| |
| <!DOCTYPE html> |
| |
| <html> |
| <head> |
| <meta charset="utf-8" /> |
| <title>pyspark.ml.functions — PySpark 3.5.5 documentation</title> |
| |
| <link href="../../../_static/styles/theme.css?digest=1999514e3f237ded88cf" rel="stylesheet"> |
| <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=1999514e3f237ded88cf" rel="stylesheet"> |
| |
| |
| <link rel="stylesheet" |
| href="../../../_static/vendor/fontawesome/5.13.0/css/all.min.css"> |
| <link rel="preload" as="font" type="font/woff2" crossorigin |
| href="../../../_static/vendor/fontawesome/5.13.0/webfonts/fa-solid-900.woff2"> |
| <link rel="preload" as="font" type="font/woff2" crossorigin |
| href="../../../_static/vendor/fontawesome/5.13.0/webfonts/fa-brands-400.woff2"> |
| |
| |
| |
| |
| |
| <link rel="stylesheet" href="../../../_static/styles/pydata-sphinx-theme.css" type="text/css" /> |
| <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> |
| <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css" /> |
| <link rel="stylesheet" type="text/css" href="../../../_static/css/pyspark.css" /> |
| |
| <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=1999514e3f237ded88cf"> |
| |
| <script id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script> |
| <script src="../../../_static/jquery.js"></script> |
| <script src="../../../_static/underscore.js"></script> |
| <script src="../../../_static/doctools.js"></script> |
| <script src="../../../_static/language_data.js"></script> |
| <script src="../../../_static/clipboard.min.js"></script> |
| <script src="../../../_static/copybutton.js"></script> |
| <script crossorigin="anonymous" integrity="sha256-Ae2Vz/4ePdIu6ZyI/5ZGsYnb+m0JlOmKPjt6XZ9JJkA=" src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.4/require.min.js"></script> |
| <script async="async" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/latest.js?config=TeX-AMS-MML_HTMLorMML"></script> |
| <script type="text/x-mathjax-config">MathJax.Hub.Config({"tex2jax": {"inlineMath": [["$", "$"], ["\\(", "\\)"]], "processEscapes": true, "ignoreClass": "document", "processClass": "math|output_area"}})</script> |
| <link rel="canonical" href="https://spark.apache.org/docs/latest/api/python/_modules/pyspark/ml/functions.html" /> |
| <link rel="search" title="Search" href="../../../search.html" /> |
| <meta name="viewport" content="width=device-width, initial-scale=1" /> |
| <meta name="docsearch:language" content="None"> |
| |
| |
| <!-- Google Analytics --> |
| |
| </head> |
| <body data-spy="scroll" data-target="#bd-toc-nav" data-offset="80"> |
| |
| <div class="container-fluid" id="banner"></div> |
| |
| |
| <nav class="navbar navbar-light navbar-expand-lg bg-light fixed-top bd-navbar" id="navbar-main"><div class="container-xl"> |
| |
| <div id="navbar-start"> |
| |
| |
| |
| <a class="navbar-brand" href="../../../index.html"> |
| <img src="../../../_static/spark-logo-reverse.png" class="logo" alt="logo"> |
| </a> |
| |
| |
| |
| </div> |
| |
| <button class="navbar-toggler" type="button" data-toggle="collapse" data-target="#navbar-collapsible" aria-controls="navbar-collapsible" aria-expanded="false" aria-label="Toggle navigation"> |
| <span class="navbar-toggler-icon"></span> |
| </button> |
| |
| |
| <div id="navbar-collapsible" class="col-lg-9 collapse navbar-collapse"> |
| <div id="navbar-center" class="mr-auto"> |
| |
| <div class="navbar-center-item"> |
| <ul id="navbar-main-elements" class="navbar-nav"> |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../index.html"> |
| Overview |
| </a> |
| </li> |
| |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../getting_started/index.html"> |
| Getting Started |
| </a> |
| </li> |
| |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../user_guide/index.html"> |
| User Guides |
| </a> |
| </li> |
| |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../reference/index.html"> |
| API Reference |
| </a> |
| </li> |
| |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../development/index.html"> |
| Development |
| </a> |
| </li> |
| |
| <li class="toctree-l1 nav-item"> |
| <a class="reference internal nav-link" href="../../../migration_guide/index.html"> |
| Migration Guides |
| </a> |
| </li> |
| |
| |
| </ul> |
| </div> |
| |
| </div> |
| |
| <div id="navbar-end"> |
| |
| <div class="navbar-end-item"> |
| <!-- |
| Licensed to the Apache Software Foundation (ASF) under one or more |
| contributor license agreements. See the NOTICE file distributed with |
| this work for additional information regarding copyright ownership. |
| The ASF licenses this file to You under the Apache License, Version 2.0 |
| (the "License"); you may not use this file except in compliance with |
| the License. You may obtain a copy of the License at |
| |
| http://www.apache.org/licenses/LICENSE-2.0 |
| |
| Unless required by applicable law or agreed to in writing, software |
| distributed under the License is distributed on an "AS IS" BASIS, |
| WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| See the License for the specific language governing permissions and |
| limitations under the License. |
| --> |
| |
| <div id="version-button" class="dropdown"> |
| <button type="button" class="btn btn-secondary btn-sm navbar-btn dropdown-toggle" id="version_switcher_button" data-toggle="dropdown"> |
| 3.5.5 |
| <span class="caret"></span> |
| </button> |
| <div id="version_switcher" class="dropdown-menu list-group-flush py-0" aria-labelledby="version_switcher_button"> |
| <!-- dropdown will be populated by javascript on page load --> |
| </div> |
| </div> |
| |
| <script type="text/javascript"> |
| // Function to construct the target URL from the JSON components |
| function buildURL(entry) { |
| var template = "https://spark.apache.org/docs/{version}/api/python/index.html"; // supplied by jinja |
| template = template.replace("{version}", entry.version); |
| return template; |
| } |
| |
| // Function to check if corresponding page path exists in other version of docs |
| // and, if so, go there instead of the homepage of the other docs version |
| function checkPageExistsAndRedirect(event) { |
| const currentFilePath = "_modules/pyspark/ml/functions.html", |
| otherDocsHomepage = event.target.getAttribute("href"); |
| let tryUrl = `${otherDocsHomepage}${currentFilePath}`; |
| $.ajax({ |
| type: 'HEAD', |
| url: tryUrl, |
| // if the page exists, go there |
| success: function() { |
| location.href = tryUrl; |
| } |
| }).fail(function() { |
| location.href = otherDocsHomepage; |
| }); |
| return false; |
| } |
| |
| // Function to populate the version switcher |
| (function () { |
| // get JSON config |
| $.getJSON("https://spark.apache.org/static/versions.json", function(data, textStatus, jqXHR) { |
| // create the nodes first (before AJAX calls) to ensure the order is |
| // correct (for now, links will go to doc version homepage) |
| $.each(data, function(index, entry) { |
| // if no custom name specified (e.g., "latest"), use version string |
| if (!("name" in entry)) { |
| entry.name = entry.version; |
| } |
| // construct the appropriate URL, and add it to the dropdown |
| entry.url = buildURL(entry); |
| const node = document.createElement("a"); |
| node.setAttribute("class", "list-group-item list-group-item-action py-1"); |
| node.setAttribute("href", `${entry.url}`); |
| node.textContent = `${entry.name}`; |
| node.onclick = checkPageExistsAndRedirect; |
| $("#version_switcher").append(node); |
| }); |
| }); |
| })(); |
| </script> |
| </div> |
| |
| </div> |
| </div> |
| </div> |
| </nav> |
| |
| |
| <div class="container-xl"> |
| <div class="row"> |
| |
| |
| <!-- Only show if we have sidebars configured, else just a small margin --> |
| <div class="col-12 col-md-3 bd-sidebar"> |
| <div class="sidebar-start-items"><form class="bd-search d-flex align-items-center" action="../../../search.html" method="get"> |
| <i class="icon fas fa-search"></i> |
| <input type="search" class="form-control" name="q" id="search-input" placeholder="Search the docs ..." aria-label="Search the docs ..." autocomplete="off" > |
| </form><nav class="bd-links" id="bd-docs-nav" aria-label="Main navigation"> |
| <div class="bd-toc-item active"> |
| |
| </div> |
| </nav> |
| </div> |
| <div class="sidebar-end-items"> |
| </div> |
| </div> |
| |
| |
| |
| |
| <div class="d-none d-xl-block col-xl-2 bd-toc"> |
| |
| </div> |
| |
| |
| |
| |
| |
| |
| <main class="col-12 col-md-9 col-xl-7 py-md-5 pl-md-5 pr-md-4 bd-content" role="main"> |
| |
| <div> |
| |
| <h1>Source code for pyspark.ml.functions</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="kn">from</span><span class="w"> </span><span class="nn">__future__</span><span class="w"> </span><span class="kn">import</span> <span class="n">annotations</span> |
| |
| <span class="kn">import</span><span class="w"> </span><span class="nn">inspect</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">pandas</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">pd</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">uuid</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark</span><span class="w"> </span><span class="kn">import</span> <span class="n">SparkContext</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.sql.functions</span><span class="w"> </span><span class="kn">import</span> <span class="n">pandas_udf</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.sql.column</span><span class="w"> </span><span class="kn">import</span> <span class="n">Column</span><span class="p">,</span> <span class="n">_to_java_column</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.sql.types</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span> |
