SkyID
/
deepke
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
deepke/deepke.name_entity_recognit...

1230 lines
191 KiB
HTML
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!DOCTYPE html>
<html class="writer-html5" lang="en" >
<head>
<meta charset="utf-8" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Models &mdash; DeepKE 1.0.0 documentation</title>
<link rel="stylesheet" href="_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
<link rel="stylesheet" href="_static/css/custom.css" type="text/css" />
<!--[if lt IE 9]>
<script src="_static/js/html5shiv.min.js"></script>
<![endif]-->
<script data-url_root="./" id="documentation_options" 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/clipboard.min.js"></script>
<script src="_static/copybutton.js"></script>
<script src="_static/js/theme.js"></script>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Module" href="deepke.name_entity_recognition.few_shot.module.html" />
<link rel="prev" title="Few Shot" href="deepke.name_entity_recognition.few_shot.html" />
</head>
<body class="wy-body-for-nav">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
<div class="wy-side-scroll">
<div class="wy-side-nav-search" >
<a href="index.html" class="icon icon-home"> DeepKE
</a>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="search.html" method="get">
<input type="text" name="q" placeholder="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div><div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="Navigation menu">
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="start.html">Start</a></li>
<li class="toctree-l1"><a class="reference internal" href="install.html">Install</a></li>
<li class="toctree-l1"><a class="reference internal" href="example.html">Example</a></li>
<li class="toctree-l1"><a class="reference internal" href="faq.html">FAQ</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Package</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="deepke.html">DeepKE</a><ul class="current">
<li class="toctree-l2"><a class="reference internal" href="deepke.attribution_extraction.html">Attribution Extraction</a></li>
<li class="toctree-l2 current"><a class="reference internal" href="deepke.name_entity_recognition.html">Name Entity Recognition</a><ul class="current">
<li class="toctree-l3 current"><a class="reference internal" href="deepke.name_entity_recognition.few_shot.html">Few Shot</a><ul class="current">
<li class="toctree-l4 current"><a class="current reference internal" href="#">Models</a></li>
<li class="toctree-l4"><a class="reference internal" href="deepke.name_entity_recognition.few_shot.module.html">Module</a></li>
<li class="toctree-l4"><a class="reference internal" href="deepke.name_entity_recognition.few_shot.utils.html">Utils</a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="deepke.name_entity_recognition.standard.html">Standard</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="deepke.relation_extraction.html">Relation Extraction</a></li>
</ul>
</li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"><nav class="wy-nav-top" aria-label="Mobile navigation menu" >
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="index.html">DeepKE</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="Page navigation">
<ul class="wy-breadcrumbs">
<li><a href="index.html" class="icon icon-home"></a> &raquo;</li>
<li><a href="deepke.html">DeepKE</a> &raquo;</li>
<li><a href="deepke.name_entity_recognition.html">Name Entity Recognition</a> &raquo;</li>
<li><a href="deepke.name_entity_recognition.few_shot.html">Few Shot</a> &raquo;</li>
<li>Models</li>
<li class="wy-breadcrumbs-aside">
<a href="https://github.com/tlk1997/test_doc/blob/main/docs/source/deepke.name_entity_recognition.few_shot.models.rst" class="fa fa-github"> Edit on GitHub</a>
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<section id="models">
<h1>Models<a class="headerlink" href="#models" title="Permalink to this headline"></a></h1>
<section id="module-deepke.name_entity_recognition.few_shot.models.model">
<span id="deepke-name-entity-recognition-few-shot-models-model-module"></span><h2>deepke.name_entity_recognition.few_shot.models.model module<a class="headerlink" href="#module-deepke.name_entity_recognition.few_shot.models.model" title="Permalink to this headline"></a></h2>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartEncoder">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">PromptBartEncoder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoder</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartEncoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartEncoder" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartEncoder.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartEncoder.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartEncoder.forward" title="Permalink to this definition"></a></dt>
<dd><p>Defines the computation performed at every call.</p>
<p>Should be overridden by all subclasses.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.</p>
</div>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">PromptBartDecoder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">decoder</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">label_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prompt_len</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">10</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">learn_weights</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartDecoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tgt_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prompt_state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartDecoder.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder.forward" title="Permalink to this definition"></a></dt>
<dd><p>Defines the computation performed at every call.</p>
<p>Should be overridden by all subclasses.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.</p>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder.decode">
<span class="sig-name descname"><span class="pre">decode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartDecoder.decode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartDecoder.decode" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartModel">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">PromptBartModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">label_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartModel" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tgt_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_seq_len</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">first</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartModel.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.forward" title="Permalink to this definition"></a></dt>
<dd><p>Defines the computation performed at every call.</p>
<p>Should be overridden by all subclasses.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.</p>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.generator">
<span class="sig-name descname"><span class="pre">generator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_seq_len</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">first</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartModel.generator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.generator" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.get_prompt">
<span class="sig-name descname"><span class="pre">get_prompt</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">batch_size</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartModel.get_prompt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartModel.get_prompt" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartState">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">PromptBartState</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoder_output</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">first</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_embed_outputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">preseqlen</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartState"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartState" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference external" href="https://docs.python.org/3/library/functions.html#object" title="(in Python v3.10)"><code class="xref py py-class docutils literal notranslate"><span class="pre">object</span></code></a></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartState.reorder_state">
<span class="sig-name descname"><span class="pre">reorder_state</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">indices</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">torch.LongTensor</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartState.reorder_state"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartState.reorder_state" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptBartState.num_samples">
<span class="sig-name descname"><span class="pre">num_samples</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptBartState.num_samples"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptBartState.num_samples" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">PromptGeneratorModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">prompt_model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">20</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_len_a</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_beams</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">do_sample</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bos_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eos_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">repetition_penalty</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">length_penalty</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">restricter</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptGeneratorModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tgt_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_seq_len</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tgt_seq_len</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">first</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptGeneratorModel.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel.forward" title="Permalink to this definition"></a></dt>
<dd><dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>src_tokens</strong> (<em>torch.LongTensor</em>) – bsz x max_len</p></li>
<li><p><strong>tgt_tokens</strong> (<em>torch.LongTensor</em>) – bsz x max_len’</p></li>
<li><p><strong>src_seq_len</strong> (<em>torch.LongTensor</em>) – bsz</p></li>
<li><p><strong>tgt_seq_len</strong> (<em>torch.LongTensor</em>) – bsz</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p></p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel.predict">
