span class="sd"> _get_sample_size_from_est_error function.</span> |
| <span class="sd"> :return: sample size</span> |
| <span class="sd"> :raises:</span> |
| <span class="sd"> ValueError: If both size and error are given, or neither is given, or</span> |
| <span class="sd"> values are out of range.</span> |
| <span class="sd"> """</span> |
| |
| <span class="k">if</span> <span class="kc">None</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error</span><span class="p">):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">_MULTI_VALUE_ERR_MSG</span> <span class="o">%</span> <span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error</span><span class="p">))</span> |
| <span class="k">elif</span> <span class="n">size</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">error</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">_NO_VALUE_ERR_MSG</span><span class="p">)</span> |
| <span class="k">elif</span> <span class="n">size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="ow">or</span> <span class="n">size</span> <span class="o"><</span> <span class="mi">16</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">_INPUT_SIZE_ERR_MSG</span> <span class="o">%</span> <span class="p">(</span><span class="n">size</span><span class="p">))</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">return</span> <span class="n">size</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">if</span> <span class="n">error</span> <span class="o"><</span> <span class="mf">0.01</span> <span class="ow">or</span> <span class="n">error</span> <span class="o">></span> <span class="mf">0.5</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">_INPUT_ERROR_ERR_MSG</span> <span class="o">%</span> <span class="p">(</span><span class="n">error</span><span class="p">))</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">return</span> <span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">_get_sample_size_from_est_error</span><span class="p">(</span><span class="n">error</span><span class="p">)</span></div> |
| |
| <span class="nd">@staticmethod</span> |
| <span class="k">def</span> <span class="nf">_get_sample_size_from_est_error</span><span class="p">(</span><span class="n">est_err</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> :return: sample size</span> |
| |
| <span class="sd"> Calculate sample size from estimation error</span> |
| <span class="sd"> """</span> |
| <span class="c1">#math.ceil in python2.7 returns a float, while it returns an int in python3.</span> |
| <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="mf">4.0</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">est_err</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">)))</span> |
| |
| <div class="viewcode-block" id="ApproximateUnique.Globally"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique.Globally">[docs]</a> <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_input_types</span><span class="p">(</span><span class="n">T</span><span class="p">)</span> |
| <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_output_types</span><span class="p">(</span><span class="nb">int</span><span class="p">)</span> |
| <span class="k">class</span> <span class="nc">Globally</span><span class="p">(</span><span class="n">PTransform</span><span class="p">):</span> |
| <span class="sd">""" Approximate.Globally approximate number of unique values"""</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">error</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="o">=</span> <span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">parse_input_params</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error</span><span class="p">)</span> |
| |
| <div class="viewcode-block" id="ApproximateUnique.Globally.expand"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique.Globally.expand">[docs]</a> <span class="k">def</span> <span class="nf">expand</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pcoll</span><span class="p">):</span> |
| <span class="n">coder</span> <span class="o">=</span> <span class="n">coders</span><span class="o">.</span><span class="n">registry</span><span class="o">.</span><span class="n">get_coder</span><span class="p">(</span><span class="n">pcoll</span><span class="p">)</span> |
| <span class="k">return</span> <span class="n">pcoll</span> \ |
| <span class="o">|</span> <span class="s1">'CountGlobalUniqueValues'</span> \ |
| <span class="o">>></span> <span class="p">(</span><span class="n">CombineGlobally</span><span class="p">(</span><span class="n">ApproximateUniqueCombineFn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">,</span> |
| <span class="n">coder</span><span class="p">)))</span></div></div> |
| |
| <div class="viewcode-block" id="ApproximateUnique.PerKey"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique.PerKey">[docs]</a> <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_input_types</span><span class="p">(</span><span class="n">typing</span><span class="o">.</span><span class="n">Tuple</span><span class="p">[</span><span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">])</span> |
| <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_output_types</span><span class="p">(</span><span class="n">typing</span><span class="o">.</span><span class="n">Tuple</span><span class="p">[</span><span class="n">K</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> |
| <span class="k">class</span> <span class="nc">PerKey</span><span class="p">(</span><span class="n">PTransform</span><span class="p">):</span> |
| <span class="sd">""" Approximate.PerKey approximate number of unique values per key"""</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">error</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="o">=</span> <span class="n">ApproximateUnique</span><span class="o">.</span><span class="n">parse_input_params</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error</span><span class="p">)</span> |
| |
| <div class="viewcode-block" id="ApproximateUnique.PerKey.expand"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateUnique.PerKey.expand">[docs]</a> <span class="k">def</span> <span class="nf">expand</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pcoll</span><span class="p">):</span> |
| <span class="n">coder</span> <span class="o">=</span> <span class="n">coders</span><span class="o">.</span><span class="n">registry</span><span class="o">.</span><span class="n">get_coder</span><span class="p">(</span><span class="n">pcoll</span><span class="p">)</span> |
| <span class="k">return</span> <span class="n">pcoll</span> \ |
| <span class="o">|</span> <span class="s1">'CountPerKeyUniqueValues'</span> \ |
| <span class="o">>></span> <span class="p">(</span><span class="n">CombinePerKey</span><span class="p">(</span><span class="n">ApproximateUniqueCombineFn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">,</span> |
| <span class="n">coder</span><span class="p">)))</span></div></div></div> |
| |
| |
| <span class="k">class</span> <span class="nc">_LargestUnique</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> An object to keep samples and calculate sample hash space. It is an</span> |
| <span class="sd"> accumulator of a combine function.</span> |
| <span class="sd"> """</span> |