| <span class="n">ArrayType</span><span class="p">,</span> |
| <span class="n">ByteType</span><span class="p">,</span> |
| <span class="n">DataType</span><span class="p">,</span> |
| <span class="n">DoubleType</span><span class="p">,</span> |
| <span class="n">FloatType</span><span class="p">,</span> |
| <span class="n">IntegerType</span><span class="p">,</span> |
| <span class="n">LongType</span><span class="p">,</span> |
| <span class="n">ShortType</span><span class="p">,</span> |
| <span class="n">StringType</span><span class="p">,</span> |
| <span class="n">StructType</span><span class="p">,</span> |
| <span class="p">)</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.ml.util</span><span class="w"> </span><span class="kn">import</span> <span class="n">try_remote_functions</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Iterator</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">,</span> <span class="n">TYPE_CHECKING</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Optional</span> |
| |
| <span class="k">if</span> <span class="n">TYPE_CHECKING</span><span class="p">:</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.sql._typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">UserDefinedFunctionLike</span> |
| |
| <span class="n">supported_scalar_types</span> <span class="o">=</span> <span class="p">(</span> |
| <span class="n">ByteType</span><span class="p">,</span> |
| <span class="n">ShortType</span><span class="p">,</span> |
| <span class="n">IntegerType</span><span class="p">,</span> |
| <span class="n">LongType</span><span class="p">,</span> |
| <span class="n">FloatType</span><span class="p">,</span> |
| <span class="n">DoubleType</span><span class="p">,</span> |
| <span class="n">StringType</span><span class="p">,</span> |
| <span class="p">)</span> |
| |
| <span class="c1"># Callable type for end user predict functions that take a variable number of ndarrays as</span> |
| <span class="c1"># input and returns one of the following as output:</span> |
| <span class="c1"># - single ndarray (single output)</span> |
| <span class="c1"># - dictionary of named ndarrays (multiple outputs represented in columnar form)</span> |
| <span class="c1"># - list of dictionaries of named ndarrays (multiple outputs represented in row form)</span> |
| <span class="n">PredictBatchFunction</span> <span class="o">=</span> <span class="n">Callable</span><span class="p">[</span> |
| <span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">Union</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">dtype</span><span class="p">]]]</span> |
| <span class="p">]</span> |
| |
| |
| <div class="viewcode-block" id="vector_to_array"><a class="viewcode-back" href="../../../reference/api/pyspark.ml.functions.vector_to_array.html#pyspark.ml.functions.vector_to_array">[docs]</a><span class="nd">@try_remote_functions</span> |
| <span class="k">def</span><span class="w"> </span><span class="nf">vector_to_array</span><span class="p">(</span><span class="n">col</span><span class="p">:</span> <span class="n">Column</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">"float64"</span><span class="p">)</span> <span class="o">-></span> <span class="n">Column</span><span class="p">:</span> |
| <span class="w"> </span><span class="sd">"""</span> |
| <span class="sd"> Converts a column of MLlib sparse/dense vectors into a column of dense arrays.</span> |
| |
| <span class="sd"> .. versionadded:: 3.0.0</span> |
| |
| <span class="sd"> .. versionchanged:: 3.5.0</span> |
| <span class="sd"> Supports Spark Connect.</span> |
| |
| <span class="sd"> Parameters</span> |
| <span class="sd"> ----------</span> |
| <span class="sd"> col : :py:class:`pyspark.sql.Column` or str</span> |
| <span class="sd"> Input column</span> |
| <span class="sd"> dtype : str, optional</span> |
| <span class="sd"> The data type of the output array. Valid values: "float64" or "float32".</span> |
| |
| <span class="sd"> Returns</span> |
| <span class="sd"> -------</span> |
| <span class="sd"> :py:class:`pyspark.sql.Column`</span> |
| <span class="sd"> The converted column of dense arrays.</span> |
| |
| <span class="sd"> Examples</span> |
| <span class="sd"> --------</span> |
| <span class="sd"> >>> from pyspark.ml.linalg import Vectors</span> |
| <span class="sd"> >>> from pyspark.ml.functions import vector_to_array</span> |
| <span class="sd"> >>> from pyspark.mllib.linalg import Vectors as OldVectors</span> |
| <span class="sd"> >>> df = spark.createDataFrame([</span> |
| <span class="sd"> ... (Vectors.dense(1.0, 2.0, 3.0), OldVectors.dense(10.0, 20.0, 30.0)),</span> |
| <span class="sd"> ... (Vectors.sparse(3, [(0, 2.0), (2, 3.0)]),</span> |
| <span class="sd"> ... OldVectors.sparse(3, [(0, 20.0), (2, 30.0)]))],</span> |
| <span class="sd"> ... ["vec", "oldVec"])</span> |
| <span class="sd"> >>> df1 = df.select(vector_to_array("vec").alias("vec"),</span> |
| <span class="sd"> ... vector_to_array("oldVec").alias("oldVec"))</span> |
| <span class="sd"> >>> df1.collect()</span> |
| <span class="sd"> [Row(vec=[1.0, 2.0, 3.0], oldVec=[10.0, 20.0, 30.0]),</span> |
| <span class="sd"> Row(vec=[2.0, 0.0, 3.0], oldVec=[20.0, 0.0, 30.0])]</span> |
| <span class="sd"> >>> df2 = df.select(vector_to_array("vec", "float32").alias("vec"),</span> |
| <span class="sd"> ... vector_to_array("oldVec", "float32").alias("oldVec"))</span> |
| <span class="sd"> >>> df2.collect()</span> |
| <span class="sd"> [Row(vec=[1.0, 2.0, 3.0], oldVec=[10.0, 20.0, 30.0]),</span> |
| <span class="sd"> Row(vec=[2.0, 0.0, 3.0], oldVec=[20.0, 0.0, 30.0])]</span> |
| <span class="sd"> >>> df1.schema.fields</span> |
| <span class="sd"> [StructField('vec', ArrayType(DoubleType(), False), False),</span> |
| <span class="sd"> StructField('oldVec', ArrayType(DoubleType(), False), False)]</span> |
| <span class="sd"> >>> df2.schema.fields</span> |
| <span class="sd"> [StructField('vec', ArrayType(FloatType(), False), False),</span> |
| <span class="sd"> StructField('oldVec', ArrayType(FloatType(), False), False)]</span> |
| <span class="sd"> """</span> |
| <span class="n">sc</span> <span class="o">=</span> <span class="n">SparkContext</span><span class="o">.</span><span class="n">_active_spark_context</span> |
| <span class="k">assert</span> <span class="n">sc</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">sc</span><span class="o">.</span><span class="n">_jvm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> |
| <span class="k">return</span> <span class="n">Column</span><span class="p">(</span> |
| <span class="n">sc</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">org</span><span class="o">.</span><span class="n">apache</span><span class="o">.</span><span class="n">spark</span><span class="o">.</span><span class="n">ml</span><span class="o">.</span><span class="n">functions</span><span class="o">.</span><span class="n">vector_to_array</span><span class="p">(</span><span class="n">_to_java_column</span><span class="p">(</span><span class="n">col</span><span class="p">),</span> <span class="n">dtype</span><span class="p">)</span> |
| <span class="p">)</span></div> |
| |
| |
| <div class="viewcode-block" id="array_to_vector"><a class="viewcode-back" href="../../../reference/api/pyspark.ml.functions.array_to_vector.html#pyspark.ml.functions.array_to_vector">[docs]</a><span class="nd">@try_remote_functions</span> |
| <span class="k">def</span><span class="w"> </span><span class="nf">array_to_vector</span><span class="p">(</span><span class="n">col</span><span class="p">:</span> <span class="n">Column</span><span class="p">)</span> <span class="o">-></span> <span class="n">Column</span><span class="p">:</span> |
| <span class="w"> </span><span class="sd">"""</span> |
| <span class="sd"> Converts a column of array of numeric type into a column of pyspark.ml.linalg.DenseVector</span> |
| <span class="sd"> instances</span> |
| |
| <span class="sd"> .. versionadded:: 3.1.0</span> |
| |
| <span class="sd"> .. versionchanged:: 3.5.0</span> |
| <span class="sd"> Supports Spark Connect.</span> |
| |
| <span class="sd"> Parameters</span> |
| <span class="sd"> ----------</span> |
| <span class="sd"> col : :py:class:`pyspark.sql.Column` or str</span> |
| <span class="sd"> Input column</span> |
| |
| <span class="sd"> Returns</span> |
| <span class="sd"> -------</span> |
| <span class="sd"> :py:class:`pyspark.sql.Column`</span> |
| <span class="sd"> The converted column of dense vectors.</span> |
| |
| <span class="sd"> Examples</span> |
| <span class="sd"> --------</span> |
| <span class="sd"> >>> from pyspark.ml.functions import array_to_vector</span> |
| <span class="sd"> >>> df1 = spark.createDataFrame([([1.5, 2.5],),], schema='v1 array<double>')</span> |
| <span class="sd"> >>> df1.select(array_to_vector('v1').alias('vec1')).collect()</span> |
| <span class="sd"> [Row(vec1=DenseVector([1.5, 2.5]))]</span> |
| <span class="sd"> >>> df2 = spark.createDataFrame([([1.5, 3.5],),], schema='v1 array<float>')</span> |
| <span class="sd"> >>> df2.select(array_to_vector('v1').alias('vec1')).collect()</span> |
| <span class="sd"> [Row(vec1=DenseVector([1.5, 3.5]))]</span> |
| <span class="sd"> >>> df3 = spark.createDataFrame([([1, 3],),], schema='v1 array<int>')</span> |
| <span class="sd"> >>> df3.select(array_to_vector('v1').alias('vec1')).collect()</span> |
| <span class="sd"> [Row(vec1=DenseVector([1.0, 3.0]))]</span> |
| <span class="sd"> """</span> |