<span class="sig-name descname"><span class="pre">predict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">src_seq_len</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">first</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#PromptGeneratorModel.predict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.PromptGeneratorModel.predict" title="Permalink to this definition"></a></dt>
<dd><dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>src_tokens</strong> (<em>torch.LongTensor</em>) – bsz x max_len</p></li>
<li><p><strong>src_seq_len</strong> (<em>torch.LongTensor</em>) – bsz</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.greedy_generate">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">greedy_generate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">decoder</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokens</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">20</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_len_a</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_beams</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bos_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eos_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">repetition_penalty</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">length_penalty</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">restricter</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#greedy_generate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.greedy_generate" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.model.</span></span><span class="sig-name descname"><span class="pre">BeamHypotheses</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_beams</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_length</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">length_penalty</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">early_stopping</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#BeamHypotheses"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference external" href="https://docs.python.org/3/library/functions.html#object" title="(in Python v3.10)"><code class="xref py py-class docutils literal notranslate"><span class="pre">object</span></code></a></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses.add">
<span class="sig-name descname"><span class="pre">add</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">hyp</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sum_logprobs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#BeamHypotheses.add"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses.add" title="Permalink to this definition"></a></dt>
<dd><p>Add a new hypothesis to the list.</p>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses.is_done">
<span class="sig-name descname"><span class="pre">is_done</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">best_sum_logprobs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/model.html#BeamHypotheses.is_done"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.model.BeamHypotheses.is_done" title="Permalink to this definition"></a></dt>
<dd><p>If there are enough hypotheses and that none of the hypotheses being generated
can become better than the worst one in the heap, then we are done with this sentence.</p>
</dd></dl>
</dd></dl>
</section>
<section id="module-deepke.name_entity_recognition.few_shot.models.modeling_bart">
<span id="deepke-name-entity-recognition-few-shot-models-modeling-bart-module"></span><h2>deepke.name_entity_recognition.few_shot.models.modeling_bart module<a class="headerlink" href="#module-deepke.name_entity_recognition.few_shot.models.modeling_bart" title="Permalink to this headline"></a></h2>
<p>PyTorch BART model, ported from the fairseq repo.</p>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.invert_mask">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">invert_mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#invert_mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.invert_mask" title="Permalink to this definition"></a></dt>
<dd><p>Turns 1-&gt;0, 0-&gt;1, False-&gt;True, True-&gt; False</p>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">PretrainedBartModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_utils.PretrainedConfig</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#PretrainedBartModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">transformers.modeling_utils.PreTrainedModel</span></code></p>
<dl class="py attribute">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.config_class">
<span class="sig-name descname"><span class="pre">config_class</span></span><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.config_class" title="Permalink to this definition"></a></dt>
<dd><p>alias of <code class="xref py py-class docutils literal notranslate"><span class="pre">transformers.configuration_bart.BartConfig</span></code></p>
</dd></dl>
<dl class="py attribute">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.base_model_prefix">
<span class="sig-name descname"><span class="pre">base_model_prefix</span></span><em class="property"> <span class="pre">=</span> <span class="pre">'model'</span></em><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.base_model_prefix" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.dummy_inputs">
<em class="property"><span class="pre">property</span> </em><span class="sig-name descname"><span class="pre">dummy_inputs</span></span><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel.dummy_inputs" title="Permalink to this definition"></a></dt>
<dd><p>Dummy inputs to do a forward pass in the network.</p>
<dl class="field-list simple">
<dt class="field-odd">Type</dt>
<dd class="field-odd"><p><code class="xref py py-obj docutils literal notranslate"><span class="pre">Dict[str,</span> <span class="pre">torch.Tensor]</span></code></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.shift_tokens_right">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">shift_tokens_right</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#shift_tokens_right"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.shift_tokens_right" title="Permalink to this definition"></a></dt>
<dd><p>Shift input ids one token to the right, and wrap the last non pad token (usually &lt;eos&gt;).</p>
</dd></dl>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.make_padding_mask">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">make_padding_mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_idx</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#make_padding_mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.make_padding_mask" title="Permalink to this definition"></a></dt>
<dd><p>True for pad tokens</p>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">EncoderLayer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#EncoderLayer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">x</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_padding_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_state</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#EncoderLayer.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer.forward" title="Permalink to this definition"></a></dt>
<dd><dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>x</strong> (<em>Tensor</em>) – input to the layer of shape <cite>(seq_len, batch, embed_dim)</cite></p></li>
<li><p><strong>encoder_padding_mask</strong> (<em>ByteTensor</em>) – binary ByteTensor of shape
<cite>(batch, src_len)</cite> where padding elements are indicated by <code class="docutils literal notranslate"><span class="pre">1</span></code>.</p></li>
<li><p><strong>t_tgt</strong> (<em>for</em>) – </p></li>
<li><p><strong>excluded</strong> (<em>t_src is</em>) – </p></li>
<li><p><strong>attention</strong> (<em>included in</em>) – </p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p>encoded output of shape <cite>(seq_len, batch, embed_dim)</cite></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartEncoder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">embed_tokens</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartEncoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<p>Transformer encoder consisting of <em>config.encoder_layers</em> self attention layers. Each layer
is a <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.EncoderLayer"><code class="xref py py-class docutils literal notranslate"><span class="pre">EncoderLayer</span></code></a>.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>config</strong> – BartConfig</p>
</dd>
</dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartEncoder.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder.forward" title="Permalink to this definition"></a></dt>
<dd><dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<em>LongTensor</em>) – tokens in the source language of shape
<cite>(batch, src_len)</cite></p></li>
<li><p><strong>attention_mask</strong> (<em>torch.LongTensor</em>) – indicating which indices are padding tokens.</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><ul class="simple">
<li><p><strong>x</strong> (Tensor): the last encoder layer’s output of
shape <cite>(src_len, batch, embed_dim)</cite></p></li>
<li><p><strong>encoder_states</strong> (tuple(torch.FloatTensor)): all intermediate
hidden states of shape <cite>(src_len, batch, embed_dim)</cite>.
Only populated if <em>output_hidden_states:</em> is True.</p></li>
<li><p><strong>all_attentions</strong> (tuple(torch.FloatTensor)): Attention weights for each layer.</p></li>
</ul>
<p>During training might not be of length n_layers because of layer dropout.</p>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p>BaseModelOutput or Tuple comprised of</p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder.forward_with_encoder_past">
<span class="sig-name descname"><span class="pre">forward_with_encoder_past</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartEncoder.forward_with_encoder_past"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartEncoder.forward_with_encoder_past" title="Permalink to this definition"></a></dt>
<dd><dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<em>LongTensor</em>) – tokens in the source language of shape
<cite>(batch, src_len)</cite></p></li>
<li><p><strong>attention_mask</strong> (<em>torch.LongTensor</em>) – indicating which indices are padding tokens.</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><ul class="simple">
<li><p><strong>x</strong> (Tensor): the last encoder layer’s output of
shape <cite>(src_len, batch, embed_dim)</cite></p></li>
<li><p><strong>encoder_states</strong> (tuple(torch.FloatTensor)): all intermediate
hidden states of shape <cite>(src_len, batch, embed_dim)</cite>.