| <span class="n">_HASH_SPACE_SIZE</span> <span class="o">=</span> <span class="mf">2.0</span> <span class="o">*</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sample_size</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="o">=</span> <span class="n">sample_size</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span> <span class="o">=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span> <span class="o">=</span> <span class="p">[]</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_set</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span> |
| |
| <span class="k">def</span> <span class="nf">add</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">element</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> :param an element from pcoll.</span> |
| <span class="sd"> :return: boolean type whether the value is in the heap</span> |
| |
| <span class="sd"> Adds a value to the heap, returning whether the value is (large enough to</span> |
| <span class="sd"> be) in the heap.</span> |
| <span class="sd"> """</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">)</span> <span class="o">>=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="ow">and</span> <span class="n">element</span> <span class="o"><</span> <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span><span class="p">:</span> |
| <span class="k">return</span> <span class="kc">False</span> |
| |
| <span class="k">if</span> <span class="n">element</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_set</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_set</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">element</span><span class="p">)</span> |
| <span class="n">heapq</span><span class="o">.</span><span class="n">heappush</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">,</span> <span class="n">element</span><span class="p">)</span> |
| |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">)</span> <span class="o">></span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">:</span> |
| <span class="n">temp</span> <span class="o">=</span> <span class="n">heapq</span><span class="o">.</span><span class="n">heappop</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">)</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_set</span><span class="o">.</span><span class="n">remove</span><span class="p">(</span><span class="n">temp</span><span class="p">)</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> |
| <span class="k">elif</span> <span class="n">element</span> <span class="o"><</span> <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span> <span class="o">=</span> <span class="n">element</span> |
| |
| <span class="k">return</span> <span class="kc">True</span> |
| |
| <span class="k">def</span> <span class="nf">get_estimate</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> :return: estimation count of unique values</span> |
| |
| <span class="sd"> If heap size is smaller than sample size, just return heap size.</span> |
| <span class="sd"> Otherwise, takes into account the possibility of hash collisions,</span> |
| <span class="sd"> which become more likely than not for 2^32 distinct elements.</span> |
| <span class="sd"> Note that log(1+x) ~ x for small x, so for sampleSize << maxHash</span> |
| <span class="sd"> log(1 - sample_size/sample_space) / log(1 - 1/sample_space) ~ sample_size</span> |
| <span class="sd"> and hence estimate ~ sample_size * hash_space / sample_space</span> |
| <span class="sd"> as one would expect.</span> |
| |
| <span class="sd"> Given sample_size / sample_space = est / hash_space</span> |
| <span class="sd"> est = sample_size * hash_space / sample_space</span> |
| |
| <span class="sd"> Given above sample_size approximate,</span> |
| <span class="sd"> est = log1p(-sample_size/sample_space) / log1p(-1/sample_space)</span> |
| <span class="sd"> * hash_space / sample_space</span> |
| <span class="sd"> """</span> |
| |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">)</span> <span class="o"><</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">:</span> |
| <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="n">sample_space_size</span> <span class="o">=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span> <span class="o">-</span> <span class="mf">1.0</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">_min_hash</span> |
| <span class="n">est</span> <span class="o">=</span> <span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">log1p</span><span class="p">(</span><span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="o">/</span> <span class="n">sample_space_size</span><span class="p">)</span> |
| <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">log1p</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span> <span class="o">/</span> <span class="n">sample_space_size</span><span class="p">)</span> |
| <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">_HASH_SPACE_SIZE</span> |
| <span class="o">/</span> <span class="n">sample_space_size</span><span class="p">)</span> |
| |
| <span class="k">return</span> <span class="nb">round</span><span class="p">(</span><span class="n">est</span><span class="p">)</span> |
| |
| |
| <span class="k">class</span> <span class="nc">ApproximateUniqueCombineFn</span><span class="p">(</span><span class="n">CombineFn</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> ApproximateUniqueCombineFn computes an estimate of the number of</span> |
| <span class="sd"> unique values that were combined.</span> |
| <span class="sd"> """</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sample_size</span><span class="p">,</span> <span class="n">coder</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span> <span class="o">=</span> <span class="n">sample_size</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_coder</span> <span class="o">=</span> <span class="n">coder</span> |
| |
| <span class="k">def</span> <span class="nf">create_accumulator</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">_LargestUnique</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">)</span> |
| |
| <span class="k">def</span> <span class="nf">add_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">accumulator</span><span class="p">,</span> <span class="n">element</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> |
| <span class="k">try</span><span class="p">:</span> |
| <span class="n">accumulator</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="nb">hash</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_coder</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">element</span><span class="p">)))</span> |
| <span class="k">return</span> <span class="n">accumulator</span> |
| <span class="k">except</span> <span class="ne">Exception</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">"Runtime exception: </span><span class="si">%s</span><span class="s2">"</span><span class="p">,</span> <span class="n">e</span><span class="p">)</span> |
| |
| <span class="c1"># created an issue https://issues.apache.org/jira/browse/BEAM-7285 to speed up</span> |
| <span class="c1"># merge process.</span> |
| <span class="k">def</span> <span class="nf">merge_accumulators</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">accumulators</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span> |
| <span class="n">merged_accumulator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_accumulator</span><span class="p">()</span> |
| <span class="k">for</span> <span class="n">accumulator</span> <span class="ow">in</span> <span class="n">accumulators</span><span class="p">:</span> |