| <span class="n">sc</span> <span class="o">=</span> <span class="n">SparkContext</span><span class="o">.</span><span class="n">_active_spark_context</span> |
| <span class="k">assert</span> <span class="n">sc</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">sc</span><span class="o">.</span><span class="n">_jvm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> |
| <span class="k">return</span> <span class="n">Column</span><span class="p">(</span><span class="n">sc</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">org</span><span class="o">.</span><span class="n">apache</span><span class="o">.</span><span class="n">spark</span><span class="o">.</span><span class="n">ml</span><span class="o">.</span><span class="n">functions</span><span class="o">.</span><span class="n">array_to_vector</span><span class="p">(</span><span class="n">_to_java_column</span><span class="p">(</span><span class="n">col</span><span class="p">)))</span></div> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_batched</span><span class="p">(</span> |
| <span class="n">data</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">]],</span> <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span> |
| <span class="p">)</span> <span class="o">-></span> <span class="n">Iterator</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">]:</span> |
| <span class="w"> </span><span class="sd">"""Generator that splits a pandas dataframe/series into batches."""</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">):</span> |
| <span class="n">df</span> <span class="o">=</span> <span class="n">data</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">):</span> |
| <span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">concat</span><span class="p">((</span><span class="n">data</span><span class="p">,),</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> <span class="c1"># isinstance(data, Tuple[pd.Series]):</span> |
| <span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">concat</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span> |
| |
| <span class="n">index</span> <span class="o">=</span> <span class="mi">0</span> |
| <span class="n">data_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">df</span><span class="p">)</span> |
| <span class="k">while</span> <span class="n">index</span> <span class="o"><</span> <span class="n">data_size</span><span class="p">:</span> |
| <span class="k">yield</span> <span class="n">df</span><span class="o">.</span><span class="n">iloc</span><span class="p">[</span><span class="n">index</span> <span class="p">:</span> <span class="n">index</span> <span class="o">+</span> <span class="n">batch_size</span><span class="p">]</span> |
| <span class="n">index</span> <span class="o">+=</span> <span class="n">batch_size</span> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_is_tensor_col</span><span class="p">(</span><span class="n">data</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">])</span> <span class="o">-></span> <span class="nb">bool</span><span class="p">:</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">):</span> |
| <span class="k">return</span> <span class="n">data</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">object_</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="o">.</span><span class="n">iloc</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">):</span> |
| <span class="k">return</span> <span class="nb">any</span><span class="p">(</span><span class="n">data</span><span class="o">.</span><span class="n">dtypes</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">object_</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">any</span><span class="p">(</span> |
| <span class="p">[</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">d</span><span class="p">,</span> <span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <span class="n">data</span><span class="o">.</span><span class="n">iloc</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span> |
| <span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Unexpected data type: </span><span class="si">{}</span><span class="s2">, expected pd.Series or pd.DataFrame."</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">data</span><span class="p">))</span> |
| <span class="p">)</span> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_has_tensor_cols</span><span class="p">(</span><span class="n">data</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">]])</span> <span class="o">-></span> <span class="nb">bool</span><span class="p">:</span> |
| <span class="w"> </span><span class="sd">"""Check if input Series/DataFrame/Tuple contains any tensor-valued columns."""</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="p">(</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">)):</span> |
| <span class="k">return</span> <span class="n">_is_tensor_col</span><span class="p">(</span><span class="n">data</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> <span class="c1"># isinstance(data, Tuple):</span> |
| <span class="k">return</span> <span class="nb">any</span><span class="p">(</span><span class="n">_is_tensor_col</span><span class="p">(</span><span class="n">elem</span><span class="p">)</span> <span class="k">for</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">data</span><span class="p">)</span> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_validate_and_transform_multiple_inputs</span><span class="p">(</span> |
| <span class="n">batch</span><span class="p">:</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">,</span> <span class="n">input_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span> <span class="n">num_input_cols</span><span class="p">:</span> <span class="nb">int</span> |
| <span class="p">)</span> <span class="o">-></span> <span class="n">List</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]:</span> |
| <span class="n">multi_inputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">batch</span><span class="p">[</span><span class="n">col</span><span class="p">]</span><span class="o">.</span><span class="n">to_numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">col</span> <span class="ow">in</span> <span class="n">batch</span><span class="o">.</span><span class="n">columns</span><span class="p">]</span> |
| <span class="k">if</span> <span class="n">input_shapes</span><span class="p">:</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">input_shapes</span><span class="p">)</span> <span class="o">==</span> <span class="n">num_input_cols</span><span class="p">:</span> |
| <span class="n">multi_inputs</span> <span class="o">=</span> <span class="p">[</span> |
| <span class="n">np</span><span class="o">.</span><span class="n">vstack</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">([</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">input_shapes</span><span class="p">[</span><span class="n">i</span><span class="p">])</span> <span class="c1"># type: ignore</span> |
| <span class="k">if</span> <span class="n">input_shapes</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> |
| <span class="k">else</span> <span class="n">v</span> |
| <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">multi_inputs</span><span class="p">)</span> |
| <span class="p">]</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="nb">all</span><span class="p">([</span><span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="o">==</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">multi_inputs</span><span class="p">]):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Input data does not match expected shape."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"input_tensor_shapes must match columns"</span><span class="p">)</span> |
| |
| <span class="k">return</span> <span class="n">multi_inputs</span> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_validate_and_transform_single_input</span><span class="p">(</span> |
| <span class="n">batch</span><span class="p">:</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">,</span> |
| <span class="n">input_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">|</span> <span class="kc">None</span><span class="p">],</span> |
| <span class="n">has_tensors</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> |
| <span class="n">has_tuple</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> |
| <span class="p">)</span> <span class="o">-></span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span> |
| <span class="c1"># multiple input columns for single expected input</span> |
| <span class="k">if</span> <span class="n">has_tensors</span><span class="p">:</span> |
| <span class="c1"># tensor columns</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">columns</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="c1"># one tensor column and one expected input, vstack rows</span> |
| <span class="n">single_input</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">vstack</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">iloc</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">])</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Multiple input columns found, but model expected a single "</span> |
| <span class="s2">"input, use `array` to combine columns into tensors."</span> |
| <span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="c1"># scalar columns</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">columns</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="c1"># single scalar column, remove extra dim</span> |