Only populated if <em>output_hidden_states:</em> is True.</p></li>
<li><p><strong>all_attentions</strong> (tuple(torch.FloatTensor)): Attention weights for each layer.</p></li>
</ul>
<p>During training might not be of length n_layers because of layer dropout.</p>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p>BaseModelOutput or Tuple comprised of</p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">DecoderLayer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#DecoderLayer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">x</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_hidden_states</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_attn_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_state</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_padding_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#DecoderLayer.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer.forward" title="Permalink to this definition"></a></dt>
<dd><p>Defines the computation performed at every call.</p>
<p>Should be overridden by all subclasses.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.</p>
</div>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartDecoder">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartDecoder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">embed_tokens</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">torch.nn.modules.sparse.Embedding</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartDecoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartDecoder" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<p>Transformer decoder consisting of <em>config.decoder_layers</em> layers. Each layer
is a <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.DecoderLayer"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecoderLayer</span></code></a>.
:param config: BartConfig
:param embed_tokens: output embedding
:type embed_tokens: torch.nn.Embedding</p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartDecoder.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_hidden_states</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_padding_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_padding_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_causal_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">unused</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartDecoder.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartDecoder.forward" title="Permalink to this definition"></a></dt>
<dd><p>Includes several features from “Jointly Learning to Align and
Translate with Transformer Models” (Garg et al., EMNLP 2019).</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<em>LongTensor</em>) – previous decoder outputs of shape
<cite>(batch, tgt_len)</cite>, for teacher forcing</p></li>
<li><p><strong>encoder_hidden_states</strong> – output from the encoder, used for
encoder-side attention</p></li>
<li><p><strong>encoder_padding_mask</strong> – for ignoring pad tokens</p></li>
<li><p><strong>past_key_values</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.10)"><em>dict</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.10)"><em>None</em></a>) – dictionary used for storing state during generation</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><ul class="simple">
<li><p>the decoder’s features of shape <cite>(batch, tgt_len, embed_dim)</cite></p></li>
<li><p>the cache</p></li>
<li><p>hidden states</p></li>
<li><p>attentions</p></li>
</ul>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p>BaseModelOutputWithPast or <a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.10)">tuple</a></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.Attention">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">Attention</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">embed_dim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_heads</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dropout</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_decoder_attention</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">cache_key</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">preseqlen</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">-</span> <span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#Attention"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.Attention" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<p>Multi-headed attention from ‘Attention Is All You Need’ paper</p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.Attention.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">query</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">key</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">key_padding_mask</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_state</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.10)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span> </span><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attn_mask</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">,</span> </span><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#Attention.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.Attention.forward" title="Permalink to this definition"></a></dt>
<dd><p>Input shape: Time(SeqLen) x Batch x Channel</p>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartClassificationHead">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartClassificationHead</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_dim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inner_dim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pooler_dropout</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartClassificationHead"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartClassificationHead" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.module.Module</span></code></p>
<p>Head for sentence-level classification tasks.</p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartClassificationHead.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartClassificationHead.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartClassificationHead.forward" title="Permalink to this definition"></a></dt>
<dd><p>Defines the computation performed at every call.</p>
<p>Should be overridden by all subclasses.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.</p>
</div>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.LearnedPositionalEmbedding">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">LearnedPositionalEmbedding</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_embeddings</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.10)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">embedding_dim</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.10)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_idx</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.10)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">offset</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#LearnedPositionalEmbedding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.LearnedPositionalEmbedding" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.sparse.Embedding</span></code></p>
<p>This module learns positional embeddings up to a fixed maximum size.
Padding ids are ignored by either offsetting based on padding_idx
or by setting padding_idx to None and ensuring that the appropriate
position ids are passed to the forward function.</p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.LearnedPositionalEmbedding.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#LearnedPositionalEmbedding.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.LearnedPositionalEmbedding.forward" title="Permalink to this definition"></a></dt>
<dd><p>Input is expected to be of size [bsz x seqlen].</p>
</dd></dl>
</dd></dl>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.LayerNorm">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">LayerNorm</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">normalized_shape</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-05</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">elementwise_affine</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#LayerNorm"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.LayerNorm" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py function">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.fill_with_neg_inf">
<span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">fill_with_neg_inf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">t</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#fill_with_neg_inf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.fill_with_neg_inf" title="Permalink to this definition"></a></dt>
<dd><p>FP16-compatible function that fills a input_ids with -inf.</p>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel"><code class="xref py py-class docutils literal notranslate"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel</span></code></a></p>
<p>The bare BART Model outputting raw hidden-states without any specific head on top.</p>
<p>This model inherits from <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code>. Check the superclass documentation for the generic
methods the library implements for all its model (such as downloading or saving, resizing the input embeddings,
pruning heads etc.)</p>
<p>This model is also a PyTorch <a class="reference external" href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general
usage and behavior.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>config</strong> (<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) – Model configuration class with all the parameters of the model.
Initializing with a config file does not load the weights associated with the model, only the configuration.
Check out the <code class="xref py py-meth docutils literal notranslate"><span class="pre">from_pretrained()</span></code> method to load the model weights.</p>
</dd>
</dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_input_ids</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_outputs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartModel.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.forward" title="Permalink to this definition"></a></dt>
<dd><p>The <code class="xref py py-class docutils literal notranslate"><span class="pre">BartModel</span></code> forward method, overrides the <code class="xref py py-func docutils literal notranslate"><span class="pre">__call__()</span></code> special method.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
pre and post processing steps while the latter silently ignores them.</p>
</div>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – <p>Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
it.</p>
<p>Indices can be obtained using <code class="xref py py-class docutils literal notranslate"><span class="pre">BartTokenizer</span></code>.
See <code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.encode()</span></code> and
<code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.__call__()</span></code> for details.</p>
<p><a class="reference external" href="../glossary.html#input-ids">What are input IDs?</a></p>
</p></li>
<li><p><strong>attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.Tensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>, <cite>optional</cite>) – <p>Mask to avoid performing attention on padding token indices.