| <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">_sample_heap</span><span class="p">:</span> |
| <span class="n">merged_accumulator</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> |
| |
| <span class="k">return</span> <span class="n">merged_accumulator</span> |
| |
| <span class="nd">@staticmethod</span> |
| <span class="k">def</span> <span class="nf">extract_output</span><span class="p">(</span><span class="n">accumulator</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">get_estimate</span><span class="p">()</span> |
| |
| <span class="k">def</span> <span class="nf">display_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="k">return</span> <span class="p">{</span><span class="s1">'sample_size'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_size</span><span class="p">}</span> |
| |
| |
| <div class="viewcode-block" id="ApproximateQuantiles"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles">[docs]</a><span class="k">class</span> <span class="nc">ApproximateQuantiles</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> PTransfrom for getting the idea of data distribution using approximate N-tile</span> |
| <span class="sd"> (e.g. quartiles, percentiles etc.) either globally or per-key.</span> |
| <span class="sd"> """</span> |
| |
| <span class="nd">@staticmethod</span> |
| <span class="k">def</span> <span class="nf">_display_data</span><span class="p">(</span><span class="n">num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">reverse</span><span class="p">):</span> |
| <span class="k">return</span> <span class="p">{</span> |
| <span class="s1">'num_quantiles'</span><span class="p">:</span> <span class="n">DisplayDataItem</span><span class="p">(</span><span class="n">num_quantiles</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Quantile Count"</span><span class="p">),</span> |
| <span class="s1">'key'</span><span class="p">:</span> <span class="n">DisplayDataItem</span><span class="p">(</span><span class="n">key</span><span class="o">.</span><span class="vm">__name__</span> <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">key</span><span class="p">,</span> <span class="s1">'__name__'</span><span class="p">)</span> |
| <span class="k">else</span> <span class="n">key</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">,</span> |
| <span class="n">label</span><span class="o">=</span><span class="s1">'Record Comparer Key'</span><span class="p">),</span> |
| <span class="s1">'reverse'</span><span class="p">:</span> <span class="n">DisplayDataItem</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">reverse</span><span class="p">),</span> <span class="n">label</span><span class="o">=</span><span class="s1">'Is reversed'</span><span class="p">)</span> |
| <span class="p">}</span> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.Globally"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.Globally">[docs]</a> <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_input_types</span><span class="p">(</span><span class="n">T</span><span class="p">)</span> |
| <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_output_types</span><span class="p">(</span><span class="n">typing</span><span class="o">.</span><span class="n">List</span><span class="p">[</span><span class="n">T</span><span class="p">])</span> |
| <span class="k">class</span> <span class="nc">Globally</span><span class="p">(</span><span class="n">PTransform</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> PTransform takes PCollection and returns a list whose single value is</span> |
| <span class="sd"> approximate N-tiles of the input collection globally.</span> |
| |
| <span class="sd"> Args:</span> |
| <span class="sd"> num_quantiles: number of elements in the resulting quantiles values list.</span> |
| <span class="sd"> key: (optional) Key is a mapping of elements to a comparable key, similar</span> |
| <span class="sd"> to the key argument of Python's sorting methods.</span> |
| <span class="sd"> reverse: (optional) whether to order things smallest to largest, rather</span> |
| <span class="sd"> than largest to smallest</span> |
| <span class="sd"> """</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">reverse</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">=</span> <span class="n">num_quantiles</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_key</span> <span class="o">=</span> <span class="n">key</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span> <span class="o">=</span> <span class="n">reverse</span> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.Globally.expand"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.Globally.expand">[docs]</a> <span class="k">def</span> <span class="nf">expand</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pcoll</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">pcoll</span> <span class="o">|</span> <span class="n">CombineGlobally</span><span class="p">(</span><span class="n">ApproximateQuantilesCombineFn</span><span class="o">.</span><span class="n">create</span><span class="p">(</span> |
| <span class="n">num_quantiles</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">))</span></div> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.Globally.display_data"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.Globally.display_data">[docs]</a> <span class="k">def</span> <span class="nf">display_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">ApproximateQuantiles</span><span class="o">.</span><span class="n">_display_data</span><span class="p">(</span> |
| <span class="n">num_quantiles</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">)</span></div></div> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.PerKey"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.PerKey">[docs]</a> <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_input_types</span><span class="p">(</span><span class="n">typing</span><span class="o">.</span><span class="n">Tuple</span><span class="p">[</span><span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">])</span> |
| <span class="nd">@typehints</span><span class="o">.</span><span class="n">with_output_types</span><span class="p">(</span><span class="n">typing</span><span class="o">.</span><span class="n">Tuple</span><span class="p">[</span><span class="n">K</span><span class="p">,</span> <span class="n">typing</span><span class="o">.</span><span class="n">List</span><span class="p">[</span><span class="n">V</span><span class="p">]])</span> |
| <span class="k">class</span> <span class="nc">PerKey</span><span class="p">(</span><span class="n">PTransform</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> PTransform takes PCollection of KV and returns a list based on each key</span> |
| <span class="sd"> whose single value is list of approximate N-tiles of the input element of</span> |
| <span class="sd"> the key.</span> |
| |
| <span class="sd"> Args:</span> |
| <span class="sd"> num_quantiles: number of elements in the resulting quantiles values list.</span> |
| <span class="sd"> key: (optional) Key is a mapping of elements to a comparable key, similar</span> |
| <span class="sd"> to the key argument of Python's sorting methods.</span> |
| <span class="sd"> reverse: (optional) whether to order things smallest to largest, rather</span> |
| <span class="sd"> than largest to smallest</span> |
| <span class="sd"> """</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">reverse</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">=</span> <span class="n">num_quantiles</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_key</span> <span class="o">=</span> <span class="n">key</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span> <span class="o">=</span> <span class="n">reverse</span> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.PerKey.expand"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.PerKey.expand">[docs]</a> <span class="k">def</span> <span class="nf">expand</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pcoll</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">pcoll</span> <span class="o">|</span> <span class="n">CombinePerKey</span><span class="p">(</span><span class="n">ApproximateQuantilesCombineFn</span><span class="o">.</span><span class="n">create</span><span class="p">(</span> |
| <span class="n">num_quantiles</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">))</span></div> |
| |
| <div class="viewcode-block" id="ApproximateQuantiles.PerKey.display_data"><a class="viewcode-back" href="../../../apache_beam.transforms.stats.html#apache_beam.transforms.stats.ApproximateQuantiles.PerKey.display_data">[docs]</a> <span class="k">def</span> <span class="nf">display_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">ApproximateQuantiles</span><span class="o">.</span><span class="n">_display_data</span><span class="p">(</span> |
| <span class="n">num_quantiles</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">)</span></div></div></div> |
| |
| |
| <span class="k">class</span> <span class="nc">_QuantileBuffer</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="sd">"""A single buffer in the sense of the referenced algorithm.</span> |
| <span class="sd"> (see http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.6.6513&rep=rep1</span> |
| <span class="sd"> &type=pdf and ApproximateQuantilesCombineFn for further information)"""</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">elements</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">elements</span> <span class="o">=</span> <span class="n">elements</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">level</span> <span class="o">=</span> <span class="n">level</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">weight</span> <span class="o">=</span> <span class="n">weight</span> |
| |
| <span class="k">def</span> <span class="fm">__lt__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">elements</span> <span class="o"><</span> <span class="n">other</span><span class="o">.</span><span class="n">elements</span> |
| |
| <span class="k">def</span> <span class="nf">sized_iterator</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| |
| <span class="k">class</span> <span class="nc">QuantileBufferIterator</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">elem</span><span class="p">,</span> <span class="n">weight</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_iter</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span><span class="n">elem</span><span class="p">)</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">weight</span> <span class="o">=</span> <span class="n">weight</span> |
| |
| <span class="k">def</span> <span class="fm">__iter__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="k">return</span> <span class="bp">self</span> |
| |
| <span class="k">def</span> <span class="fm">__next__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="n">value</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_iter</span><span class="p">)</span> |
| <span class="k">return</span> <span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span> |
| |
| <span class="nb">next</span> <span class="o">=</span> <span class="fm">__next__</span> <span class="c1"># For Python 2</span> |
| |
| <span class="k">return</span> <span class="n">QuantileBufferIterator</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">elements</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span> |
| |
| |
| <span class="k">class</span> <span class="nc">_QuantileState</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Compact summarization of a collection on which quantiles can be estimated.</span> |
| <span class="sd"> """</span> |
| <span class="n">min_val</span> <span class="o">=</span> <span class="kc">None</span> <span class="c1"># Holds smallest item in the list</span> |
| <span class="n">max_val</span> <span class="o">=</span> <span class="kc">None</span> <span class="c1"># Holds largest item in the list</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">buffer_size</span><span class="p">,</span> <span class="n">num_buffers</span><span class="p">,</span> <span class="n">unbuffered_elements</span><span class="p">,</span> <span class="n">buffers</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">buffer_size</span> <span class="o">=</span> <span class="n">buffer_size</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">num_buffers</span> <span class="o">=</span> <span class="n">num_buffers</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">buffers</span> <span class="o">=</span> <span class="n">buffers</span> |
| |
| <span class="c1"># The algorithm requires that the manipulated buffers always be filled to</span> |
| <span class="c1"># capacity to perform the collapse operation. This operation can be extended</span> |
| <span class="c1"># to buffers of varying sizes by introducing the notion of fractional</span> |
| <span class="c1"># weights, but it's easier to simply combine the remainders from all shards</span> |
| <span class="c1"># into new, full buffers and then take them into account when computing the</span> |
| <span class="c1"># final output.</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">unbuffered_elements</span> <span class="o">=</span> <span class="n">unbuffered_elements</span> |
| |
| <span class="k">def</span> <span class="nf">is_empty</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="sd">"""Check if the buffered & unbuffered elements are empty or not."""</span> |
| <span class="k">return</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">unbuffered_elements</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">buffers</span> |
| |
| |
| <span class="k">class</span> <span class="nc">ApproximateQuantilesCombineFn</span><span class="p">(</span><span class="n">CombineFn</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> This combiner gives an idea of the distribution of a collection of values</span> |
| <span class="sd"> using approximate N-tiles. The output of this combiner is the list of size of</span> |
| <span class="sd"> the number of quantiles (num_quantiles), containing the input values of the</span> |
| <span class="sd"> minimum value item of the list, the intermediate values (n-tiles) and the</span> |
| <span class="sd"> maximum value item of the list, in the sort order provided via key (similar</span> |
| <span class="sd"> to the key argument of Python's sorting methods).</span> |
| |
| <span class="sd"> If there are fewer values to combine than the number of quantile</span> |
| <span class="sd"> (num_quantiles), then the resulting list will contain all the values being</span> |
| <span class="sd"> combined, in sorted order.</span> |
| |
| <span class="sd"> If no `key` is provided, then the results are sorted in the natural order.</span> |
| |
| <span class="sd"> To evaluate the quantiles, we use the "New Algorithm" described here:</span> |
| |
| <span class="sd"> [MRL98] Manku, Rajagopalan & Lindsay, "Approximate Medians and other</span> |
| <span class="sd"> Quantiles in One Pass and with Limited Memory", Proc. 1998 ACM SIGMOD,</span> |
| <span class="sd"> Vol 27, No 2, p 426-435, June 1998.</span> |
| <span class="sd"> http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.6.6513&rep=rep1</span> |
| <span class="sd"> &type=pdf</span> |
| |
| <span class="sd"> The default error bound is (1 / N), though in practice the accuracy</span> |
| <span class="sd"> tends to be much better.</span> |
| |
| <span class="sd"> Args:</span> |
| <span class="sd"> num_quantiles: Number of quantiles to produce. It is the size of the final</span> |
| <span class="sd"> output list, including the mininum and maximum value items.</span> |
| <span class="sd"> buffer_size: The size of the buffers, corresponding to k in the referenced</span> |
| <span class="sd"> paper.</span> |
| <span class="sd"> num_buffers: The number of buffers, corresponding to b in the referenced</span> |
| <span class="sd"> paper.</span> |
| <span class="sd"> key: (optional) Key is a mapping of elements to a comparable key, similar</span> |
| <span class="sd"> to the key argument of Python's sorting methods.</span> |
| <span class="sd"> reverse: (optional) whether to order things smallest to largest, rather</span> |
| <span class="sd"> than largest to smallest</span> |
| <span class="sd"> """</span> |
| |
| <span class="c1"># For alternating between biasing up and down in the above even weight</span> |
| <span class="c1"># collapse operation.</span> |
| <span class="n">_offset_jitter</span> <span class="o">=</span> <span class="mi">0</span> |
| |
| <span class="c1"># The cost (in time and space) to compute quantiles to a given accuracy is a</span> |
| <span class="c1"># function of the total number of elements in the data set. If an estimate is</span> |
| <span class="c1"># not known or specified, we use this as an upper bound. If this is too low,</span> |
| <span class="c1"># errors may exceed the requested tolerance; if too high, efficiency may be</span> |
| <span class="c1"># non-optimal. The impact is logarithmic with respect to this value, so this</span> |
| <span class="c1"># default should be fine for most uses.</span> |
| <span class="n">_MAX_NUM_ELEMENTS</span> <span class="o">=</span> <span class="mf">1e9</span> |
| <span class="n">_qs</span> <span class="o">=</span> <span class="kc">None</span> <span class="c1"># Refers to the _QuantileState</span> |
| |
| <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_quantiles</span><span class="p">,</span> <span class="n">buffer_size</span><span class="p">,</span> <span class="n">num_buffers</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span> |
| <span class="k">if</span> <span class="n">key</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">:</span> <span class="p">(</span><span class="n">key</span><span class="p">(</span><span class="n">a</span><span class="p">)</span> <span class="o"><</span> <span class="n">key</span><span class="p">(</span><span class="n">b</span><span class="p">))</span> <span class="o">-</span> <span class="p">(</span><span class="n">key</span><span class="p">(</span><span class="n">a</span><span class="p">)</span> <span class="o">></span> <span class="n">key</span><span class="p">(</span><span class="n">b</span><span class="p">))</span> \ |
| <span class="k">if</span> <span class="n">reverse</span> <span class="k">else</span> <span class="p">(</span><span class="n">key</span><span class="p">(</span><span class="n">a</span><span class="p">)</span> <span class="o">></span> <span class="n">key</span><span class="p">(</span><span class="n">b</span><span class="p">))</span> <span class="o">-</span> <span class="p">(</span><span class="n">key</span><span class="p">(</span><span class="n">a</span><span class="p">)</span> <span class="o"><</span> <span class="n">key</span><span class="p">(</span><span class="n">b</span><span class="p">))</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">:</span> <span class="p">(</span><span class="n">a</span> <span class="o"><</span> <span class="n">b</span><span class="p">)</span> <span class="o">-</span> <span class="p">(</span><span class="n">a</span> <span class="o">></span> <span class="n">b</span><span class="p">)</span> <span class="k">if</span> <span class="n">reverse</span> \ |
| <span class="k">else</span> <span class="p">(</span><span class="n">a</span> <span class="o">></span> <span class="n">b</span><span class="p">)</span> <span class="o">-</span> <span class="p">(</span><span class="n">a</span> <span class="o"><</span> <span class="n">b</span><span class="p">)</span> |
| |
| <span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">=</span> <span class="n">num_quantiles</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_buffer_size</span> <span class="o">=</span> <span class="n">buffer_size</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_num_buffers</span> <span class="o">=</span> <span class="n">num_buffers</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_key</span> <span class="o">=</span> <span class="n">key</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span> <span class="o">=</span> <span class="n">reverse</span> |
| |
| <span class="nd">@classmethod</span> |
| <span class="k">def</span> <span class="nf">create</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">num_quantiles</span><span class="p">,</span> <span class="n">epsilon</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">max_num_elements</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Creates an approximate quantiles combiner with the given key and desired</span> |
| <span class="sd"> number of quantiles.</span> |
| |
| <span class="sd"> Args:</span> |
| <span class="sd"> num_quantiles: Number of quantiles to produce. It is the size of the</span> |
| <span class="sd"> final output list, including the mininum and maximum value items.</span> |
| <span class="sd"> epsilon: (optional) The default error bound is `epsilon`, which holds as</span> |
| <span class="sd"> long as the number of elements is less than `_MAX_NUM_ELEMENTS`.</span> |
| <span class="sd"> Specifically, if one considers the input as a sorted list x_1, ...,</span> |
| <span class="sd"> x_N, then the distance between each exact quantile x_c and its</span> |
| <span class="sd"> approximation x_c' is bounded by `|c - c'| < epsilon * N`. Note that</span> |
| <span class="sd"> these errors are worst-case scenarios. In practice the accuracy tends</span> |
| <span class="sd"> to be much better.</span> |
| <span class="sd"> max_num_elements: (optional) The cost (in time and space) to compute</span> |
| <span class="sd"> quantiles to a given accuracy is a function of the total number of</span> |
| <span class="sd"> elements in the data set.</span> |
| <span class="sd"> key: (optional) Key is a mapping of elements to a comparable key, similar</span> |
| <span class="sd"> to the key argument of Python's sorting methods.</span> |
| <span class="sd"> reverse: (optional) whether to order things smallest to largest, rather</span> |
| <span class="sd"> than largest to smallest</span> |
| <span class="sd"> """</span> |
| <span class="n">max_num_elements</span> <span class="o">=</span> <span class="n">max_num_elements</span> <span class="ow">or</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_MAX_NUM_ELEMENTS</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="n">epsilon</span><span class="p">:</span> |
| <span class="n">epsilon</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">num_quantiles</span> |
| <span class="n">b</span> <span class="o">=</span> <span class="mi">2</span> |
| <span class="k">while</span> <span class="p">(</span><span class="n">b</span> <span class="o">-</span> <span class="mi">2</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o"><<</span> <span class="p">(</span><span class="n">b</span> <span class="o">-</span> <span class="mi">2</span><span class="p">))</span> <span class="o"><</span> <span class="n">epsilon</span> <span class="o">*</span> <span class="n">max_num_elements</span><span class="p">:</span> |
| <span class="n">b</span> <span class="o">=</span> <span class="n">b</span> <span class="o">+</span> <span class="mi">1</span> |
| <span class="n">b</span> <span class="o">=</span> <span class="n">b</span> <span class="o">-</span> <span class="mi">1</span> |
| <span class="n">k</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">max_num_elements</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="mi">1</span> <span class="o"><<</span> <span class="p">(</span><span class="n">b</span> <span class="o">-</span> <span class="mi">1</span><span class="p">))))</span> |
| <span class="k">return</span> <span class="bp">cls</span><span class="p">(</span><span class="n">num_quantiles</span><span class="o">=</span><span class="n">num_quantiles</span><span class="p">,</span> <span class="n">buffer_size</span><span class="o">=</span><span class="n">k</span><span class="p">,</span> <span class="n">num_buffers</span><span class="o">=</span><span class="n">b</span><span class="p">,</span> |
| <span class="n">key</span><span class="o">=</span><span class="n">key</span><span class="p">,</span> <span class="n">reverse</span><span class="o">=</span><span class="n">reverse</span><span class="p">)</span> |
| |
| <span class="k">def</span> <span class="nf">_add_unbuffered</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">qs</span><span class="p">,</span> <span class="n">elem</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Add a new buffer to the unbuffered list, creating a new buffer and</span> |
| <span class="sd"> collapsing if needed.</span> |
| <span class="sd"> """</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">elem</span><span class="p">)</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">)</span> <span class="o">==</span> <span class="n">qs</span><span class="o">.</span><span class="n">buffer_size</span><span class="p">:</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">)</span> |
| <span class="n">heapq</span><span class="o">.</span><span class="n">heappush</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">,</span> |
| <span class="n">_QuantileBuffer</span><span class="p">(</span><span class="n">elements</span><span class="o">=</span><span class="n">qs</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">))</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">unbuffered_elements</span> <span class="o">=</span> <span class="p">[]</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_collapse_if_needed</span><span class="p">(</span><span class="n">qs</span><span class="p">)</span> |
| |
| <span class="k">def</span> <span class="nf">_offset</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">newWeight</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> If the weight is even, we must round up or down. Alternate between these</span> |
| <span class="sd"> two options to avoid a bias.</span> |
| <span class="sd"> """</span> |
| <span class="k">if</span> <span class="n">newWeight</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="k">return</span> <span class="p">(</span><span class="n">newWeight</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_offset_jitter</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">_offset_jitter</span> |
| <span class="k">return</span> <span class="p">(</span><span class="n">newWeight</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">_offset_jitter</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span> |
| |
| <span class="k">def</span> <span class="nf">_collapse</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">buffers</span><span class="p">):</span> |
| <span class="n">new_level</span> <span class="o">=</span> <span class="mi">0</span> |
| <span class="n">new_weight</span> <span class="o">=</span> <span class="mi">0</span> |
| <span class="k">for</span> <span class="n">buffer_elem</span> <span class="ow">in</span> <span class="n">buffers</span><span class="p">:</span> |
| <span class="c1"># As presented in the paper, there should always be at least two</span> |
| <span class="c1"># buffers of the same (minimal) level to collapse, but it is possible</span> |
| <span class="c1"># to violate this condition when combining buffers from independently</span> |
| <span class="c1"># computed shards. If they differ we take the max.</span> |
| <span class="n">new_level</span> <span class="o">=</span> <span class="nb">max</span><span class="p">([</span><span class="n">new_level</span><span class="p">,</span> <span class="n">buffer_elem</span><span class="o">.</span><span class="n">level</span> <span class="o">+</span> <span class="mi">1</span><span class="p">])</span> |
| <span class="n">new_weight</span> <span class="o">=</span> <span class="n">new_weight</span> <span class="o">+</span> <span class="n">buffer_elem</span><span class="o">.</span><span class="n">weight</span> |
| <span class="n">new_elements</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_interpolate</span><span class="p">(</span><span class="n">buffers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_buffer_size</span><span class="p">,</span> <span class="n">new_weight</span><span class="p">,</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_offset</span><span class="p">(</span><span class="n">new_weight</span><span class="p">))</span> |
| <span class="k">return</span> <span class="n">_QuantileBuffer</span><span class="p">(</span><span class="n">new_elements</span><span class="p">,</span> <span class="n">new_level</span><span class="p">,</span> <span class="n">new_weight</span><span class="p">)</span> |
| |
| <span class="k">def</span> <span class="nf">_collapse_if_needed</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">qs</span><span class="p">):</span> |
| <span class="k">while</span> <span class="nb">len</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">)</span> <span class="o">></span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_buffers</span><span class="p">:</span> |
| <span class="n">toCollapse</span> <span class="o">=</span> <span class="p">[]</span> |
| <span class="n">toCollapse</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">heapq</span><span class="o">.</span><span class="n">heappop</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">))</span> |
| <span class="n">toCollapse</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">heapq</span><span class="o">.</span><span class="n">heappop</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">))</span> |
| <span class="n">minLevel</span> <span class="o">=</span> <span class="n">toCollapse</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">level</span> |
| |
| <span class="k">while</span> <span class="nb">len</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">)</span> <span class="o">></span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">level</span> <span class="o">==</span> <span class="n">minLevel</span><span class="p">:</span> |
| <span class="n">toCollapse</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">heapq</span><span class="o">.</span><span class="n">heappop</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">))</span> |
| |
| <span class="n">heapq</span><span class="o">.</span><span class="n">heappush</span><span class="p">(</span><span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_collapse</span><span class="p">(</span><span class="n">toCollapse</span><span class="p">))</span> |
| |
| <span class="k">def</span> <span class="nf">_interpolate</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">i_buffers</span><span class="p">,</span> <span class="n">count</span><span class="p">,</span> <span class="n">step</span><span class="p">,</span> <span class="n">offset</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Emulates taking the ordered union of all elements in buffers, repeated</span> |
| <span class="sd"> according to their weight, and picking out the (k * step + offset)-th</span> |
| <span class="sd"> elements of this list for `0 <= k < count`.</span> |
| <span class="sd"> """</span> |
| |
| <span class="n">iterators</span> <span class="o">=</span> <span class="p">[]</span> |
| <span class="n">new_elements</span> <span class="o">=</span> <span class="p">[]</span> |
| <span class="n">compare_key</span> <span class="o">=</span> <span class="kc">None</span> |
| <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">:</span> |
| <span class="n">compare_key</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> |
| <span class="k">for</span> <span class="n">buffer_elem</span> <span class="ow">in</span> <span class="n">i_buffers</span><span class="p">:</span> |
| <span class="n">iterators</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">buffer_elem</span><span class="o">.</span><span class="n">sized_iterator</span><span class="p">())</span> |
| |
| <span class="c1"># Python 3 `heapq.merge` support key comparison and returns an iterator and</span> |
| <span class="c1"># does not pull the data into memory all at once. Python 2 does not</span> |
| <span class="c1"># support comparison on its `heapq.merge` api, so we use the itertools</span> |
| <span class="c1"># which takes the `key` function for comparison and creates an iterator</span> |
| <span class="c1"># from it.</span> |
| <span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">version_info</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o"><</span> <span class="mi">3</span><span class="p">:</span> |
| <span class="n">sorted_elem</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span> |
| <span class="nb">sorted</span><span class="p">(</span><span class="n">itertools</span><span class="o">.</span><span class="n">chain</span><span class="o">.</span><span class="n">from_iterable</span><span class="p">(</span><span class="n">iterators</span><span class="p">),</span> <span class="n">key</span><span class="o">=</span><span class="n">compare_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">))</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="n">sorted_elem</span> <span class="o">=</span> <span class="n">heapq</span><span class="o">.</span><span class="n">merge</span><span class="p">(</span><span class="o">*</span><span class="n">iterators</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">compare_key</span><span class="p">,</span> |
| <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">)</span> |
| |
| <span class="n">weighted_element</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">sorted_elem</span><span class="p">)</span> |
| <span class="n">current</span> <span class="o">=</span> <span class="n">weighted_element</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> |
| <span class="n">j</span> <span class="o">=</span> <span class="mi">0</span> |
| <span class="k">while</span> <span class="n">j</span> <span class="o"><</span> <span class="n">count</span><span class="p">:</span> |
| <span class="n">target</span> <span class="o">=</span> <span class="n">j</span> <span class="o">*</span> <span class="n">step</span> <span class="o">+</span> <span class="n">offset</span> |
| <span class="n">j</span> <span class="o">=</span> <span class="n">j</span> <span class="o">+</span> <span class="mi">1</span> |
| <span class="k">try</span><span class="p">:</span> |
| <span class="k">while</span> <span class="n">current</span> <span class="o"><=</span> <span class="n">target</span><span class="p">:</span> |
| <span class="n">weighted_element</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">sorted_elem</span><span class="p">)</span> |
| <span class="n">current</span> <span class="o">=</span> <span class="n">current</span> <span class="o">+</span> <span class="n">weighted_element</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> |
| <span class="k">except</span> <span class="ne">StopIteration</span><span class="p">:</span> |
| <span class="k">pass</span> |
| <span class="n">new_elements</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">weighted_element</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> |
| <span class="k">return</span> <span class="n">new_elements</span> |
| |
| <span class="k">def</span> <span class="nf">create_accumulator</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_qs</span> <span class="o">=</span> <span class="n">_QuantileState</span><span class="p">(</span><span class="n">buffer_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_buffer_size</span><span class="p">,</span> |
| <span class="n">num_buffers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_buffers</span><span class="p">,</span> |
| <span class="n">unbuffered_elements</span><span class="o">=</span><span class="p">[],</span> <span class="n">buffers</span><span class="o">=</span><span class="p">[])</span> |
| <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_qs</span> |
| |
| <span class="k">def</span> <span class="nf">add_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">quantile_state</span><span class="p">,</span> <span class="n">element</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Add a new element to the collection being summarized by quantile state.</span> |
| <span class="sd"> """</span> |
| <span class="k">if</span> <span class="n">quantile_state</span><span class="o">.</span><span class="n">is_empty</span><span class="p">():</span> |
| <span class="n">quantile_state</span><span class="o">.</span><span class="n">min_val</span> <span class="o">=</span> <span class="n">quantile_state</span><span class="o">.</span><span class="n">max_val</span> <span class="o">=</span> <span class="n">element</span> |
| <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">quantile_state</span><span class="o">.</span><span class="n">min_val</span><span class="p">)</span> <span class="o"><</span> <span class="mi">0</span><span class="p">:</span> |
| <span class="n">quantile_state</span><span class="o">.</span><span class="n">min_val</span> <span class="o">=</span> <span class="n">element</span> |
| <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">quantile_state</span><span class="o">.</span><span class="n">max_val</span><span class="p">)</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span> |
| <span class="n">quantile_state</span><span class="o">.</span><span class="n">max_val</span> <span class="o">=</span> <span class="n">element</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_add_unbuffered</span><span class="p">(</span><span class="n">quantile_state</span><span class="p">,</span> <span class="n">elem</span><span class="o">=</span><span class="n">element</span><span class="p">)</span> |
| <span class="k">return</span> <span class="n">quantile_state</span> |
| |
| <span class="k">def</span> <span class="nf">merge_accumulators</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">accumulators</span><span class="p">):</span> |
| <span class="sd">"""Merges all the accumulators (quantile state) as one."""</span> |
| <span class="n">qs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_accumulator</span><span class="p">()</span> |
| <span class="k">for</span> <span class="n">accumulator</span> <span class="ow">in</span> <span class="n">accumulators</span><span class="p">:</span> |
| <span class="k">if</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">is_empty</span><span class="p">():</span> |
| <span class="k">continue</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="n">qs</span><span class="o">.</span><span class="n">min_val</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">min_val</span><span class="p">,</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">min_val</span><span class="p">)</span> <span class="o"><</span> <span class="mi">0</span><span class="p">:</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">min_val</span> <span class="o">=</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">min_val</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="n">qs</span><span class="o">.</span><span class="n">max_val</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">_comparator</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">max_val</span><span class="p">,</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">max_val</span><span class="p">)</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span> |
| <span class="n">qs</span><span class="o">.</span><span class="n">max_val</span> <span class="o">=</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">max_val</span> |
| |
| <span class="k">for</span> <span class="n">unbuffered_element</span> <span class="ow">in</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">:</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_add_unbuffered</span><span class="p">(</span><span class="n">qs</span><span class="p">,</span> <span class="n">unbuffered_element</span><span class="p">)</span> |
| |
| <span class="n">qs</span><span class="o">.</span><span class="n">buffers</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">buffers</span><span class="p">)</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_collapse_if_needed</span><span class="p">(</span><span class="n">qs</span><span class="p">)</span> |
| <span class="k">return</span> <span class="n">qs</span> |
| |
| <span class="k">def</span> <span class="nf">extract_output</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">accumulator</span><span class="p">):</span> |
| <span class="sd">"""</span> |
| <span class="sd"> Outputs num_quantiles elements consisting of the minimum, maximum and</span> |
| <span class="sd"> num_quantiles - 2 evenly spaced intermediate elements. Returns the empty</span> |
| <span class="sd"> list if no elements have been added.</span> |
| <span class="sd"> """</span> |
| <span class="k">if</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">is_empty</span><span class="p">():</span> |
| <span class="k">return</span> <span class="p">[]</span> |
| |
| <span class="n">all_elems</span> <span class="o">=</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">buffers</span> |
| <span class="n">total_count</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">)</span> |
| <span class="k">for</span> <span class="n">buffer_elem</span> <span class="ow">in</span> <span class="n">all_elems</span><span class="p">:</span> |
| <span class="n">total_count</span> <span class="o">=</span> <span class="n">total_count</span> <span class="o">+</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">buffer_size</span> <span class="o">*</span> <span class="n">buffer_elem</span><span class="o">.</span><span class="n">weight</span> |
| |
| <span class="k">if</span> <span class="n">accumulator</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">:</span> |
| <span class="n">accumulator</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="n">key</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_key</span><span class="p">,</span> <span class="n">reverse</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_reverse</span><span class="p">)</span> |
| <span class="n">all_elems</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">_QuantileBuffer</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">unbuffered_elements</span><span class="p">))</span> |
| |
| <span class="n">step</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">*</span> <span class="n">total_count</span> <span class="o">/</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> |
| <span class="n">offset</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">*</span> <span class="n">total_count</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> |
| <span class="n">quantiles</span> <span class="o">=</span> <span class="p">[</span><span class="n">accumulator</span><span class="o">.</span><span class="n">min_val</span><span class="p">]</span> |
| <span class="n">quantiles</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span> |
| <span class="bp">self</span><span class="o">.</span><span class="n">_interpolate</span><span class="p">(</span><span class="n">all_elems</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_num_quantiles</span> <span class="o">-</span> <span class="mi">2</span><span class="p">,</span> <span class="n">step</span><span class="p">,</span> <span class="n">offset</span><span class="p">))</span> |
| <span class="n">quantiles</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">accumulator</span><span class="o">.</span><span class="n">max_val</span><span class="p">)</span> |
| <span class="k">return</span> <span class="n">quantiles</span> |
| </pre></div> |
| |
| </div> |
| <div class="articleComments"> |
| |
| </div> |
| </div> |
| <footer> |
| |
| |
| <hr/> |
| |
| <div role="contentinfo"> |
| <p> |
| © Copyright . |
| |
| </p> |
| </div> |
| Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. |
| |
| </footer> |
| |
| </div> |
| </div> |
| |
| </section> |
| |
| </div> |
| |
| |
| |
| |
| |
| <script type="text/javascript"> |
| var DOCUMENTATION_OPTIONS = { |
| URL_ROOT:'../../../', |
| VERSION:'', |
| COLLAPSE_INDEX:false, |
| FILE_SUFFIX:'.html', |
| HAS_SOURCE: true, |
| SOURCELINK_SUFFIX: '.txt' |
| }; |
| </script> |
| <script type="text/javascript" src="../../../_static/jquery.js"></script> |
| <script type="text/javascript" src="../../../_static/underscore.js"></script> |
| <script type="text/javascript" src="../../../_static/doctools.js"></script> |
| |
| |
| |
| |
| |
| <script type="text/javascript" src="../../../_static/js/theme.js"></script> |
| |
| |
| |
| |
| <script type="text/javascript"> |
| jQuery(function () { |
| SphinxRtdTheme.StickyNav.enable(); |
| }); |
| </script> |
| |
| |
| </body> |
| </html> |