| <span class="n">np_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="o">.</span><span class="n">to_numpy</span><span class="p">()</span> |
| <span class="n">single_input</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">np_batch</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">np_batch</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">></span> <span class="mi">1</span> <span class="k">else</span> <span class="n">np_batch</span> |
| <span class="k">if</span> <span class="n">input_shapes</span> <span class="ow">and</span> <span class="n">input_shapes</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">[],</span> <span class="p">[</span><span class="mi">1</span><span class="p">]]:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Invalid input_tensor_shape for scalar column."</span><span class="p">)</span> |
| <span class="k">elif</span> <span class="ow">not</span> <span class="n">has_tuple</span><span class="p">:</span> |
| <span class="c1"># columns grouped via `array`, convert to single tensor</span> |
| <span class="n">single_input</span> <span class="o">=</span> <span class="n">batch</span><span class="o">.</span><span class="n">to_numpy</span><span class="p">()</span> |
| <span class="k">if</span> <span class="n">input_shapes</span> <span class="ow">and</span> <span class="n">input_shapes</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">columns</span><span class="p">)]:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Input data does not match expected shape."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Multiple input columns found, but model expected a single "</span> |
| <span class="s2">"input, use `array` to combine columns into tensors."</span> |
| <span class="p">)</span> |
| |
| <span class="c1"># if input_tensor_shapes provided, try to reshape input</span> |
| <span class="k">if</span> <span class="n">input_shapes</span><span class="p">:</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">input_shapes</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="n">single_input</span> <span class="o">=</span> <span class="n">single_input</span><span class="o">.</span><span class="n">reshape</span><span class="p">([</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">input_shapes</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="c1"># type: ignore</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">single_input</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Input data does not match expected shape."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Multiple input_tensor_shapes found, but model expected one input"</span><span class="p">)</span> |
| |
| <span class="k">return</span> <span class="n">single_input</span> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_validate_and_transform_prediction_result</span><span class="p">(</span> |
| <span class="n">preds</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span> <span class="o">|</span> <span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]</span> <span class="o">|</span> <span class="n">List</span><span class="p">[</span><span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]],</span> |
| <span class="n">num_input_rows</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> |
| <span class="n">return_type</span><span class="p">:</span> <span class="n">DataType</span><span class="p">,</span> |
| <span class="p">)</span> <span class="o">-></span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span> <span class="o">|</span> <span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">:</span> |
| <span class="w"> </span><span class="sd">"""Validate numpy-based model predictions against the expected pandas_udf return_type and</span> |
| <span class="sd"> transforms the predictions into an equivalent pandas DataFrame or Series."""</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">return_type</span><span class="p">,</span> <span class="n">StructType</span><span class="p">):</span> |
| <span class="n">struct_rtype</span><span class="p">:</span> <span class="n">StructType</span> <span class="o">=</span> <span class="n">return_type</span> |
| <span class="n">fieldNames</span> <span class="o">=</span> <span class="n">struct_rtype</span><span class="o">.</span><span class="n">names</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">preds</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span> |
| <span class="c1"># dictionary of columns</span> |
| <span class="n">predNames</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">preds</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span> |
| <span class="k">for</span> <span class="n">field</span> <span class="ow">in</span> <span class="n">struct_rtype</span><span class="o">.</span><span class="n">fields</span><span class="p">:</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">field</span><span class="o">.</span><span class="n">dataType</span><span class="p">,</span> <span class="n">ArrayType</span><span class="p">):</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span> |
| <span class="n">preds</span><span class="p">[</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">])</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Prediction results for ArrayType must be two-dimensional."</span> |
| <span class="p">)</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">field</span><span class="o">.</span><span class="n">dataType</span><span class="p">,</span> <span class="n">supported_scalar_types</span><span class="p">):</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Prediction results for scalar types must be one-dimensional."</span> |
| <span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Unsupported field type in return struct type."</span><span class="p">)</span> |
| |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">])</span> <span class="o">!=</span> <span class="n">num_input_rows</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results must have same length as input data"</span><span class="p">)</span> |
| |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">preds</span><span class="p">,</span> <span class="nb">list</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">dict</span><span class="p">):</span> |
| <span class="c1"># rows of dictionaries</span> |
| <span class="n">predNames</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">)</span> <span class="o">!=</span> <span class="n">num_input_rows</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results must have same length as input data."</span><span class="p">)</span> |
| <span class="k">for</span> <span class="n">field</span> <span class="ow">in</span> <span class="n">struct_rtype</span><span class="o">.</span><span class="n">fields</span><span class="p">:</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">field</span><span class="o">.</span><span class="n">dataType</span><span class="p">,</span> <span class="n">ArrayType</span><span class="p">):</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Prediction results for ArrayType must be one-dimensional."</span> |
| <span class="p">)</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">field</span><span class="o">.</span><span class="n">dataType</span><span class="p">,</span> <span class="n">supported_scalar_types</span><span class="p">):</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="n">np</span><span class="o">.</span><span class="n">isscalar</span><span class="p">(</span><span class="n">preds</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="p">]):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Invalid scalar prediction result."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Unsupported field type in return struct type."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Prediction results for StructType must be a dictionary or "</span> |
| <span class="s2">"a list of dictionary, got: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">preds</span><span class="p">))</span> |
| <span class="p">)</span> |
| |
| <span class="c1"># check column names</span> |
| <span class="k">if</span> <span class="nb">set</span><span class="p">(</span><span class="n">predNames</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">set</span><span class="p">(</span><span class="n">fieldNames</span><span class="p">):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span> |
| <span class="s2">"Prediction result columns did not match expected return_type "</span> |
| <span class="s2">"columns: expected </span><span class="si">{}</span><span class="s2">, got: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">fieldNames</span><span class="p">,</span> <span class="n">predNames</span><span class="p">)</span> |
| <span class="p">)</span> |
| |
| <span class="k">return</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">(</span><span class="n">preds</span><span class="p">)</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">return_type</span><span class="p">,</span> <span class="n">ArrayType</span><span class="p">):</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">preds</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="p">)</span> <span class="o">!=</span> <span class="n">num_input_rows</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results must have same length as input data."</span><span class="p">)</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results for ArrayType must be two-dimensional."</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results for ArrayType must be an ndarray."</span><span class="p">)</span> |
| |
| <span class="k">return</span> <span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">preds</span><span class="p">))</span> |
| <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">return_type</span><span class="p">,</span> <span class="n">supported_scalar_types</span><span class="p">):</span> |
| <span class="n">preds_array</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span> <span class="o">=</span> <span class="n">preds</span> <span class="c1"># type: ignore</span> |
| <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds_array</span><span class="p">)</span> <span class="o">!=</span> <span class="n">num_input_rows</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Prediction results must have same length as input data."</span><span class="p">)</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span> |
| <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">preds_array</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">preds_array</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">)</span> |
| <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds_array</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span> |
| <span class="p">):</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Invalid shape for scalar prediction result."</span><span class="p">)</span> |
| |
| <span class="n">output</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">preds_array</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preds_array</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">></span> <span class="mi">1</span> <span class="k">else</span> <span class="n">preds_array</span> |
| <span class="k">return</span> <span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">(</span><span class="n">output</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Unsupported return type"</span><span class="p">)</span> |
| |
| |
| <div class="viewcode-block" id="predict_batch_udf"><a class="viewcode-back" href="../../../reference/api/pyspark.ml.functions.predict_batch_udf.html#pyspark.ml.functions.predict_batch_udf">[docs]</a><span class="k">def</span><span class="w"> </span><span class="nf">predict_batch_udf</span><span class="p">(</span> |
| <span class="n">make_predict_fn</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span> |
| <span class="p">[],</span> |
| <span class="n">PredictBatchFunction</span><span class="p">,</span> |
| <span class="p">],</span> |
| <span class="o">*</span><span class="p">,</span> |
| <span class="n">return_type</span><span class="p">:</span> <span class="n">DataType</span><span class="p">,</span> |
| <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> |
| <span class="n">input_tensor_shapes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span> <span class="n">Mapping</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> |
| <span class="p">)</span> <span class="o">-></span> <span class="n">UserDefinedFunctionLike</span><span class="p">:</span> |
| <span class="w"> </span><span class="sd">"""Given a function which loads a model and returns a `predict` function for inference over a</span> |
| <span class="sd"> batch of numpy inputs, returns a Pandas UDF wrapper for inference over a Spark DataFrame.</span> |
| |
| <span class="sd"> The returned Pandas UDF does the following on each DataFrame partition:</span> |
| |
| <span class="sd"> * calls the `make_predict_fn` to load the model and cache its `predict` function.</span> |
| <span class="sd"> * batches the input records as numpy arrays and invokes `predict` on each batch.</span> |
| |
| <span class="sd"> Note: this assumes that the `make_predict_fn` encapsulates all of the necessary dependencies for</span> |
| <span class="sd"> running the model, or the Spark executor environment already satisfies all runtime requirements.</span> |
| |
| <span class="sd"> For the conversion of the Spark DataFrame to numpy arrays, there is a one-to-one mapping between</span> |
| <span class="sd"> the input arguments of the `predict` function (returned by the `make_predict_fn`) and the input</span> |
| <span class="sd"> columns sent to the Pandas UDF (returned by the `predict_batch_udf`) at runtime. Each input</span> |
| <span class="sd"> column will be converted as follows:</span> |
| |
| <span class="sd"> * scalar column -> 1-dim np.ndarray</span> |
| <span class="sd"> * tensor column + tensor shape -> N-dim np.ndarray</span> |
| |
| <span class="sd"> Note that any tensor columns in the Spark DataFrame must be represented as a flattened</span> |
| <span class="sd"> one-dimensional array, and multiple scalar columns can be combined into a single tensor column</span> |
| <span class="sd"> using the standard :py:func:`pyspark.sql.functions.array()` function.</span> |
| |
| <span class="sd"> .. versionadded:: 3.4.0</span> |
| |
| <span class="sd"> Parameters</span> |
| <span class="sd"> ----------</span> |
| <span class="sd"> make_predict_fn : callable</span> |
| <span class="sd"> Function which is responsible for loading a model and returning a</span> |
| <span class="sd"> :py:class:`PredictBatchFunction` which takes one or more numpy arrays as input and returns</span> |
| <span class="sd"> one of the following:</span> |
| |
| <span class="sd"> * a numpy array (for a single output)</span> |
| <span class="sd"> * a dictionary of named numpy arrays (for multiple outputs)</span> |
| <span class="sd"> * a row-oriented list of dictionaries (for multiple outputs).</span> |
| |
| <span class="sd"> For a dictionary of named numpy arrays, the arrays can only be one or two dimensional, since</span> |
| <span class="sd"> higher dimensional arrays are not supported. For a row-oriented list of dictionaries, each</span> |
| <span class="sd"> element in the dictionary must be either a scalar or one-dimensional array.</span> |
| <span class="sd"> return_type : :py:class:`pyspark.sql.types.DataType` or str.</span> |
| <span class="sd"> Spark SQL datatype for the expected output:</span> |
| |
| <span class="sd"> * Scalar (e.g. IntegerType, FloatType) --> 1-dim numpy array.</span> |
| <span class="sd"> * ArrayType --> 2-dim numpy array.</span> |
| <span class="sd"> * StructType --> dict with keys matching struct fields.</span> |
| <span class="sd"> * StructType --> list of dict with keys matching struct fields, for models like the</span> |
| <span class="sd"> `Huggingface pipeline for sentiment analysis</span> |
| <span class="sd"> <https://huggingface.co/docs/transformers/quicktour#pipeline-usage>`_.</span> |
| |
| <span class="sd"> batch_size : int</span> |
| <span class="sd"> Batch size to use for inference. This is typically a limitation of the model</span> |
| <span class="sd"> and/or available hardware resources and is usually smaller than the Spark partition size.</span> |
| <span class="sd"> input_tensor_shapes : list, dict, optional.</span> |
| <span class="sd"> A list of ints or a dictionary of ints (key) and list of ints (value).</span> |
| <span class="sd"> Input tensor shapes for models with tensor inputs. This can be a list of shapes,</span> |
| <span class="sd"> where each shape is a list of integers or None (for scalar inputs). Alternatively, this</span> |
| <span class="sd"> can be represented by a "sparse" dictionary, where the keys are the integer indices of the</span> |
| <span class="sd"> inputs, and the values are the shapes. Each tensor input value in the Spark DataFrame must</span> |
| <span class="sd"> be represented as a single column containing a flattened 1-D array. The provided</span> |
| <span class="sd"> `input_tensor_shapes` will be used to reshape the flattened array into the expected tensor</span> |
| <span class="sd"> shape. For the list form, the order of the tensor shapes must match the order of the</span> |
| <span class="sd"> selected DataFrame columns. The batch dimension (typically -1 or None in the first</span> |
| <span class="sd"> dimension) should not be included, since it will be determined by the batch_size argument.</span> |
| <span class="sd"> Tabular datasets with scalar-valued columns should not provide this argument.</span> |
| |
| <span class="sd"> Returns</span> |
| <span class="sd"> -------</span> |
| <span class="sd"> :py:class:`UserDefinedFunctionLike`</span> |
| <span class="sd"> A Pandas UDF for model inference on a Spark DataFrame.</span> |
| |
| <span class="sd"> Examples</span> |
| <span class="sd"> --------</span> |
| <span class="sd"> For a pre-trained TensorFlow MNIST model with two-dimensional input images represented as a</span> |
| <span class="sd"> flattened tensor value stored in a single Spark DataFrame column of type `array<float>`.</span> |
| |
| <span class="sd"> .. code-block:: python</span> |
| |
| <span class="sd"> from pyspark.ml.functions import predict_batch_udf</span> |
| |
| <span class="sd"> def make_mnist_fn():</span> |
| <span class="sd"> # load/init happens once per python worker</span> |
| <span class="sd"> import tensorflow as tf</span> |
| <span class="sd"> model = tf.keras.models.load_model('/path/to/mnist_model')</span> |
| |
| <span class="sd"> # predict on batches of tasks/partitions, using cached model</span> |
| <span class="sd"> def predict(inputs: np.ndarray) -> np.ndarray:</span> |
| <span class="sd"> # inputs.shape = [batch_size, 784], see input_tensor_shapes</span> |
| <span class="sd"> # outputs.shape = [batch_size, 10], see return_type</span> |
| <span class="sd"> return model.predict(inputs)</span> |
| |
| <span class="sd"> return predict</span> |
| |
| <span class="sd"> mnist_udf = predict_batch_udf(make_mnist_fn,</span> |
| <span class="sd"> return_type=ArrayType(FloatType()),</span> |
| <span class="sd"> batch_size=100,</span> |
| <span class="sd"> input_tensor_shapes=[[784]])</span> |
| |
| <span class="sd"> df = spark.read.parquet("/path/to/mnist_data")</span> |
| <span class="sd"> df.show(5)</span> |
| <span class="sd"> # +--------------------+</span> |
| <span class="sd"> # | data|</span> |
| <span class="sd"> # +--------------------+</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|</span> |
| <span class="sd"> # +--------------------+</span> |
| |
| <span class="sd"> df.withColumn("preds", mnist_udf("data")).show(5)</span> |
| <span class="sd"> # +--------------------+--------------------+</span> |
| <span class="sd"> # | data| preds|</span> |
| <span class="sd"> # +--------------------+--------------------+</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|[-13.511008, 8.84...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|[-5.3957458, -2.2...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|[-7.2014456, -8.8...|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|[-19.466187, -13....|</span> |
| <span class="sd"> # |[0.0, 0.0, 0.0, 0...|[-5.7757926, -7.8...|</span> |
| <span class="sd"> # +--------------------+--------------------+</span> |
| |
| <span class="sd"> To demonstrate usage with different combinations of input and output types, the following</span> |
| <span class="sd"> examples just use simple mathematical transforms as the models.</span> |
| |
| <span class="sd"> * Single scalar column</span> |
| <span class="sd"> Input DataFrame has a single scalar column, which will be passed to the `predict`</span> |
| <span class="sd"> function as a 1-D numpy array.</span> |
| |
| <span class="sd"> >>> import numpy as np</span> |
| <span class="sd"> >>> import pandas as pd</span> |
| <span class="sd"> >>> from pyspark.ml.functions import predict_batch_udf</span> |
| <span class="sd"> >>> from pyspark.sql.types import FloatType</span> |
| <span class="sd"> >>></span> |
| <span class="sd"> >>> df = spark.createDataFrame(pd.DataFrame(np.arange(100)))</span> |
| <span class="sd"> >>> df.show(5)</span> |
| <span class="sd"> +---+</span> |
| <span class="sd"> | 0|</span> |
| <span class="sd"> +---+</span> |
| <span class="sd"> | 0|</span> |
| <span class="sd"> | 1|</span> |
| <span class="sd"> | 2|</span> |
| <span class="sd"> | 3|</span> |
| <span class="sd"> | 4|</span> |
| <span class="sd"> +---+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> >>> def make_times_two_fn():</span> |
| <span class="sd"> ... def predict(inputs: np.ndarray) -> np.ndarray:</span> |
| <span class="sd"> ... # inputs.shape = [batch_size]</span> |
| <span class="sd"> ... # outputs.shape = [batch_size]</span> |
| <span class="sd"> ... return inputs * 2</span> |
| <span class="sd"> ... return predict</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> times_two_udf = predict_batch_udf(make_times_two_fn,</span> |
| <span class="sd"> ... return_type=FloatType(),</span> |
| <span class="sd"> ... batch_size=10)</span> |
| <span class="sd"> >>> df = spark.createDataFrame(pd.DataFrame(np.arange(100)))</span> |
| <span class="sd"> >>> df.withColumn("x2", times_two_udf("0")).show(5)</span> |
| <span class="sd"> +---+---+</span> |
| <span class="sd"> | 0| x2|</span> |
| <span class="sd"> +---+---+</span> |
| <span class="sd"> | 0|0.0|</span> |
| <span class="sd"> | 1|2.0|</span> |
| <span class="sd"> | 2|4.0|</span> |
| <span class="sd"> | 3|6.0|</span> |
| <span class="sd"> | 4|8.0|</span> |
| <span class="sd"> +---+---+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> * Multiple scalar columns</span> |
| <span class="sd"> Input DataFrame has multiple columns of scalar values. If the user-provided `predict`</span> |
| <span class="sd"> function expects a single input, then the user must combine the multiple columns into a</span> |
| <span class="sd"> single tensor using `pyspark.sql.functions.array`.</span> |
| |
| <span class="sd"> >>> import numpy as np</span> |
| <span class="sd"> >>> import pandas as pd</span> |
| <span class="sd"> >>> from pyspark.ml.functions import predict_batch_udf</span> |
| <span class="sd"> >>> from pyspark.sql.functions import array</span> |
| <span class="sd"> >>></span> |
| <span class="sd"> >>> data = np.arange(0, 1000, dtype=np.float64).reshape(-1, 4)</span> |
| <span class="sd"> >>> pdf = pd.DataFrame(data, columns=['a','b','c','d'])</span> |
| <span class="sd"> >>> df = spark.createDataFrame(pdf)</span> |
| <span class="sd"> >>> df.show(5)</span> |
| <span class="sd"> +----+----+----+----+</span> |
| <span class="sd"> | a| b| c| d|</span> |
| <span class="sd"> +----+----+----+----+</span> |
| <span class="sd"> | 0.0| 1.0| 2.0| 3.0|</span> |
| <span class="sd"> | 4.0| 5.0| 6.0| 7.0|</span> |
| <span class="sd"> | 8.0| 9.0|10.0|11.0|</span> |
| <span class="sd"> |12.0|13.0|14.0|15.0|</span> |
| <span class="sd"> |16.0|17.0|18.0|19.0|</span> |
| <span class="sd"> +----+----+----+----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> >>> def make_sum_fn():</span> |
| <span class="sd"> ... def predict(inputs: np.ndarray) -> np.ndarray:</span> |
| <span class="sd"> ... # inputs.shape = [batch_size, 4]</span> |
| <span class="sd"> ... # outputs.shape = [batch_size]</span> |
| <span class="sd"> ... return np.sum(inputs, axis=1)</span> |
| <span class="sd"> ... return predict</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> sum_udf = predict_batch_udf(make_sum_fn,</span> |
| <span class="sd"> ... return_type=FloatType(),</span> |
| <span class="sd"> ... batch_size=10,</span> |
| <span class="sd"> ... input_tensor_shapes=[[4]])</span> |
| <span class="sd"> >>> df.withColumn("sum", sum_udf(array("a", "b", "c", "d"))).show(5)</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> | a| b| c| d| sum|</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> | 0.0| 1.0| 2.0| 3.0| 6.0|</span> |
| <span class="sd"> | 4.0| 5.0| 6.0| 7.0|22.0|</span> |
| <span class="sd"> | 8.0| 9.0|10.0|11.0|38.0|</span> |
| <span class="sd"> |12.0|13.0|14.0|15.0|54.0|</span> |
| <span class="sd"> |16.0|17.0|18.0|19.0|70.0|</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> If the `predict` function expects multiple inputs, then the number of selected input columns</span> |
| <span class="sd"> must match the number of expected inputs.</span> |
| |
| <span class="sd"> >>> def make_sum_fn():</span> |
| <span class="sd"> ... def predict(x1: np.ndarray,</span> |
| <span class="sd"> ... x2: np.ndarray,</span> |
| <span class="sd"> ... x3: np.ndarray,</span> |
| <span class="sd"> ... x4: np.ndarray) -> np.ndarray:</span> |
| <span class="sd"> ... # xN.shape = [batch_size]</span> |
| <span class="sd"> ... # outputs.shape = [batch_size]</span> |
| <span class="sd"> ... return x1 + x2 + x3 + x4</span> |
| <span class="sd"> ... return predict</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> sum_udf = predict_batch_udf(make_sum_fn,</span> |
| <span class="sd"> ... return_type=FloatType(),</span> |
| <span class="sd"> ... batch_size=10)</span> |
| <span class="sd"> >>> df.withColumn("sum", sum_udf("a", "b", "c", "d")).show(5)</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> | a| b| c| d| sum|</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> | 0.0| 1.0| 2.0| 3.0| 6.0|</span> |
| <span class="sd"> | 4.0| 5.0| 6.0| 7.0|22.0|</span> |
| <span class="sd"> | 8.0| 9.0|10.0|11.0|38.0|</span> |
| <span class="sd"> |12.0|13.0|14.0|15.0|54.0|</span> |
| <span class="sd"> |16.0|17.0|18.0|19.0|70.0|</span> |
| <span class="sd"> +----+----+----+----+----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> * Multiple tensor columns</span> |
| <span class="sd"> Input DataFrame has multiple columns, where each column is a tensor. The number of columns</span> |
| <span class="sd"> should match the number of expected inputs for the user-provided `predict` function.</span> |
| |
| <span class="sd"> >>> import numpy as np</span> |
| <span class="sd"> >>> import pandas as pd</span> |
| <span class="sd"> >>> from pyspark.ml.functions import predict_batch_udf</span> |
| <span class="sd"> >>> from pyspark.sql.types import ArrayType, FloatType, StructType, StructField</span> |
| <span class="sd"> >>> from typing import Mapping</span> |
| <span class="sd"> >>></span> |
| <span class="sd"> >>> data = np.arange(0, 1000, dtype=np.float64).reshape(-1, 4)</span> |
| <span class="sd"> >>> pdf = pd.DataFrame(data, columns=['a','b','c','d'])</span> |
| <span class="sd"> >>> pdf_tensor = pd.DataFrame()</span> |
| <span class="sd"> >>> pdf_tensor['t1'] = pdf.values.tolist()</span> |
| <span class="sd"> >>> pdf_tensor['t2'] = pdf.drop(columns='d').values.tolist()</span> |
| <span class="sd"> >>> df = spark.createDataFrame(pdf_tensor)</span> |
| <span class="sd"> >>> df.show(5)</span> |
| <span class="sd"> +--------------------+------------------+</span> |
| <span class="sd"> | t1| t2|</span> |
| <span class="sd"> +--------------------+------------------+</span> |
| <span class="sd"> |[0.0, 1.0, 2.0, 3.0]| [0.0, 1.0, 2.0]|</span> |
| <span class="sd"> |[4.0, 5.0, 6.0, 7.0]| [4.0, 5.0, 6.0]|</span> |
| <span class="sd"> |[8.0, 9.0, 10.0, ...| [8.0, 9.0, 10.0]|</span> |
| <span class="sd"> |[12.0, 13.0, 14.0...|[12.0, 13.0, 14.0]|</span> |
| <span class="sd"> |[16.0, 17.0, 18.0...|[16.0, 17.0, 18.0]|</span> |
| <span class="sd"> +--------------------+------------------+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> >>> def make_multi_sum_fn():</span> |
| <span class="sd"> ... def predict(x1: np.ndarray, x2: np.ndarray) -> np.ndarray:</span> |
| <span class="sd"> ... # x1.shape = [batch_size, 4]</span> |
| <span class="sd"> ... # x2.shape = [batch_size, 3]</span> |
| <span class="sd"> ... # outputs.shape = [batch_size]</span> |
| <span class="sd"> ... return np.sum(x1, axis=1) + np.sum(x2, axis=1)</span> |
| <span class="sd"> ... return predict</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> multi_sum_udf = predict_batch_udf(</span> |
| <span class="sd"> ... make_multi_sum_fn,</span> |
| <span class="sd"> ... return_type=FloatType(),</span> |
| <span class="sd"> ... batch_size=5,</span> |
| <span class="sd"> ... input_tensor_shapes=[[4], [3]],</span> |
| <span class="sd"> ... )</span> |
| <span class="sd"> >>> df.withColumn("sum", multi_sum_udf("t1", "t2")).show(5)</span> |
| <span class="sd"> +--------------------+------------------+-----+</span> |
| <span class="sd"> | t1| t2| sum|</span> |
| <span class="sd"> +--------------------+------------------+-----+</span> |
| <span class="sd"> |[0.0, 1.0, 2.0, 3.0]| [0.0, 1.0, 2.0]| 9.0|</span> |
| <span class="sd"> |[4.0, 5.0, 6.0, 7.0]| [4.0, 5.0, 6.0]| 37.0|</span> |
| <span class="sd"> |[8.0, 9.0, 10.0, ...| [8.0, 9.0, 10.0]| 65.0|</span> |
| <span class="sd"> |[12.0, 13.0, 14.0...|[12.0, 13.0, 14.0]| 93.0|</span> |
| <span class="sd"> |[16.0, 17.0, 18.0...|[16.0, 17.0, 18.0]|121.0|</span> |
| <span class="sd"> +--------------------+------------------+-----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> * Multiple outputs</span> |
| <span class="sd"> Some models can provide multiple outputs. These can be returned as a dictionary of named</span> |
| <span class="sd"> values, which can be represented in either columnar or row-based formats.</span> |
| |
| <span class="sd"> >>> def make_multi_sum_fn():</span> |
| <span class="sd"> ... def predict_columnar(x1: np.ndarray, x2: np.ndarray) -> Mapping[str, np.ndarray]:</span> |
| <span class="sd"> ... # x1.shape = [batch_size, 4]</span> |
| <span class="sd"> ... # x2.shape = [batch_size, 3]</span> |
| <span class="sd"> ... return {</span> |
| <span class="sd"> ... "sum1": np.sum(x1, axis=1),</span> |
| <span class="sd"> ... "sum2": np.sum(x2, axis=1)</span> |
| <span class="sd"> ... }</span> |
| <span class="sd"> ... return predict_columnar</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> multi_sum_udf = predict_batch_udf(</span> |
| <span class="sd"> ... make_multi_sum_fn,</span> |
| <span class="sd"> ... return_type=StructType([</span> |
| <span class="sd"> ... StructField("sum1", FloatType(), True),</span> |
| <span class="sd"> ... StructField("sum2", FloatType(), True)</span> |
| <span class="sd"> ... ]),</span> |
| <span class="sd"> ... batch_size=5,</span> |
| <span class="sd"> ... input_tensor_shapes=[[4], [3]],</span> |
| <span class="sd"> ... )</span> |
| <span class="sd"> >>> df.withColumn("preds", multi_sum_udf("t1", "t2")).select("t1", "t2", "preds.*").show(5)</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> | t1| t2|sum1|sum2|</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> |[0.0, 1.0, 2.0, 3.0]| [0.0, 1.0, 2.0]| 6.0| 3.0|</span> |
| <span class="sd"> |[4.0, 5.0, 6.0, 7.0]| [4.0, 5.0, 6.0]|22.0|15.0|</span> |
| <span class="sd"> |[8.0, 9.0, 10.0, ...| [8.0, 9.0, 10.0]|38.0|27.0|</span> |
| <span class="sd"> |[12.0, 13.0, 14.0...|[12.0, 13.0, 14.0]|54.0|39.0|</span> |
| <span class="sd"> |[16.0, 17.0, 18.0...|[16.0, 17.0, 18.0]|70.0|51.0|</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> >>> def make_multi_sum_fn():</span> |
| <span class="sd"> ... def predict_row(x1: np.ndarray, x2: np.ndarray) -> list[Mapping[str, float]]:</span> |
| <span class="sd"> ... # x1.shape = [batch_size, 4]</span> |
| <span class="sd"> ... # x2.shape = [batch_size, 3]</span> |
| <span class="sd"> ... return [{'sum1': np.sum(x1[i]), 'sum2': np.sum(x2[i])} for i in range(len(x1))]</span> |
| <span class="sd"> ... return predict_row</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> multi_sum_udf = predict_batch_udf(</span> |
| <span class="sd"> ... make_multi_sum_fn,</span> |
| <span class="sd"> ... return_type=StructType([</span> |
| <span class="sd"> ... StructField("sum1", FloatType(), True),</span> |
| <span class="sd"> ... StructField("sum2", FloatType(), True)</span> |
| <span class="sd"> ... ]),</span> |
| <span class="sd"> ... batch_size=5,</span> |
| <span class="sd"> ... input_tensor_shapes=[[4], [3]],</span> |
| <span class="sd"> ... )</span> |
| <span class="sd"> >>> df.withColumn("sum", multi_sum_udf("t1", "t2")).select("t1", "t2", "sum.*").show(5)</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> | t1| t2|sum1|sum2|</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> |[0.0, 1.0, 2.0, 3.0]| [0.0, 1.0, 2.0]| 6.0| 3.0|</span> |
| <span class="sd"> |[4.0, 5.0, 6.0, 7.0]| [4.0, 5.0, 6.0]|22.0|15.0|</span> |
| <span class="sd"> |[8.0, 9.0, 10.0, ...| [8.0, 9.0, 10.0]|38.0|27.0|</span> |
| <span class="sd"> |[12.0, 13.0, 14.0...|[12.0, 13.0, 14.0]|54.0|39.0|</span> |
| <span class="sd"> |[16.0, 17.0, 18.0...|[16.0, 17.0, 18.0]|70.0|51.0|</span> |
| <span class="sd"> +--------------------+------------------+----+----+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| |
| <span class="sd"> Note that the multiple outputs can be arrays as well.</span> |
| |
| <span class="sd"> >>> def make_multi_times_two_fn():</span> |
| <span class="sd"> ... def predict(x1: np.ndarray, x2: np.ndarray) -> Mapping[str, np.ndarray]:</span> |
| <span class="sd"> ... # x1.shape = [batch_size, 4]</span> |
| <span class="sd"> ... # x2.shape = [batch_size, 3]</span> |
| <span class="sd"> ... return {"t1x2": x1 * 2, "t2x2": x2 * 2}</span> |
| <span class="sd"> ... return predict</span> |
| <span class="sd"> ...</span> |
| <span class="sd"> >>> multi_times_two_udf = predict_batch_udf(</span> |
| <span class="sd"> ... make_multi_times_two_fn,</span> |
| <span class="sd"> ... return_type=StructType([</span> |
| <span class="sd"> ... StructField("t1x2", ArrayType(FloatType()), True),</span> |
| <span class="sd"> ... StructField("t2x2", ArrayType(FloatType()), True)</span> |
| <span class="sd"> ... ]),</span> |
| <span class="sd"> ... batch_size=5,</span> |
| <span class="sd"> ... input_tensor_shapes=[[4], [3]],</span> |
| <span class="sd"> ... )</span> |
| <span class="sd"> >>> df.withColumn("x2", multi_times_two_udf("t1", "t2")).select("t1", "t2", "x2.*").show(5)</span> |
| <span class="sd"> +--------------------+------------------+--------------------+------------------+</span> |
| <span class="sd"> | t1| t2| t1x2| t2x2|</span> |
| <span class="sd"> +--------------------+------------------+--------------------+------------------+</span> |
| <span class="sd"> |[0.0, 1.0, 2.0, 3.0]| [0.0, 1.0, 2.0]|[0.0, 2.0, 4.0, 6.0]| [0.0, 2.0, 4.0]|</span> |
| <span class="sd"> |[4.0, 5.0, 6.0, 7.0]| [4.0, 5.0, 6.0]|[8.0, 10.0, 12.0,...| [8.0, 10.0, 12.0]|</span> |
| <span class="sd"> |[8.0, 9.0, 10.0, ...| [8.0, 9.0, 10.0]|[16.0, 18.0, 20.0...|[16.0, 18.0, 20.0]|</span> |
| <span class="sd"> |[12.0, 13.0, 14.0...|[12.0, 13.0, 14.0]|[24.0, 26.0, 28.0...|[24.0, 26.0, 28.0]|</span> |
| <span class="sd"> |[16.0, 17.0, 18.0...|[16.0, 17.0, 18.0]|[32.0, 34.0, 36.0...|[32.0, 34.0, 36.0]|</span> |
| <span class="sd"> +--------------------+------------------+--------------------+------------------+</span> |
| <span class="sd"> only showing top 5 rows</span> |
| <span class="sd"> """</span> |
| <span class="c1"># generate a new uuid each time this is invoked on the driver to invalidate executor-side cache.</span> |
| <span class="n">model_uuid</span> <span class="o">=</span> <span class="n">uuid</span><span class="o">.</span><span class="n">uuid4</span><span class="p">()</span> |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">predict</span><span class="p">(</span><span class="n">data</span><span class="p">:</span> <span class="n">Iterator</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">Series</span><span class="p">,</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">]])</span> <span class="o">-></span> <span class="n">Iterator</span><span class="p">[</span><span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">]:</span> |
| <span class="c1"># TODO: adjust return type hint when Iterator[Union[pd.Series, pd.DataFrame]] is supported</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.ml.model_cache</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelCache</span> |
| |
| <span class="c1"># get predict function (from cache or from running user-provided make_predict_fn)</span> |
| <span class="n">predict_fn</span> <span class="o">=</span> <span class="n">ModelCache</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">model_uuid</span><span class="p">)</span> |
| <span class="k">if</span> <span class="ow">not</span> <span class="n">predict_fn</span><span class="p">:</span> |
| <span class="n">predict_fn</span> <span class="o">=</span> <span class="n">make_predict_fn</span><span class="p">()</span> |
| <span class="n">ModelCache</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">model_uuid</span><span class="p">,</span> <span class="n">predict_fn</span><span class="p">)</span> |
| |
| <span class="c1"># get number of expected parameters for predict function</span> |
| <span class="n">signature</span> <span class="o">=</span> <span class="n">inspect</span><span class="o">.</span><span class="n">signature</span><span class="p">(</span><span class="n">predict_fn</span><span class="p">)</span> |
| <span class="n">num_expected_cols</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">signature</span><span class="o">.</span><span class="n">parameters</span><span class="p">)</span> |
| |
| <span class="c1"># convert sparse input_tensor_shapes to dense if needed</span> |
| <span class="n">input_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">|</span> <span class="kc">None</span><span class="p">]</span> |
| <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">input_tensor_shapes</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">):</span> |
| <span class="n">input_shapes</span> <span class="o">=</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_expected_cols</span> |
| <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">shape</span> <span class="ow">in</span> <span class="n">input_tensor_shapes</span><span class="o">.</span><span class="n">items</span><span class="p">():</span> |
| <span class="n">input_shapes</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="o">=</span> <span class="n">shape</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="n">input_shapes</span> <span class="o">=</span> <span class="n">input_tensor_shapes</span> <span class="c1"># type: ignore</span> |
| |
| <span class="c1"># iterate over pandas batch, invoking predict_fn with ndarrays</span> |
| <span class="k">for</span> <span class="n">pandas_batch</span> <span class="ow">in</span> <span class="n">data</span><span class="p">:</span> |
| <span class="n">has_tuple</span> <span class="o">=</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pandas_batch</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">)</span> <span class="c1"># type: ignore</span> |
| <span class="n">has_tensors</span> <span class="o">=</span> <span class="n">_has_tensor_cols</span><span class="p">(</span><span class="n">pandas_batch</span><span class="p">)</span> |
| |
| <span class="c1"># require input_tensor_shapes for any tensor columns</span> |
| <span class="k">if</span> <span class="n">has_tensors</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">input_shapes</span><span class="p">:</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">"Tensor columns require input_tensor_shapes"</span><span class="p">)</span> |
| |
| <span class="k">for</span> <span class="n">batch</span> <span class="ow">in</span> <span class="n">_batched</span><span class="p">(</span><span class="n">pandas_batch</span><span class="p">,</span> <span class="n">batch_size</span><span class="p">):</span> |
| <span class="n">num_input_rows</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">)</span> |
| <span class="n">num_input_cols</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">columns</span><span class="p">)</span> |
| <span class="k">if</span> <span class="n">num_input_cols</span> <span class="o">==</span> <span class="n">num_expected_cols</span> <span class="ow">and</span> <span class="n">num_expected_cols</span> <span class="o">></span> <span class="mi">1</span><span class="p">:</span> |
| <span class="c1"># input column per expected input for multiple inputs</span> |
| <span class="n">multi_inputs</span> <span class="o">=</span> <span class="n">_validate_and_transform_multiple_inputs</span><span class="p">(</span> |
| <span class="n">batch</span><span class="p">,</span> <span class="n">input_shapes</span><span class="p">,</span> <span class="n">num_input_cols</span> |
| <span class="p">)</span> |
| <span class="c1"># run model prediction function on multiple (numpy) inputs</span> |
| <span class="n">preds</span> <span class="o">=</span> <span class="n">predict_fn</span><span class="p">(</span><span class="o">*</span><span class="n">multi_inputs</span><span class="p">)</span> |
| <span class="k">elif</span> <span class="n">num_expected_cols</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span> |
| <span class="c1"># one or more input columns for single expected input</span> |
| <span class="n">single_input</span> <span class="o">=</span> <span class="n">_validate_and_transform_single_input</span><span class="p">(</span> |
| <span class="n">batch</span><span class="p">,</span> <span class="n">input_shapes</span><span class="p">,</span> <span class="n">has_tensors</span><span class="p">,</span> <span class="n">has_tuple</span> |
| <span class="p">)</span> |
| <span class="c1"># run model prediction function on single (numpy) inputs</span> |
| <span class="n">preds</span> <span class="o">=</span> <span class="n">predict_fn</span><span class="p">(</span><span class="n">single_input</span><span class="p">)</span> |
| <span class="k">else</span><span class="p">:</span> |
| <span class="n">msg</span> <span class="o">=</span> <span class="s2">"Model expected </span><span class="si">{}</span><span class="s2"> inputs, but received </span><span class="si">{}</span><span class="s2"> columns"</span> |
| <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">num_expected_cols</span><span class="p">,</span> <span class="n">num_input_cols</span><span class="p">))</span> |
| |
| <span class="c1"># return transformed predictions to Spark</span> |
| <span class="k">yield</span> <span class="n">_validate_and_transform_prediction_result</span><span class="p">(</span> |
| <span class="n">preds</span><span class="p">,</span> <span class="n">num_input_rows</span><span class="p">,</span> <span class="n">return_type</span> |
| <span class="p">)</span> <span class="c1"># type: ignore</span> |
| |
| <span class="k">return</span> <span class="n">pandas_udf</span><span class="p">(</span><span class="n">predict</span><span class="p">,</span> <span class="n">return_type</span><span class="p">)</span> <span class="c1"># type: ignore[call-overload]</span></div> |
| |
| |
| <span class="k">def</span><span class="w"> </span><span class="nf">_test</span><span class="p">()</span> <span class="o">-></span> <span class="kc">None</span><span class="p">:</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">doctest</span> |
| <span class="kn">from</span><span class="w"> </span><span class="nn">pyspark.sql</span><span class="w"> </span><span class="kn">import</span> <span class="n">SparkSession</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">pyspark.ml.functions</span> |
| <span class="kn">import</span><span class="w"> </span><span class="nn">sys</span> |
| |
| <span class="n">globs</span> <span class="o">=</span> <span class="n">pyspark</span><span class="o">.</span><span class="n">ml</span><span class="o">.</span><span class="n">functions</span><span class="o">.</span><span class="vm">__dict__</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span> |
| <span class="n">spark</span> <span class="o">=</span> <span class="n">SparkSession</span><span class="o">.</span><span class="n">builder</span><span class="o">.</span><span class="n">master</span><span class="p">(</span><span class="s2">"local[2]"</span><span class="p">)</span><span class="o">.</span><span class="n">appName</span><span class="p">(</span><span class="s2">"ml.functions tests"</span><span class="p">)</span><span class="o">.</span><span class="n">getOrCreate</span><span class="p">()</span> |
| <span class="n">sc</span> <span class="o">=</span> <span class="n">spark</span><span class="o">.</span><span class="n">sparkContext</span> |
| <span class="n">globs</span><span class="p">[</span><span class="s2">"sc"</span><span class="p">]</span> <span class="o">=</span> <span class="n">sc</span> |
| <span class="n">globs</span><span class="p">[</span><span class="s2">"spark"</span><span class="p">]</span> <span class="o">=</span> <span class="n">spark</span> |
| |
| <span class="p">(</span><span class="n">failure_count</span><span class="p">,</span> <span class="n">test_count</span><span class="p">)</span> <span class="o">=</span> <span class="n">doctest</span><span class="o">.</span><span class="n">testmod</span><span class="p">(</span> |
| <span class="n">pyspark</span><span class="o">.</span><span class="n">ml</span><span class="o">.</span><span class="n">functions</span><span class="p">,</span> |
| <span class="n">globs</span><span class="o">=</span><span class="n">globs</span><span class="p">,</span> |
| <span class="n">optionflags</span><span class="o">=</span><span class="n">doctest</span><span class="o">.</span><span class="n">ELLIPSIS</span> <span class="o">|</span> <span class="n">doctest</span><span class="o">.</span><span class="n">NORMALIZE_WHITESPACE</span><span class="p">,</span> |
| <span class="p">)</span> |
| <span class="n">spark</span><span class="o">.</span><span class="n">stop</span><span class="p">()</span> |
| <span class="k">if</span> <span class="n">failure_count</span><span class="p">:</span> |
| <span class="n">sys</span><span class="o">.</span><span class="n">exit</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> |
| |
| |
| <span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">"__main__"</span><span class="p">:</span> |
| <span class="n">_test</span><span class="p">()</span> |
| </pre></div> |
| |
| </div> |
| |
| |
| <!-- Previous / next buttons --> |
| <div class='prev-next-area'> |
| </div> |
| |
| </main> |
| |
| |
| </div> |
| </div> |
| |
| <script src="../../../_static/scripts/pydata-sphinx-theme.js?digest=1999514e3f237ded88cf"></script> |
| <footer class="footer mt-5 mt-md-0"> |
| <div class="container"> |
| |
| <div class="footer-item"> |
| <p class="copyright"> |
| © Copyright .<br> |
| </p> |
| </div> |
| |
| <div class="footer-item"> |
| <p class="sphinx-version"> |
| Created using <a href="http://sphinx-doc.org/">Sphinx</a> 3.0.4.<br> |
| </p> |
| </div> |
| |
| </div> |
| </footer> |
| </body> |
| </html> |