Mask values selected in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code>:</p>
<ul>
<li><p>1 for tokens that are <strong>not masked</strong>,</p></li>
<li><p>0 for tokens that are <strong>masked</strong>.</p></li>
</ul>
<p><a class="reference external" href="../glossary.html#attention-mask">What are attention masks?</a></p>
</p></li>
<li><p><strong>decoder_input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">target_sequence_length)</span></code>, <cite>optional</cite>) – Provide for translation and summarization training. By default, the model will create this tensor by
shifting the <code class="xref py py-obj docutils literal notranslate"><span class="pre">input_ids</span></code> to the right, following the paper.</p></li>
<li><p><strong>decoder_attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.BoolTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">tgt_seq_len)</span></code>, <cite>optional</cite>) – <p>Default behavior: generate a tensor that ignores pad tokens in <code class="xref py py-obj docutils literal notranslate"><span class="pre">decoder_input_ids</span></code>. Causal mask will
also be used by default.</p>
<p>If you want to change padding behavior, you should read <code class="xref py py-func docutils literal notranslate"><span class="pre">modeling_bart._prepare_decoder_inputs()</span></code> and
modify to your needs. See diagram 1 in <a class="reference external" href="https://arxiv.org/abs/1910.13461">the paper</a> for more
information on the default strategy.</p>
</p></li>
<li><p><strong>encoder_outputs</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>) – Tuple consists of (<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">hidden_states</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">attentions</span></code>)
<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) is a
sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of
the decoder.</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor))</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code> with each tuple having 4 tensors of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length</span> <span class="pre">-</span> <span class="pre">1,</span> <span class="pre">embed_size_per_head)</span></code>) – <p>Contains precomputed key and value hidden-states of the attention blocks. Can be used to speed up decoding.</p>
<p>If <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> are used, the user can optionally input only the last
<code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> (those that don’t have their past key value states given to this model) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">1)</span></code> instead of all <code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>.</p>
</p></li>
<li><p><strong>use_cache</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – If set to <a class="reference external" href="https://docs.python.org/3/library/constants.html#True" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">True</span></code></a>, <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> key value states are returned and can be used to speed up
decoding (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code>).</p></li>
<li><p><strong>output_attentions</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the attentions tensors of all attention layers. See <code class="docutils literal notranslate"><span class="pre">attentions</span></code> under returned
tensors for more detail.</p></li>
<li><p><strong>output_hidden_states</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the hidden states of all layers. See <code class="docutils literal notranslate"><span class="pre">hidden_states</span></code> under returned tensors for
more detail.</p></li>
<li><p><strong>return_dict</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return a <code class="xref py py-class docutils literal notranslate"><span class="pre">ModelOutput</span></code> instead of a plain tuple.</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><p>A <code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqModelOutput</span></code> (if <code class="docutils literal notranslate"><span class="pre">return_dict=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.return_dict=True</span></code>) or a
tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> comprising various elements depending on the configuration
(<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) and inputs.</p>
<ul>
<li><p><strong>last_hidden_state</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>) – Sequence of hidden-states at the output of the last layer of the decoder of the model.</p>
<p>If <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> is used only the last hidden-state of the sequences of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">1,</span> <span class="pre">hidden_size)</span></code> is output.</p>
</li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">List[torch.FloatTensor]</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">use_cache=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.use_cache=True</span></code>) – List of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code>, with each tensor of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(2,</span> <span class="pre">batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">embed_size_per_head)</span></code>).</p>
<p>Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be
used (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> input) to speed up sequential decoding.</p>
</li>
<li><p><strong>decoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>decoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
<li><p><strong>encoder_last_hidden_state</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) – Sequence of hidden-states at the output of the last layer of the encoder of the model.</p></li>
<li><p><strong>encoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>encoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
</ul>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqModelOutput</span></code> or <code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code></p>
</dd>
</dl>
<p>Example:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">transformers</span> <span class="kn">import</span> <span class="n">BartTokenizer</span><span class="p">,</span> <span class="n">BartModel</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">import</span> <span class="nn">torch</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">BartTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">BartModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">,</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="s2">&quot;Hello, my dog is cute&quot;</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">last_hidden_states</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">last_hidden_state</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.get_input_embeddings">
<span class="sig-name descname"><span class="pre">get_input_embeddings</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartModel.get_input_embeddings"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.get_input_embeddings" title="Permalink to this definition"></a></dt>
<dd><p>Returns the model’s input embeddings.</p>
<dl class="field-list simple">
<dt class="field-odd">Returns</dt>
<dd class="field-odd"><p>A torch module mapping vocabulary to hidden states.</p>
</dd>
<dt class="field-even">Return type</dt>
<dd class="field-even"><p><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Module</span></code></p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.set_input_embeddings">
<span class="sig-name descname"><span class="pre">set_input_embeddings</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartModel.set_input_embeddings"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.set_input_embeddings" title="Permalink to this definition"></a></dt>
<dd><p>Set model’s input embeddings.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>value</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Module</span></code>) – A module mapping vocabulary to hidden states.</p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.get_output_embeddings">
<span class="sig-name descname"><span class="pre">get_output_embeddings</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartModel.get_output_embeddings"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartModel.get_output_embeddings" title="Permalink to this definition"></a></dt>
<dd><p>Returns the model’s output embeddings.</p>
<dl class="field-list simple">
<dt class="field-odd">Returns</dt>
<dd class="field-odd"><p>A torch module mapping hidden states to vocabulary.</p>
</dd>
<dt class="field-even">Return type</dt>
<dd class="field-even"><p><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Module</span></code></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartForConditionalGeneration</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel"><code class="xref py py-class docutils literal notranslate"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel</span></code></a></p>
<p>The BART Model with a language modeling head. Can be used for summarization.</p>
<p>This model inherits from <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code>. Check the superclass documentation for the generic
methods the library implements for all its model (such as downloading or saving, resizing the input embeddings,
pruning heads etc.)</p>
<p>This model is also a PyTorch <a class="reference external" href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general
usage and behavior.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>config</strong> (<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) – Model configuration class with all the parameters of the model.
Initializing with a config file does not load the weights associated with the model, only the configuration.
Check out the <code class="xref py py-meth docutils literal notranslate"><span class="pre">from_pretrained()</span></code> method to load the model weights.</p>
</dd>
</dl>
<dl class="py attribute">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.base_model_prefix">
<span class="sig-name descname"><span class="pre">base_model_prefix</span></span><em class="property"> <span class="pre">=</span> <span class="pre">'model'</span></em><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.base_model_prefix" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py attribute">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.authorized_missing_keys">
<span class="sig-name descname"><span class="pre">authorized_missing_keys</span></span><em class="property"> <span class="pre">=</span> <span class="pre">['final_logits_bias',</span> <span class="pre">'encoder\\.version',</span> <span class="pre">'decoder\\.version']</span></em><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.authorized_missing_keys" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.resize_token_embeddings">
<span class="sig-name descname"><span class="pre">resize_token_embeddings</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">new_num_tokens</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.10)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.nn.modules.sparse.Embedding</span></span></span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.resize_token_embeddings"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.resize_token_embeddings" title="Permalink to this definition"></a></dt>
<dd><p>Resizes input token embeddings matrix of the model if <code class="xref py py-obj docutils literal notranslate"><span class="pre">new_num_tokens</span> <span class="pre">!=</span> <span class="pre">config.vocab_size</span></code>.</p>
<p>Takes care of tying weights embeddings afterwards if the model class has a <code class="xref py py-obj docutils literal notranslate"><span class="pre">tie_weights()</span></code> method.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>new_num_tokens</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">int</span></code></a>, <cite>optional</cite>) – The number of new tokens in the embedding matrix. Increasing the size will add newly initialized
vectors at the end. Reducing the size will remove vectors from the end. If not provided or <a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">None</span></code></a>,
just returns a pointer to the input tokens <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.nn.Embedding</span></code> module of the model wihtout doing
anything.</p>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p>Pointer to the input tokens Embeddings Module of the model.</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p><code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.nn.Embedding</span></code></p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_input_ids</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">unused</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.forward" title="Permalink to this definition"></a></dt>
<dd><p>The <code class="xref py py-class docutils literal notranslate"><span class="pre">BartForConditionalGeneration</span></code> forward method, overrides the <code class="xref py py-func docutils literal notranslate"><span class="pre">__call__()</span></code> special method.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
pre and post processing steps while the latter silently ignores them.</p>
</div>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – <p>Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
it.</p>
<p>Indices can be obtained using <code class="xref py py-class docutils literal notranslate"><span class="pre">BartTokenizer</span></code>.
See <code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.encode()</span></code> and
<code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.__call__()</span></code> for details.</p>
<p><a class="reference external" href="../glossary.html#input-ids">What are input IDs?</a></p>
</p></li>
<li><p><strong>attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.Tensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>, <cite>optional</cite>) – <p>Mask to avoid performing attention on padding token indices.
Mask values selected in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code>:</p>
<ul>
<li><p>1 for tokens that are <strong>not masked</strong>,</p></li>
<li><p>0 for tokens that are <strong>masked</strong>.</p></li>
</ul>
<p><a class="reference external" href="../glossary.html#attention-mask">What are attention masks?</a></p>
</p></li>
<li><p><strong>decoder_input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">target_sequence_length)</span></code>, <cite>optional</cite>) – Provide for translation and summarization training. By default, the model will create this tensor by
shifting the <code class="xref py py-obj docutils literal notranslate"><span class="pre">input_ids</span></code> to the right, following the paper.</p></li>
<li><p><strong>decoder_attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.BoolTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">tgt_seq_len)</span></code>, <cite>optional</cite>) – <p>Default behavior: generate a tensor that ignores pad tokens in <code class="xref py py-obj docutils literal notranslate"><span class="pre">decoder_input_ids</span></code>. Causal mask will
also be used by default.</p>
<p>If you want to change padding behavior, you should read <code class="xref py py-func docutils literal notranslate"><span class="pre">modeling_bart._prepare_decoder_inputs()</span></code> and
modify to your needs. See diagram 1 in <a class="reference external" href="https://arxiv.org/abs/1910.13461">the paper</a> for more
information on the default strategy.</p>
</p></li>
<li><p><strong>encoder_outputs</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>) – Tuple consists of (<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">hidden_states</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">attentions</span></code>)
<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) is a
sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of
the decoder.</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor))</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code> with each tuple having 4 tensors of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length</span> <span class="pre">-</span> <span class="pre">1,</span> <span class="pre">embed_size_per_head)</span></code>) – <p>Contains precomputed key and value hidden-states of the attention blocks. Can be used to speed up decoding.</p>
<p>If <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> are used, the user can optionally input only the last
<code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> (those that don’t have their past key value states given to this model) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">1)</span></code> instead of all <code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>.</p>
</p></li>
<li><p><strong>use_cache</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – If set to <a class="reference external" href="https://docs.python.org/3/library/constants.html#True" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">True</span></code></a>, <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> key value states are returned and can be used to speed up
decoding (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code>).</p></li>
<li><p><strong>output_attentions</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the attentions tensors of all attention layers. See <code class="docutils literal notranslate"><span class="pre">attentions</span></code> under returned
tensors for more detail.</p></li>
<li><p><strong>output_hidden_states</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the hidden states of all layers. See <code class="docutils literal notranslate"><span class="pre">hidden_states</span></code> under returned tensors for
more detail.</p></li>
<li><p><strong>return_dict</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return a <code class="xref py py-class docutils literal notranslate"><span class="pre">ModelOutput</span></code> instead of a plain tuple.</p></li>
<li><p><strong>labels</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>, <cite>optional</cite>) – Labels for computing the masked language modeling loss.
Indices should either be in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">...,</span> <span class="pre">config.vocab_size]</span></code> or -100 (see <code class="docutils literal notranslate"><span class="pre">input_ids</span></code> docstring).
Tokens with indices set to <code class="docutils literal notranslate"><span class="pre">-100</span></code> are ignored (masked), the loss is only computed for the tokens
with labels in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">...,</span> <span class="pre">config.vocab_size]</span></code>.</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><p>A <code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqLMOutput</span></code> (if <code class="docutils literal notranslate"><span class="pre">return_dict=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.return_dict=True</span></code>) or a
tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> comprising various elements depending on the configuration
(<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) and inputs.</p>
<ul>
<li><p><strong>loss</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(1,)</span></code>, <cite>optional</cite>, returned when <code class="xref py py-obj docutils literal notranslate"><span class="pre">labels</span></code> is provided) – Languaged modeling loss.</p></li>
<li><p><strong>logits</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">config.vocab_size)</span></code>) – Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">List[torch.FloatTensor]</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">use_cache=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.use_cache=True</span></code>) – List of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code>, with each tensor of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(2,</span> <span class="pre">batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">embed_size_per_head)</span></code>).</p>
<p>Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be
used (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> input) to speed up sequential decoding.</p>
</li>
<li><p><strong>decoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>decoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
<li><p><strong>encoder_last_hidden_state</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) – Sequence of hidden-states at the output of the last layer of the encoder of the model.</p></li>
<li><p><strong>encoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>encoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
</ul>
<p>Conditional generation example:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="c1"># Mask filling only works for bart-large</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">transformers</span> <span class="kn">import</span> <span class="n">BartTokenizer</span><span class="p">,</span> <span class="n">BartForConditionalGeneration</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">BartTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">TXT</span> <span class="o">=</span> <span class="s2">&quot;My friends are &lt;mask&gt; but they eat too many carbs.&quot;</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">BartForConditionalGeneration</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">input_ids</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">([</span><span class="n">TXT</span><span class="p">],</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">]</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">logits</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span><span class="o">.</span><span class="n">logits</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">masked_index</span> <span class="o">=</span> <span class="p">(</span><span class="n">input_ids</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">mask_token_id</span><span class="p">)</span><span class="o">.</span><span class="n">nonzero</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">probs</span> <span class="o">=</span> <span class="n">logits</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">masked_index</span><span class="p">]</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">values</span><span class="p">,</span> <span class="n">predictions</span> <span class="o">=</span> <span class="n">probs</span><span class="o">.</span><span class="n">topk</span><span class="p">(</span><span class="mi">5</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">predictions</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
<span class="gp">&gt;&gt;&gt; </span><span class="c1"># [&#39;good&#39;, &#39;great&#39;, &#39;all&#39;, &#39;really&#39;, &#39;very&#39;]</span>
</pre></div>
</div>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqLMOutput</span></code> or <code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code></p>
</dd>
</dl>
<p>Summarization example:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">transformers</span> <span class="kn">import</span> <span class="n">BartTokenizer</span><span class="p">,</span> <span class="n">BartForConditionalGeneration</span><span class="p">,</span> <span class="n">BartConfig</span>
<span class="gp">&gt;&gt;&gt; </span><span class="c1"># see ``examples/summarization/bart/run_eval.py`` for a longer example</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">BartForConditionalGeneration</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large-cnn&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">BartTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large-cnn&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">ARTICLE_TO_SUMMARIZE</span> <span class="o">=</span> <span class="s2">&quot;My friends are cool but they eat too many carbs.&quot;</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">([</span><span class="n">ARTICLE_TO_SUMMARIZE</span><span class="p">],</span> <span class="n">max_length</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="c1"># Generate Summary</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">summary_ids</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">],</span> <span class="n">num_beams</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">max_length</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">early_stopping</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="nb">print</span><span class="p">([</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">g</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">clean_up_tokenization_spaces</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> <span class="k">for</span> <span class="n">g</span> <span class="ow">in</span> <span class="n">summary_ids</span><span class="p">])</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.prepare_inputs_for_generation">
<span class="sig-name descname"><span class="pre">prepare_inputs_for_generation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">decoder_input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_outputs</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.prepare_inputs_for_generation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.prepare_inputs_for_generation" title="Permalink to this definition"></a></dt>
<dd><p>Implement in subclasses of <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code> for custom behavior to prepare inputs in the
generate method.</p>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.adjust_logits_during_generation">
<span class="sig-name descname"><span class="pre">adjust_logits_during_generation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">logits</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">cur_len</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_length</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.adjust_logits_during_generation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.adjust_logits_during_generation" title="Permalink to this definition"></a></dt>
<dd><p>Implement in subclasses of <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code> for custom behavior to adjust the logits in
the generate method.</p>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.get_encoder">
<span class="sig-name descname"><span class="pre">get_encoder</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.get_encoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.get_encoder" title="Permalink to this definition"></a></dt>
<dd></dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.get_output_embeddings">
<span class="sig-name descname"><span class="pre">get_output_embeddings</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForConditionalGeneration.get_output_embeddings"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForConditionalGeneration.get_output_embeddings" title="Permalink to this definition"></a></dt>
<dd><p>Returns the model’s output embeddings.</p>
<dl class="field-list simple">
<dt class="field-odd">Returns</dt>
<dd class="field-odd"><p>A torch module mapping hidden states to vocabulary.</p>
</dd>
<dt class="field-even">Return type</dt>
<dd class="field-even"><p><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Module</span></code></p>
</dd>
</dl>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForSequenceClassification">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartForSequenceClassification</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">transformers.configuration_bart.BartConfig</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForSequenceClassification"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForSequenceClassification" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel"><code class="xref py py-class docutils literal notranslate"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel</span></code></a></p>
<p>Bart model with a sequence classification/head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks.</p>
<p>This model inherits from <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code>. Check the superclass documentation for the generic
methods the library implements for all its model (such as downloading or saving, resizing the input embeddings,
pruning heads etc.)</p>
<p>This model is also a PyTorch <a class="reference external" href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general
usage and behavior.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>config</strong> (<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) – Model configuration class with all the parameters of the model.
Initializing with a config file does not load the weights associated with the model, only the configuration.
Check out the <code class="xref py py-meth docutils literal notranslate"><span class="pre">from_pretrained()</span></code> method to load the model weights.</p>
</dd>
</dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForSequenceClassification.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_input_ids</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForSequenceClassification.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForSequenceClassification.forward" title="Permalink to this definition"></a></dt>
<dd><p>The <code class="xref py py-class docutils literal notranslate"><span class="pre">BartForSequenceClassification</span></code> forward method, overrides the <code class="xref py py-func docutils literal notranslate"><span class="pre">__call__()</span></code> special method.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
pre and post processing steps while the latter silently ignores them.</p>
</div>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – <p>Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
it.</p>
<p>Indices can be obtained using <code class="xref py py-class docutils literal notranslate"><span class="pre">BartTokenizer</span></code>.
See <code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.encode()</span></code> and
<code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.__call__()</span></code> for details.</p>
<p><a class="reference external" href="../glossary.html#input-ids">What are input IDs?</a></p>
</p></li>
<li><p><strong>attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.Tensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>, <cite>optional</cite>) – <p>Mask to avoid performing attention on padding token indices.
Mask values selected in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code>:</p>
<ul>
<li><p>1 for tokens that are <strong>not masked</strong>,</p></li>
<li><p>0 for tokens that are <strong>masked</strong>.</p></li>
</ul>
<p><a class="reference external" href="../glossary.html#attention-mask">What are attention masks?</a></p>
</p></li>
<li><p><strong>decoder_input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">target_sequence_length)</span></code>, <cite>optional</cite>) – Provide for translation and summarization training. By default, the model will create this tensor by
shifting the <code class="xref py py-obj docutils literal notranslate"><span class="pre">input_ids</span></code> to the right, following the paper.</p></li>
<li><p><strong>decoder_attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.BoolTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">tgt_seq_len)</span></code>, <cite>optional</cite>) – <p>Default behavior: generate a tensor that ignores pad tokens in <code class="xref py py-obj docutils literal notranslate"><span class="pre">decoder_input_ids</span></code>. Causal mask will
also be used by default.</p>
<p>If you want to change padding behavior, you should read <code class="xref py py-func docutils literal notranslate"><span class="pre">modeling_bart._prepare_decoder_inputs()</span></code> and
modify to your needs. See diagram 1 in <a class="reference external" href="https://arxiv.org/abs/1910.13461">the paper</a> for more
information on the default strategy.</p>
</p></li>
<li><p><strong>encoder_outputs</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>) – Tuple consists of (<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">hidden_states</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">attentions</span></code>)
<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) is a
sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of
the decoder.</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor))</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code> with each tuple having 4 tensors of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length</span> <span class="pre">-</span> <span class="pre">1,</span> <span class="pre">embed_size_per_head)</span></code>) – <p>Contains precomputed key and value hidden-states of the attention blocks. Can be used to speed up decoding.</p>
<p>If <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> are used, the user can optionally input only the last
<code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> (those that don’t have their past key value states given to this model) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">1)</span></code> instead of all <code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>.</p>
</p></li>
<li><p><strong>use_cache</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – If set to <a class="reference external" href="https://docs.python.org/3/library/constants.html#True" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">True</span></code></a>, <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> key value states are returned and can be used to speed up
decoding (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code>).</p></li>
<li><p><strong>output_attentions</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the attentions tensors of all attention layers. See <code class="docutils literal notranslate"><span class="pre">attentions</span></code> under returned
tensors for more detail.</p></li>
<li><p><strong>output_hidden_states</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the hidden states of all layers. See <code class="docutils literal notranslate"><span class="pre">hidden_states</span></code> under returned tensors for
more detail.</p></li>
<li><p><strong>return_dict</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return a <code class="xref py py-class docutils literal notranslate"><span class="pre">ModelOutput</span></code> instead of a plain tuple.</p></li>
<li><p><strong>labels</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,)</span></code>, <cite>optional</cite>) – Labels for computing the sequence classification/regression loss.
Indices should be in <code class="xref py py-obj docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">...,</span> <span class="pre">config.num_labels</span> <span class="pre">-</span> <span class="pre">1]</span></code>.
If <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.num_labels</span> <span class="pre">&gt;</span> <span class="pre">1</span></code> a classification loss is computed (Cross-Entropy).</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><p>A <code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqSequenceClassifierOutput</span></code> (if <code class="docutils literal notranslate"><span class="pre">return_dict=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.return_dict=True</span></code>) or a
tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> comprising various elements depending on the configuration
(<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) and inputs.</p>
<ul>
<li><p><strong>loss</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(1,)</span></code>, <cite>optional</cite>, returned when <code class="xref py py-obj docutils literal notranslate"><span class="pre">label</span></code> is provided) – Classification (or regression if config.num_labels==1) loss.</p></li>
<li><p><strong>logits</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">config.num_labels)</span></code>) – Classification (or regression if config.num_labels==1) scores (before SoftMax).</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">List[torch.FloatTensor]</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">use_cache=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.use_cache=True</span></code>) – List of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code>, with each tensor of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(2,</span> <span class="pre">batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">embed_size_per_head)</span></code>).</p>
<p>Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be
used (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> input) to speed up sequential decoding.</p>
</li>
<li><p><strong>decoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>decoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
<li><p><strong>encoder_last_hidden_state</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) – Sequence of hidden-states at the output of the last layer of the encoder of the model.</p></li>
<li><p><strong>encoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>encoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
</ul>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqSequenceClassifierOutput</span></code> or <code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code></p>
</dd>
</dl>
<p>Example:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">transformers</span> <span class="kn">import</span> <span class="n">BartTokenizer</span><span class="p">,</span> <span class="n">BartForSequenceClassification</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">import</span> <span class="nn">torch</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">BartTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">BartForSequenceClassification</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">,</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="s2">&quot;Hello, my dog is cute&quot;</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="c1"># Batch size 1</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">,</span> <span class="n">labels</span><span class="o">=</span><span class="n">labels</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">loss</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">loss</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">logits</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">logits</span>
</pre></div>
</div>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForQuestionAnswering">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">BartForQuestionAnswering</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForQuestionAnswering"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForQuestionAnswering" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel" title="deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel"><code class="xref py py-class docutils literal notranslate"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.PretrainedBartModel</span></code></a></p>
<p>BART Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layer on top of
the hidden-states output to compute <cite>span start logits</cite> and <cite>span end logits</cite>).</p>
<p>This model inherits from <code class="xref py py-class docutils literal notranslate"><span class="pre">PreTrainedModel</span></code>. Check the superclass documentation for the generic
methods the library implements for all its model (such as downloading or saving, resizing the input embeddings,
pruning heads etc.)</p>
<p>This model is also a PyTorch <a class="reference external" href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general
usage and behavior.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>config</strong> (<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) – Model configuration class with all the parameters of the model.
Initializing with a config file does not load the weights associated with the model, only the configuration.
Check out the <code class="xref py py-meth docutils literal notranslate"><span class="pre">from_pretrained()</span></code> method to load the model weights.</p>
</dd>
</dl>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForQuestionAnswering.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_input_ids</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">decoder_attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">start_positions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">end_positions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#BartForQuestionAnswering.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.BartForQuestionAnswering.forward" title="Permalink to this definition"></a></dt>
<dd><p>The <code class="xref py py-class docutils literal notranslate"><span class="pre">BartForQuestionAnswering</span></code> forward method, overrides the <code class="xref py py-func docutils literal notranslate"><span class="pre">__call__()</span></code> special method.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Although the recipe for forward pass needs to be defined within
this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
instead of this since the former takes care of running the
pre and post processing steps while the latter silently ignores them.</p>
</div>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – <p>Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
it.</p>
<p>Indices can be obtained using <code class="xref py py-class docutils literal notranslate"><span class="pre">BartTokenizer</span></code>.
See <code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.encode()</span></code> and
<code class="xref py py-meth docutils literal notranslate"><span class="pre">transformers.PreTrainedTokenizer.__call__()</span></code> for details.</p>
<p><a class="reference external" href="../glossary.html#input-ids">What are input IDs?</a></p>
</p></li>
<li><p><strong>attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.Tensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>, <cite>optional</cite>) – <p>Mask to avoid performing attention on padding token indices.
Mask values selected in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code>:</p>
<ul>
<li><p>1 for tokens that are <strong>not masked</strong>,</p></li>
<li><p>0 for tokens that are <strong>masked</strong>.</p></li>
</ul>
<p><a class="reference external" href="../glossary.html#attention-mask">What are attention masks?</a></p>
</p></li>
<li><p><strong>decoder_input_ids</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">target_sequence_length)</span></code>, <cite>optional</cite>) – Provide for translation and summarization training. By default, the model will create this tensor by
shifting the <code class="xref py py-obj docutils literal notranslate"><span class="pre">input_ids</span></code> to the right, following the paper.</p></li>
<li><p><strong>decoder_attention_mask</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.BoolTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">tgt_seq_len)</span></code>, <cite>optional</cite>) – <p>Default behavior: generate a tensor that ignores pad tokens in <code class="xref py py-obj docutils literal notranslate"><span class="pre">decoder_input_ids</span></code>. Causal mask will
also be used by default.</p>
<p>If you want to change padding behavior, you should read <code class="xref py py-func docutils literal notranslate"><span class="pre">modeling_bart._prepare_decoder_inputs()</span></code> and
modify to your needs. See diagram 1 in <a class="reference external" href="https://arxiv.org/abs/1910.13461">the paper</a> for more
information on the default strategy.</p>
</p></li>
<li><p><strong>encoder_outputs</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>) – Tuple consists of (<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">hidden_states</span></code>, <cite>optional</cite>: <code class="xref py py-obj docutils literal notranslate"><span class="pre">attentions</span></code>)
<code class="xref py py-obj docutils literal notranslate"><span class="pre">last_hidden_state</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) is a
sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of
the decoder.</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(tuple(torch.FloatTensor))</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code> with each tuple having 4 tensors of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length</span> <span class="pre">-</span> <span class="pre">1,</span> <span class="pre">embed_size_per_head)</span></code>) – <p>Contains precomputed key and value hidden-states of the attention blocks. Can be used to speed up decoding.</p>
<p>If <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> are used, the user can optionally input only the last
<code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> (those that don’t have their past key value states given to this model) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">1)</span></code> instead of all <code class="docutils literal notranslate"><span class="pre">decoder_input_ids</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>.</p>
</p></li>
<li><p><strong>use_cache</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – If set to <a class="reference external" href="https://docs.python.org/3/library/constants.html#True" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">True</span></code></a>, <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> key value states are returned and can be used to speed up
decoding (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code>).</p></li>
<li><p><strong>output_attentions</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the attentions tensors of all attention layers. See <code class="docutils literal notranslate"><span class="pre">attentions</span></code> under returned
tensors for more detail.</p></li>
<li><p><strong>output_hidden_states</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return the hidden states of all layers. See <code class="docutils literal notranslate"><span class="pre">hidden_states</span></code> under returned tensors for
more detail.</p></li>
<li><p><strong>return_dict</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.10)"><code class="xref py py-obj docutils literal notranslate"><span class="pre">bool</span></code></a>, <cite>optional</cite>) – Whether or not to return a <code class="xref py py-class docutils literal notranslate"><span class="pre">ModelOutput</span></code> instead of a plain tuple.</p></li>
<li><p><strong>start_positions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,)</span></code>, <cite>optional</cite>) – Labels for position (index) of the start of the labelled span for computing the token classification loss.
Positions are clamped to the length of the sequence (<cite>sequence_length</cite>).
Position outside of the sequence are not taken into account for computing the loss.</p></li>
<li><p><strong>end_positions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.LongTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,)</span></code>, <cite>optional</cite>) – Labels for position (index) of the end of the labelled span for computing the token classification loss.
Positions are clamped to the length of the sequence (<cite>sequence_length</cite>).
Position outside of the sequence are not taken into account for computing the loss.</p></li>
</ul>
</dd>
<dt class="field-even">Returns</dt>
<dd class="field-even"><p><p>A <code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqQuestionAnsweringModelOutput</span></code> (if <code class="docutils literal notranslate"><span class="pre">return_dict=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.return_dict=True</span></code>) or a
tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> comprising various elements depending on the configuration
(<code class="xref py py-class docutils literal notranslate"><span class="pre">BartConfig</span></code>) and inputs.</p>
<ul>
<li><p><strong>loss</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(1,)</span></code>, <cite>optional</cite>, returned when <code class="xref py py-obj docutils literal notranslate"><span class="pre">labels</span></code> is provided) – Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.</p></li>
<li><p><strong>start_logits</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – Span-start scores (before SoftMax).</p></li>
<li><p><strong>end_logits</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length)</span></code>) – Span-end scores (before SoftMax).</p></li>
<li><p><strong>past_key_values</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">List[torch.FloatTensor]</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">use_cache=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.use_cache=True</span></code>) – List of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of length <code class="xref py py-obj docutils literal notranslate"><span class="pre">config.n_layers</span></code>, with each tensor of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(2,</span> <span class="pre">batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">embed_size_per_head)</span></code>).</p>
<p>Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be
used (see <code class="xref py py-obj docutils literal notranslate"><span class="pre">past_key_values</span></code> input) to speed up sequential decoding.</p>
</li>
<li><p><strong>decoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>decoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
<li><p><strong>encoder_last_hidden_state</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>, <cite>optional</cite>) – Sequence of hidden-states at the output of the last layer of the encoder of the model.</p></li>
<li><p><strong>encoder_hidden_states</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_hidden_states=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_hidden_states=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for the output of the embeddings + one for the output of each layer)
of shape <code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">sequence_length,</span> <span class="pre">hidden_size)</span></code>.</p>
<p>Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.</p>
</li>
<li><p><strong>encoder_attentions</strong> (<code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code>, <cite>optional</cite>, returned when <code class="docutils literal notranslate"><span class="pre">output_attentions=True</span></code> is passed or when <code class="docutils literal notranslate"><span class="pre">config.output_attentions=True</span></code>) – Tuple of <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.FloatTensor</span></code> (one for each layer) of shape
<code class="xref py py-obj docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_heads,</span> <span class="pre">sequence_length,</span> <span class="pre">sequence_length)</span></code>.</p>
<p>Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the
self-attention heads.</p>
</li>
</ul>
</p>
</dd>
<dt class="field-odd">Return type</dt>
<dd class="field-odd"><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Seq2SeqQuestionAnsweringModelOutput</span></code> or <code class="xref py py-obj docutils literal notranslate"><span class="pre">tuple(torch.FloatTensor)</span></code></p>
</dd>
</dl>
<p>Example:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">transformers</span> <span class="kn">import</span> <span class="n">BartTokenizer</span><span class="p">,</span> <span class="n">BartForQuestionAnswering</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">import</span> <span class="nn">torch</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">BartTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">BartForQuestionAnswering</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="s1">&#39;facebook/bart-large&#39;</span><span class="p">,</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">question</span><span class="p">,</span> <span class="n">text</span> <span class="o">=</span> <span class="s2">&quot;Who was Jim Henson?&quot;</span><span class="p">,</span> <span class="s2">&quot;Jim Henson was a nice puppet&quot;</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">question</span><span class="p">,</span> <span class="n">text</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">start_positions</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">1</span><span class="p">])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">end_positions</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">3</span><span class="p">])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">,</span> <span class="n">start_positions</span><span class="o">=</span><span class="n">start_positions</span><span class="p">,</span> <span class="n">end_positions</span><span class="o">=</span><span class="n">end_positions</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">loss</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">loss</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">start_scores</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">start_logits</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">end_scores</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">end_logits</span>
</pre></div>
</div>
</dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.SinusoidalPositionalEmbedding">
<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">deepke.name_entity_recognition.few_shot.models.modeling_bart.</span></span><span class="sig-name descname"><span class="pre">SinusoidalPositionalEmbedding</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_positions</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">embedding_dim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_idx</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#SinusoidalPositionalEmbedding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.SinusoidalPositionalEmbedding" title="Permalink to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.modules.sparse.Embedding</span></code></p>
<p>This module produces sinusoidal positional embeddings of any length.</p>
<dl class="py method">
<dt class="sig sig-object py" id="deepke.name_entity_recognition.few_shot.models.modeling_bart.SinusoidalPositionalEmbedding.forward">
<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/deepke/name_entity_recognition/few_shot/models/modeling_bart.html#SinusoidalPositionalEmbedding.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#deepke.name_entity_recognition.few_shot.models.modeling_bart.SinusoidalPositionalEmbedding.forward" title="Permalink to this definition"></a></dt>
<dd><p>Input is expected to be of size [bsz x seqlen].</p>
</dd></dl>
</dd></dl>
</section>
</section>
</div>
</div>
<footer><div class="rst-footer-buttons" role="navigation" aria-label="Footer">
<a href="deepke.name_entity_recognition.few_shot.html" class="btn btn-neutral float-left" title="Few Shot" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left" aria-hidden="true"></span> Previous</a>
<a href="deepke.name_entity_recognition.few_shot.module.html" class="btn btn-neutral float-right" title="Module" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right" aria-hidden="true"></span></a>
</div>
<hr/>
<div role="contentinfo">
<p>&#169; Copyright 2021, ZJUNLP.</p>
</div>
Built with <a href="https://www.sphinx-doc.org/">Sphinx</a> using a
<a href="https://github.com/readthedocs/sphinx_rtd_theme">theme</a>
provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script>
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink