715 lines
94 KiB
HTML
715 lines
94 KiB
HTML
<!DOCTYPE html>
|
|
<html class="writer-html5" lang="en" data-content_root="../../../">
|
|
<head>
|
|
<meta charset="utf-8" />
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
|
<title>quapy.classification.neural — QuaPy: A Python-based open-source framework for quantification 0.1.8 documentation</title>
|
|
<link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=92fd9be5" />
|
|
<link rel="stylesheet" type="text/css" href="../../../_static/css/theme.css?v=19f00094" />
|
|
|
|
|
|
<!--[if lt IE 9]>
|
|
<script src="../../../_static/js/html5shiv.min.js"></script>
|
|
<![endif]-->
|
|
|
|
<script src="../../../_static/jquery.js?v=5d32c60e"></script>
|
|
<script src="../../../_static/_sphinx_javascript_frameworks_compat.js?v=2cd50e6c"></script>
|
|
<script src="../../../_static/documentation_options.js?v=22607128"></script>
|
|
<script src="../../../_static/doctools.js?v=9a2dae69"></script>
|
|
<script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
|
|
<script src="../../../_static/js/theme.js"></script>
|
|
<link rel="index" title="Index" href="../../../genindex.html" />
|
|
<link rel="search" title="Search" href="../../../search.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">
|
|
QuaPy: A Python-based open-source framework for quantification
|
|
</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" aria-label="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">
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../modules.html">quapy</a></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">QuaPy: A Python-based open-source framework for quantification</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" aria-label="Home"></a></li>
|
|
<li class="breadcrumb-item"><a href="../../index.html">Module code</a></li>
|
|
<li class="breadcrumb-item active">quapy.classification.neural</li>
|
|
<li class="wy-breadcrumbs-aside">
|
|
</li>
|
|
</ul>
|
|
<hr/>
|
|
</div>
|
|
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
|
<div itemprop="articleBody">
|
|
|
|
<h1>Source code for quapy.classification.neural</h1><div class="highlight"><pre>
|
|
<span></span><span class="kn">import</span> <span class="nn">os</span>
|
|
<span class="kn">from</span> <span class="nn">abc</span> <span class="kn">import</span> <span class="n">ABCMeta</span><span class="p">,</span> <span class="n">abstractmethod</span>
|
|
<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
|
|
|
|
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
|
<span class="kn">import</span> <span class="nn">torch</span>
|
|
<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
|
|
<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
|
|
<span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <span class="n">accuracy_score</span><span class="p">,</span> <span class="n">f1_score</span>
|
|
<span class="kn">from</span> <span class="nn">torch.nn.utils.rnn</span> <span class="kn">import</span> <span class="n">pad_sequence</span>
|
|
<span class="kn">from</span> <span class="nn">tqdm</span> <span class="kn">import</span> <span class="n">tqdm</span>
|
|
|
|
<span class="kn">import</span> <span class="nn">quapy</span> <span class="k">as</span> <span class="nn">qp</span>
|
|
<span class="kn">from</span> <span class="nn">quapy.data</span> <span class="kn">import</span> <span class="n">LabelledCollection</span>
|
|
<span class="kn">from</span> <span class="nn">quapy.util</span> <span class="kn">import</span> <span class="n">EarlyStop</span>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">NeuralClassifierTrainer</span><span class="p">:</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Trains a neural network for text classification.</span>
|
|
|
|
<span class="sd"> :param net: an instance of `TextClassifierNet` implementing the forward pass</span>
|
|
<span class="sd"> :param lr: learning rate (default 1e-3)</span>
|
|
<span class="sd"> :param weight_decay: weight decay (default 0)</span>
|
|
<span class="sd"> :param patience: number of epochs that do not show any improvement in validation</span>
|
|
<span class="sd"> to wait before applying early stop (default 10)</span>
|
|
<span class="sd"> :param epochs: maximum number of training epochs (default 200)</span>
|
|
<span class="sd"> :param batch_size: batch size for training (default 64)</span>
|
|
<span class="sd"> :param batch_size_test: batch size for test (default 512)</span>
|
|
<span class="sd"> :param padding_length: maximum number of tokens to consider in a document (default 300)</span>
|
|
<span class="sd"> :param device: specify 'cpu' (default) or 'cuda' for enabling gpu</span>
|
|
<span class="sd"> :param checkpointpath: where to store the parameters of the best model found so far</span>
|
|
<span class="sd"> according to the evaluation in the held-out validation split (default '../checkpoint/classifier_net.dat')</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
|
<span class="n">net</span><span class="p">:</span> <span class="s1">'TextClassifierNet'</span><span class="p">,</span>
|
|
<span class="n">lr</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
|
|
<span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
|
|
<span class="n">patience</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span>
|
|
<span class="n">epochs</span><span class="o">=</span><span class="mi">200</span><span class="p">,</span>
|
|
<span class="n">batch_size</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span>
|
|
<span class="n">batch_size_test</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
|
|
<span class="n">padding_length</span><span class="o">=</span><span class="mi">300</span><span class="p">,</span>
|
|
<span class="n">device</span><span class="o">=</span><span class="s1">'cuda'</span><span class="p">,</span>
|
|
<span class="n">checkpointpath</span><span class="o">=</span><span class="s1">'../checkpoint/classifier_net.dat'</span><span class="p">):</span>
|
|
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
|
|
<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">net</span><span class="p">,</span> <span class="n">TextClassifierNet</span><span class="p">),</span> <span class="sa">f</span><span class="s1">'net is not an instance of </span><span class="si">{</span><span class="n">TextClassifierNet</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s1">'</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span> <span class="o">=</span> <span class="n">net</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">vocabulary_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span><span class="o">=</span><span class="p">{</span>
|
|
<span class="s1">'lr'</span><span class="p">:</span> <span class="n">lr</span><span class="p">,</span>
|
|
<span class="s1">'weight_decay'</span><span class="p">:</span> <span class="n">weight_decay</span><span class="p">,</span>
|
|
<span class="s1">'patience'</span><span class="p">:</span> <span class="n">patience</span><span class="p">,</span>
|
|
<span class="s1">'epochs'</span><span class="p">:</span> <span class="n">epochs</span><span class="p">,</span>
|
|
<span class="s1">'batch_size'</span><span class="p">:</span> <span class="n">batch_size</span><span class="p">,</span>
|
|
<span class="s1">'batch_size_test'</span><span class="p">:</span> <span class="n">batch_size_test</span><span class="p">,</span>
|
|
<span class="s1">'padding_length'</span><span class="p">:</span> <span class="n">padding_length</span><span class="p">,</span>
|
|
<span class="s1">'device'</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
|
|
<span class="p">}</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">learner_hyperparams</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">get_params</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">checkpointpath</span> <span class="o">=</span> <span class="n">checkpointpath</span>
|
|
|
|
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s1">'[NeuralNetwork running on </span><span class="si">{</span><span class="n">device</span><span class="si">}</span><span class="s1">]'</span><span class="p">)</span>
|
|
<span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">Path</span><span class="p">(</span><span class="n">checkpointpath</span><span class="p">)</span><span class="o">.</span><span class="n">parent</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.reset_net_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.reset_net_params">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">reset_net_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">vocab_size</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Reinitialize the network parameters</span>
|
|
|
|
<span class="sd"> :param vocab_size: the size of the vocabulary</span>
|
|
<span class="sd"> :param n_classes: the number of target classes</span>
|
|
<span class="sd"> """</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="vm">__class__</span><span class="p">(</span><span class="n">vocab_size</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">,</span> <span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">learner_hyperparams</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span><span class="p">[</span><span class="s1">'device'</span><span class="p">])</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">xavier_uniform</span><span class="p">()</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.get_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.get_params">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Get hyper-parameters for this estimator</span>
|
|
|
|
<span class="sd"> :return: a dictionary with parameter names mapped to their values</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="p">{</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">get_params</span><span class="p">(),</span> <span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span><span class="p">}</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.set_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.set_params">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">set_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Set the parameters of this trainer and the learner it is training.</span>
|
|
<span class="sd"> In this current version, parameter names for the trainer and learner should</span>
|
|
<span class="sd"> be disjoint.</span>
|
|
|
|
<span class="sd"> :param params: a `**kwargs` dictionary with the parameters</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">trainer_hyperparams</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span>
|
|
<span class="n">learner_hyperparams</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">get_params</span><span class="p">()</span>
|
|
<span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">val</span> <span class="ow">in</span> <span class="n">params</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
|
|
<span class="k">if</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">trainer_hyperparams</span> <span class="ow">and</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">learner_hyperparams</span><span class="p">:</span>
|
|
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">'the use of parameter </span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s1"> is ambiguous since it can refer to '</span>
|
|
<span class="sa">f</span><span class="s1">'a parameters of the Trainer or the learner </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s1">'</span><span class="p">)</span>
|
|
<span class="k">elif</span> <span class="n">key</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">trainer_hyperparams</span> <span class="ow">and</span> <span class="n">key</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">learner_hyperparams</span><span class="p">:</span>
|
|
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">'parameter </span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s1"> is not valid'</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">trainer_hyperparams</span><span class="p">:</span>
|
|
<span class="n">trainer_hyperparams</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">val</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">learner_hyperparams</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">val</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span> <span class="o">=</span> <span class="n">trainer_hyperparams</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">learner_hyperparams</span> <span class="o">=</span> <span class="n">learner_hyperparams</span> </div>
|
|
|
|
|
|
<span class="nd">@property</span>
|
|
<span class="k">def</span> <span class="nf">device</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">""" Gets the device in which the network is allocated</span>
|
|
|
|
<span class="sd"> :return: device</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="nb">next</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">parameters</span><span class="p">())</span><span class="o">.</span><span class="n">device</span>
|
|
|
|
<span class="k">def</span> <span class="nf">_train_epoch</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">,</span> <span class="n">status</span><span class="p">,</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
|
|
<span class="n">criterion</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">CrossEntropyLoss</span><span class="p">()</span>
|
|
<span class="n">losses</span><span class="p">,</span> <span class="n">predictions</span><span class="p">,</span> <span class="n">true_labels</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[]</span>
|
|
<span class="k">for</span> <span class="n">xi</span><span class="p">,</span> <span class="n">yi</span> <span class="ow">in</span> <span class="n">data</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">zero_grad</span><span class="p">()</span>
|
|
<span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="n">xi</span><span class="p">)</span>
|
|
<span class="n">loss</span> <span class="o">=</span> <span class="n">criterion</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">yi</span><span class="p">)</span>
|
|
<span class="n">loss</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
|
|
<span class="n">losses</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">loss</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
|
|
<span class="n">preds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
|
|
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"loss"</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">losses</span><span class="p">)</span>
|
|
<span class="n">predictions</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">preds</span><span class="o">.</span><span class="n">tolist</span><span class="p">())</span>
|
|
<span class="n">true_labels</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">yi</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">.</span><span class="n">tolist</span><span class="p">())</span>
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"acc"</span><span class="p">]</span> <span class="o">=</span> <span class="n">accuracy_score</span><span class="p">(</span><span class="n">true_labels</span><span class="p">,</span> <span class="n">predictions</span><span class="p">)</span>
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"f1"</span><span class="p">]</span> <span class="o">=</span> <span class="n">f1_score</span><span class="p">(</span><span class="n">true_labels</span><span class="p">,</span> <span class="n">predictions</span><span class="p">,</span> <span class="n">average</span><span class="o">=</span><span class="s1">'macro'</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">__update_progress_bar</span><span class="p">(</span><span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">_test_epoch</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">,</span> <span class="n">status</span><span class="p">,</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
|
|
<span class="n">criterion</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">CrossEntropyLoss</span><span class="p">()</span>
|
|
<span class="n">losses</span><span class="p">,</span> <span class="n">predictions</span><span class="p">,</span> <span class="n">true_labels</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[]</span>
|
|
<span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
|
|
<span class="k">for</span> <span class="n">xi</span><span class="p">,</span> <span class="n">yi</span> <span class="ow">in</span> <span class="n">data</span><span class="p">:</span>
|
|
<span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="n">xi</span><span class="p">)</span>
|
|
<span class="n">loss</span> <span class="o">=</span> <span class="n">criterion</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">yi</span><span class="p">)</span>
|
|
<span class="n">losses</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">loss</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
|
|
<span class="n">preds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
|
|
<span class="n">predictions</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">preds</span><span class="o">.</span><span class="n">tolist</span><span class="p">())</span>
|
|
<span class="n">true_labels</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">yi</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">.</span><span class="n">tolist</span><span class="p">())</span>
|
|
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"loss"</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">losses</span><span class="p">)</span>
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"acc"</span><span class="p">]</span> <span class="o">=</span> <span class="n">accuracy_score</span><span class="p">(</span><span class="n">true_labels</span><span class="p">,</span> <span class="n">predictions</span><span class="p">)</span>
|
|
<span class="n">status</span><span class="p">[</span><span class="s2">"f1"</span><span class="p">]</span> <span class="o">=</span> <span class="n">f1_score</span><span class="p">(</span><span class="n">true_labels</span><span class="p">,</span> <span class="n">predictions</span><span class="p">,</span> <span class="n">average</span><span class="o">=</span><span class="s1">'macro'</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">__update_progress_bar</span><span class="p">(</span><span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">__update_progress_bar</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">):</span>
|
|
<span class="n">pbar</span><span class="o">.</span><span class="n">set_description</span><span class="p">(</span><span class="sa">f</span><span class="s1">'[</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s1">] training epoch=</span><span class="si">{</span><span class="n">epoch</span><span class="si">}</span><span class="s1"> '</span>
|
|
<span class="sa">f</span><span class="s1">'tr-loss=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"tr"</span><span class="p">][</span><span class="s2">"loss"</span><span class="p">]</span><span class="si">:</span><span class="s1">.5f</span><span class="si">}</span><span class="s1"> '</span>
|
|
<span class="sa">f</span><span class="s1">'tr-acc=</span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"tr"</span><span class="p">][</span><span class="s2">"acc"</span><span class="p">]</span><span class="si">:</span><span class="s1">.2f</span><span class="si">}</span><span class="s1">% '</span>
|
|
<span class="sa">f</span><span class="s1">'tr-macroF1=</span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"tr"</span><span class="p">][</span><span class="s2">"f1"</span><span class="p">]</span><span class="si">:</span><span class="s1">.2f</span><span class="si">}</span><span class="s1">% '</span>
|
|
<span class="sa">f</span><span class="s1">'patience=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="o">.</span><span class="n">patience</span><span class="si">}</span><span class="s1">/</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="o">.</span><span class="n">PATIENCE_LIMIT</span><span class="si">}</span><span class="s1"> '</span>
|
|
<span class="sa">f</span><span class="s1">'val-loss=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"va"</span><span class="p">][</span><span class="s2">"loss"</span><span class="p">]</span><span class="si">:</span><span class="s1">.5f</span><span class="si">}</span><span class="s1"> '</span>
|
|
<span class="sa">f</span><span class="s1">'val-acc=</span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"va"</span><span class="p">][</span><span class="s2">"acc"</span><span class="p">]</span><span class="si">:</span><span class="s1">.2f</span><span class="si">}</span><span class="s1">% '</span>
|
|
<span class="sa">f</span><span class="s1">'macroF1=</span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s2">"va"</span><span class="p">][</span><span class="s2">"f1"</span><span class="p">]</span><span class="si">:</span><span class="s1">.2f</span><span class="si">}</span><span class="s1">%'</span><span class="p">)</span>
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.fit">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.fit">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">fit</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">val_split</span><span class="o">=</span><span class="mf">0.3</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Fits the model according to the given training data.</span>
|
|
|
|
<span class="sd"> :param instances: list of lists of indexed tokens</span>
|
|
<span class="sd"> :param labels: array-like of shape `(n_samples, n_classes)` with the class labels</span>
|
|
<span class="sd"> :param val_split: proportion of training documents to be taken as the validation set (default 0.3)</span>
|
|
<span class="sd"> :return:</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">train</span><span class="p">,</span> <span class="n">val</span> <span class="o">=</span> <span class="n">LabelledCollection</span><span class="p">(</span><span class="n">instances</span><span class="p">,</span> <span class="n">labels</span><span class="p">)</span><span class="o">.</span><span class="n">split_stratified</span><span class="p">(</span><span class="mi">1</span><span class="o">-</span><span class="n">val_split</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">classes_</span> <span class="o">=</span> <span class="n">train</span><span class="o">.</span><span class="n">classes_</span>
|
|
<span class="n">opt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span>
|
|
<span class="n">checkpoint</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">checkpointpath</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">reset_net_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">,</span> <span class="n">train</span><span class="o">.</span><span class="n">n_classes</span><span class="p">)</span>
|
|
|
|
<span class="n">train_generator</span> <span class="o">=</span> <span class="n">TorchDataset</span><span class="p">(</span><span class="n">train</span><span class="o">.</span><span class="n">instances</span><span class="p">,</span> <span class="n">train</span><span class="o">.</span><span class="n">labels</span><span class="p">)</span><span class="o">.</span><span class="n">asDataloader</span><span class="p">(</span>
|
|
<span class="n">opt</span><span class="p">[</span><span class="s1">'batch_size'</span><span class="p">],</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">pad_length</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'padding_length'</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'device'</span><span class="p">])</span>
|
|
<span class="n">valid_generator</span> <span class="o">=</span> <span class="n">TorchDataset</span><span class="p">(</span><span class="n">val</span><span class="o">.</span><span class="n">instances</span><span class="p">,</span> <span class="n">val</span><span class="o">.</span><span class="n">labels</span><span class="p">)</span><span class="o">.</span><span class="n">asDataloader</span><span class="p">(</span>
|
|
<span class="n">opt</span><span class="p">[</span><span class="s1">'batch_size_test'</span><span class="p">],</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">pad_length</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'padding_length'</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'device'</span><span class="p">])</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">status</span> <span class="o">=</span> <span class="p">{</span><span class="s1">'tr'</span><span class="p">:</span> <span class="p">{</span><span class="s1">'loss'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="s1">'acc'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="s1">'f1'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">},</span>
|
|
<span class="s1">'va'</span><span class="p">:</span> <span class="p">{</span><span class="s1">'loss'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="s1">'acc'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="s1">'f1'</span><span class="p">:</span> <span class="o">-</span><span class="mi">1</span><span class="p">}}</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">optim</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">parameters</span><span class="p">(),</span> <span class="n">lr</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'lr'</span><span class="p">],</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'weight_decay'</span><span class="p">])</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span> <span class="o">=</span> <span class="n">EarlyStop</span><span class="p">(</span><span class="n">opt</span><span class="p">[</span><span class="s1">'patience'</span><span class="p">],</span> <span class="n">lower_is_better</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
|
|
|
|
<span class="k">with</span> <span class="n">tqdm</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">opt</span><span class="p">[</span><span class="s1">'epochs'</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">))</span> <span class="k">as</span> <span class="n">pbar</span><span class="p">:</span>
|
|
<span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="n">pbar</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">_train_epoch</span><span class="p">(</span><span class="n">train_generator</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s1">'tr'</span><span class="p">],</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">_test_epoch</span><span class="p">(</span><span class="n">valid_generator</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s1">'va'</span><span class="p">],</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s1">'va'</span><span class="p">][</span><span class="s1">'f1'</span><span class="p">],</span> <span class="n">epoch</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="o">.</span><span class="n">IMPROVED</span><span class="p">:</span>
|
|
<span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">checkpoint</span><span class="p">)</span>
|
|
<span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="o">.</span><span class="n">STOP</span><span class="p">:</span>
|
|
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s1">'training ended by patience exhasted; loading best model parameters in </span><span class="si">{</span><span class="n">checkpoint</span><span class="si">}</span><span class="s1"> '</span>
|
|
<span class="sa">f</span><span class="s1">'for epoch </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">early_stop</span><span class="o">.</span><span class="n">best_epoch</span><span class="si">}</span><span class="s1">'</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">))</span>
|
|
<span class="k">break</span>
|
|
|
|
<span class="nb">print</span><span class="p">(</span><span class="s1">'performing one training pass over the validation set...'</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">_train_epoch</span><span class="p">(</span><span class="n">valid_generator</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">status</span><span class="p">[</span><span class="s1">'tr'</span><span class="p">],</span> <span class="n">pbar</span><span class="p">,</span> <span class="n">epoch</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
|
<span class="nb">print</span><span class="p">(</span><span class="s1">'[done]'</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="bp">self</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.predict">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.predict">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">predict</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Predicts labels for the instances</span>
|
|
|
|
<span class="sd"> :param instances: list of lists of indexed tokens</span>
|
|
<span class="sd"> :return: a `numpy` array of length `n` containing the label predictions, where `n` is the number of</span>
|
|
<span class="sd"> instances in `X`</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">instances</span><span class="p">),</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.predict_proba">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.predict_proba">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">predict_proba</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Predicts posterior probabilities for the instances</span>
|
|
|
|
<span class="sd"> :param X: array-like of shape `(n_samples, n_features)` instances to classify</span>
|
|
<span class="sd"> :return: array-like of shape `(n_samples, n_classes)` with the posterior probabilities</span>
|
|
<span class="sd"> """</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
|
|
<span class="n">opt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span>
|
|
<span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
|
|
<span class="n">posteriors</span> <span class="o">=</span> <span class="p">[]</span>
|
|
<span class="k">for</span> <span class="n">xi</span> <span class="ow">in</span> <span class="n">TorchDataset</span><span class="p">(</span><span class="n">instances</span><span class="p">)</span><span class="o">.</span><span class="n">asDataloader</span><span class="p">(</span>
|
|
<span class="n">opt</span><span class="p">[</span><span class="s1">'batch_size_test'</span><span class="p">],</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">pad_length</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'padding_length'</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'device'</span><span class="p">]):</span>
|
|
<span class="n">posteriors</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">xi</span><span class="p">))</span>
|
|
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">posteriors</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="NeuralClassifierTrainer.transform">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.NeuralClassifierTrainer.transform">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Returns the embeddings of the instances</span>
|
|
|
|
<span class="sd"> :param instances: list of lists of indexed tokens</span>
|
|
<span class="sd"> :return: array-like of shape `(n_samples, embed_size)` with the embedded instances,</span>
|
|
<span class="sd"> where `embed_size` is defined by the classification network</span>
|
|
<span class="sd"> """</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
|
|
<span class="n">embeddings</span> <span class="o">=</span> <span class="p">[]</span>
|
|
<span class="n">opt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">trainer_hyperparams</span>
|
|
<span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
|
|
<span class="k">for</span> <span class="n">xi</span> <span class="ow">in</span> <span class="n">TorchDataset</span><span class="p">(</span><span class="n">instances</span><span class="p">)</span><span class="o">.</span><span class="n">asDataloader</span><span class="p">(</span>
|
|
<span class="n">opt</span><span class="p">[</span><span class="s1">'batch_size_test'</span><span class="p">],</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">pad_length</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'padding_length'</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">opt</span><span class="p">[</span><span class="s1">'device'</span><span class="p">]):</span>
|
|
<span class="n">embeddings</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">net</span><span class="o">.</span><span class="n">document_embedding</span><span class="p">(</span><span class="n">xi</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">())</span>
|
|
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">embeddings</span><span class="p">)</span></div>
|
|
</div>
|
|
|
|
|
|
|
|
<div class="viewcode-block" id="TorchDataset">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TorchDataset">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">TorchDataset</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Transforms labelled instances into a Torch's :class:`torch.utils.data.DataLoader` object</span>
|
|
|
|
<span class="sd"> :param instances: list of lists of indexed tokens</span>
|
|
<span class="sd"> :param labels: array-like of shape `(n_samples, n_classes)` with the class labels</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">,</span> <span class="n">labels</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">instances</span> <span class="o">=</span> <span class="n">instances</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">labels</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">instances</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">):</span>
|
|
<span class="k">return</span> <span class="p">{</span><span class="s1">'doc'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">instances</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="s1">'label'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span><span class="p">}</span>
|
|
|
|
<div class="viewcode-block" id="TorchDataset.asDataloader">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TorchDataset.asDataloader">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">asDataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch_size</span><span class="p">,</span> <span class="n">shuffle</span><span class="p">,</span> <span class="n">pad_length</span><span class="p">,</span> <span class="n">device</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Converts the labelled collection into a Torch DataLoader with dynamic padding for</span>
|
|
<span class="sd"> the batch</span>
|
|
|
|
<span class="sd"> :param batch_size: batch size</span>
|
|
<span class="sd"> :param shuffle: whether or not to shuffle instances</span>
|
|
<span class="sd"> :param pad_length: the maximum length for the list of tokens (dynamic padding is</span>
|
|
<span class="sd"> applied, meaning that if the longest document in the batch is shorter than</span>
|
|
<span class="sd"> `pad_length`, then the batch is padded up to its length, and not to `pad_length`.</span>
|
|
<span class="sd"> :param device: whether to allocate tensors in cpu or in cuda</span>
|
|
<span class="sd"> :return: a :class:`torch.utils.data.DataLoader` object</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">def</span> <span class="nf">collate</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span>
|
|
<span class="n">data</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">item</span><span class="p">[</span><span class="s1">'doc'</span><span class="p">][:</span><span class="n">pad_length</span><span class="p">])</span> <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
|
|
<span class="n">data</span> <span class="o">=</span> <span class="n">pad_sequence</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">padding_value</span><span class="o">=</span><span class="n">qp</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s1">'PAD_INDEX'</span><span class="p">])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
|
|
<span class="n">targets</span> <span class="o">=</span> <span class="p">[</span><span class="n">item</span><span class="p">[</span><span class="s1">'label'</span><span class="p">]</span> <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
|
|
<span class="k">if</span> <span class="n">targets</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="k">return</span> <span class="n">data</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">targets</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="p">[</span><span class="n">data</span><span class="p">,</span> <span class="n">targets</span><span class="p">]</span>
|
|
|
|
<span class="n">torchDataset</span> <span class="o">=</span> <span class="n">TorchDataset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">instances</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">DataLoader</span><span class="p">(</span><span class="n">torchDataset</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="n">shuffle</span><span class="p">,</span> <span class="n">collate_fn</span><span class="o">=</span><span class="n">collate</span><span class="p">)</span></div>
|
|
</div>
|
|
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">TextClassifierNet</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">metaclass</span><span class="o">=</span><span class="n">ABCMeta</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Abstract Text classifier (`torch.nn.Module`)</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.document_embedding">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.document_embedding">[docs]</a>
|
|
<span class="nd">@abstractmethod</span>
|
|
<span class="k">def</span> <span class="nf">document_embedding</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Embeds documents (i.e., performs the forward pass up to the</span>
|
|
<span class="sd"> next-to-last layer).</span>
|
|
|
|
<span class="sd"> :param x: a batch of instances, typically generated by a torch's `DataLoader`</span>
|
|
<span class="sd"> instance (see :class:`quapy.classification.neural.TorchDataset`)</span>
|
|
<span class="sd"> :return: a torch tensor of shape `(n_samples, n_dimensions)`, where</span>
|
|
<span class="sd"> `n_samples` is the number of documents, and `n_dimensions` is the</span>
|
|
<span class="sd"> dimensionality of the embedding</span>
|
|
<span class="sd"> """</span>
|
|
<span class="o">...</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.forward">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.forward">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Performs the forward pass.</span>
|
|
|
|
<span class="sd"> :param x: a batch of instances, typically generated by a torch's `DataLoader`</span>
|
|
<span class="sd"> instance (see :class:`quapy.classification.neural.TorchDataset`)</span>
|
|
<span class="sd"> :return: a tensor of shape `(n_instances, n_classes)` with the decision scores</span>
|
|
<span class="sd"> for each of the instances and classes</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">doc_embedded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">document_embedding</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">output</span><span class="p">(</span><span class="n">doc_embedded</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.dimensions">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.dimensions">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">dimensions</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Gets the number of dimensions of the embedding space</span>
|
|
|
|
<span class="sd"> :return: integer</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">dim</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.predict_proba">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.predict_proba">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">predict_proba</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Predicts posterior probabilities for the instances in `x`</span>
|
|
|
|
<span class="sd"> :param x: a torch tensor of indexed tokens with shape `(n_instances, pad_length)`</span>
|
|
<span class="sd"> where `n_instances` is the number of instances in the batch, and `pad_length`</span>
|
|
<span class="sd"> is length of the pad in the batch</span>
|
|
<span class="sd"> :return: array-like of shape `(n_samples, n_classes)` with the posterior probabilities</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.xavier_uniform">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.xavier_uniform">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">xavier_uniform</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Performs Xavier initialization of the network parameters</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
|
|
<span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">></span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">:</span>
|
|
<span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">xavier_uniform_</span><span class="p">(</span><span class="n">p</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="TextClassifierNet.get_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.TextClassifierNet.get_params">[docs]</a>
|
|
<span class="nd">@abstractmethod</span>
|
|
<span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Get hyper-parameters for this estimator</span>
|
|
|
|
<span class="sd"> :return: a dictionary with parameter names mapped to their values</span>
|
|
<span class="sd"> """</span>
|
|
<span class="o">...</span></div>
|
|
|
|
|
|
<span class="nd">@property</span>
|
|
<span class="k">def</span> <span class="nf">vocabulary_size</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Return the size of the vocabulary</span>
|
|
|
|
<span class="sd"> :return: integer</span>
|
|
<span class="sd"> """</span>
|
|
<span class="o">...</span></div>
|
|
|
|
|
|
|
|
<div class="viewcode-block" id="LSTMnet">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.LSTMnet">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">LSTMnet</span><span class="p">(</span><span class="n">TextClassifierNet</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> An implementation of :class:`quapy.classification.neural.TextClassifierNet` based on</span>
|
|
<span class="sd"> Long Short Term Memory networks.</span>
|
|
|
|
<span class="sd"> :param vocabulary_size: the size of the vocabulary</span>
|
|
<span class="sd"> :param n_classes: number of target classes</span>
|
|
<span class="sd"> :param embedding_size: the dimensionality of the word embeddings space (default 100)</span>
|
|
<span class="sd"> :param hidden_size: the dimensionality of the hidden space (default 256)</span>
|
|
<span class="sd"> :param repr_size: the dimensionality of the document embeddings space (default 100)</span>
|
|
<span class="sd"> :param lstm_class_nlayers: number of LSTM layers (default 1)</span>
|
|
<span class="sd"> :param drop_p: drop probability for dropout (default 0.5)</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">vocabulary_size</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">,</span> <span class="n">embedding_size</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> <span class="n">hidden_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">repr_size</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> <span class="n">lstm_class_nlayers</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
|
|
<span class="n">drop_p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
|
|
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">vocabulary_size_</span> <span class="o">=</span> <span class="n">vocabulary_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">n_classes</span> <span class="o">=</span> <span class="n">n_classes</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">hyperparams</span><span class="o">=</span><span class="p">{</span>
|
|
<span class="s1">'embedding_size'</span><span class="p">:</span> <span class="n">embedding_size</span><span class="p">,</span>
|
|
<span class="s1">'hidden_size'</span><span class="p">:</span> <span class="n">hidden_size</span><span class="p">,</span>
|
|
<span class="s1">'repr_size'</span><span class="p">:</span> <span class="n">repr_size</span><span class="p">,</span>
|
|
<span class="s1">'lstm_class_nlayers'</span><span class="p">:</span> <span class="n">lstm_class_nlayers</span><span class="p">,</span>
|
|
<span class="s1">'drop_p'</span><span class="p">:</span> <span class="n">drop_p</span>
|
|
<span class="p">}</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">word_embedding</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Embedding</span><span class="p">(</span><span class="n">vocabulary_size</span><span class="p">,</span> <span class="n">embedding_size</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">lstm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">LSTM</span><span class="p">(</span><span class="n">embedding_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">lstm_class_nlayers</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">drop_p</span><span class="p">,</span> <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">drop_p</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">dim</span> <span class="o">=</span> <span class="n">repr_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">doc_embedder</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">hidden_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">dim</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dim</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">__init_hidden</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">set_size</span><span class="p">):</span>
|
|
<span class="n">opt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hyperparams</span>
|
|
<span class="n">var_hidden</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">opt</span><span class="p">[</span><span class="s1">'lstm_class_nlayers'</span><span class="p">],</span> <span class="n">set_size</span><span class="p">,</span> <span class="n">opt</span><span class="p">[</span><span class="s1">'hidden_size'</span><span class="p">])</span>
|
|
<span class="n">var_cell</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">opt</span><span class="p">[</span><span class="s1">'lstm_class_nlayers'</span><span class="p">],</span> <span class="n">set_size</span><span class="p">,</span> <span class="n">opt</span><span class="p">[</span><span class="s1">'hidden_size'</span><span class="p">])</span>
|
|
<span class="k">if</span> <span class="nb">next</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lstm</span><span class="o">.</span><span class="n">parameters</span><span class="p">())</span><span class="o">.</span><span class="n">is_cuda</span><span class="p">:</span>
|
|
<span class="n">var_hidden</span><span class="p">,</span> <span class="n">var_cell</span> <span class="o">=</span> <span class="n">var_hidden</span><span class="o">.</span><span class="n">cuda</span><span class="p">(),</span> <span class="n">var_cell</span><span class="o">.</span><span class="n">cuda</span><span class="p">()</span>
|
|
<span class="k">return</span> <span class="n">var_hidden</span><span class="p">,</span> <span class="n">var_cell</span>
|
|
|
|
<div class="viewcode-block" id="LSTMnet.document_embedding">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.LSTMnet.document_embedding">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">document_embedding</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Embeds documents (i.e., performs the forward pass up to the</span>
|
|
<span class="sd"> next-to-last layer).</span>
|
|
|
|
<span class="sd"> :param x: a batch of instances, typically generated by a torch's `DataLoader`</span>
|
|
<span class="sd"> instance (see :class:`quapy.classification.neural.TorchDataset`)</span>
|
|
<span class="sd"> :return: a torch tensor of shape `(n_samples, n_dimensions)`, where</span>
|
|
<span class="sd"> `n_samples` is the number of documents, and `n_dimensions` is the</span>
|
|
<span class="sd"> dimensionality of the embedding</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">embedded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">word_embedding</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
|
|
<span class="n">rnn_output</span><span class="p">,</span> <span class="n">rnn_hidden</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">lstm</span><span class="p">(</span><span class="n">embedded</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">__init_hidden</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">()[</span><span class="mi">0</span><span class="p">]))</span>
|
|
<span class="n">abstracted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">rnn_hidden</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">]))</span>
|
|
<span class="n">abstracted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">doc_embedder</span><span class="p">(</span><span class="n">abstracted</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">abstracted</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="LSTMnet.get_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.LSTMnet.get_params">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Get hyper-parameters for this estimator</span>
|
|
|
|
<span class="sd"> :return: a dictionary with parameter names mapped to their values</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">hyperparams</span></div>
|
|
|
|
|
|
<span class="nd">@property</span>
|
|
<span class="k">def</span> <span class="nf">vocabulary_size</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Return the size of the vocabulary</span>
|
|
|
|
<span class="sd"> :return: integer</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">vocabulary_size_</span></div>
|
|
|
|
|
|
|
|
<div class="viewcode-block" id="CNNnet">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.CNNnet">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">CNNnet</span><span class="p">(</span><span class="n">TextClassifierNet</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> An implementation of :class:`quapy.classification.neural.TextClassifierNet` based on</span>
|
|
<span class="sd"> Convolutional Neural Networks.</span>
|
|
|
|
<span class="sd"> :param vocabulary_size: the size of the vocabulary</span>
|
|
<span class="sd"> :param n_classes: number of target classes</span>
|
|
<span class="sd"> :param embedding_size: the dimensionality of the word embeddings space (default 100)</span>
|
|
<span class="sd"> :param hidden_size: the dimensionality of the hidden space (default 256)</span>
|
|
<span class="sd"> :param repr_size: the dimensionality of the document embeddings space (default 100)</span>
|
|
<span class="sd"> :param kernel_heights: list of kernel lengths (default [3,5,7]), i.e., the number of</span>
|
|
<span class="sd"> consecutive tokens that each kernel covers</span>
|
|
<span class="sd"> :param stride: convolutional stride (default 1)</span>
|
|
<span class="sd"> :param stride: convolutional pad (default 0)</span>
|
|
<span class="sd"> :param drop_p: drop probability for dropout (default 0.5)</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">vocabulary_size</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">,</span> <span class="n">embedding_size</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> <span class="n">hidden_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">repr_size</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
|
|
<span class="n">kernel_heights</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">drop_p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">(</span><span class="n">CNNnet</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">vocabulary_size_</span> <span class="o">=</span> <span class="n">vocabulary_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">n_classes</span> <span class="o">=</span> <span class="n">n_classes</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">hyperparams</span><span class="o">=</span><span class="p">{</span>
|
|
<span class="s1">'embedding_size'</span><span class="p">:</span> <span class="n">embedding_size</span><span class="p">,</span>
|
|
<span class="s1">'hidden_size'</span><span class="p">:</span> <span class="n">hidden_size</span><span class="p">,</span>
|
|
<span class="s1">'repr_size'</span><span class="p">:</span> <span class="n">repr_size</span><span class="p">,</span>
|
|
<span class="s1">'kernel_heights'</span><span class="p">:</span><span class="n">kernel_heights</span><span class="p">,</span>
|
|
<span class="s1">'stride'</span><span class="p">:</span> <span class="n">stride</span><span class="p">,</span>
|
|
<span class="s1">'drop_p'</span><span class="p">:</span> <span class="n">drop_p</span>
|
|
<span class="p">}</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">word_embedding</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Embedding</span><span class="p">(</span><span class="n">vocabulary_size</span><span class="p">,</span> <span class="n">embedding_size</span><span class="p">)</span>
|
|
<span class="n">in_channels</span> <span class="o">=</span> <span class="mi">1</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="p">(</span><span class="n">kernel_heights</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">embedding_size</span><span class="p">),</span> <span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="p">(</span><span class="n">kernel_heights</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">embedding_size</span><span class="p">),</span> <span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">conv3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="p">(</span><span class="n">kernel_heights</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">embedding_size</span><span class="p">),</span> <span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">drop_p</span><span class="p">)</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">dim</span> <span class="o">=</span> <span class="n">repr_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">doc_embedder</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">kernel_heights</span><span class="p">)</span> <span class="o">*</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">dim</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">output</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dim</span><span class="p">,</span> <span class="n">n_classes</span><span class="p">)</span>
|
|
|
|
<span class="k">def</span> <span class="nf">__conv_block</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">,</span> <span class="n">conv_layer</span><span class="p">):</span>
|
|
<span class="n">conv_out</span> <span class="o">=</span> <span class="n">conv_layer</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span> <span class="c1"># conv_out.size() = (batch_size, out_channels, dim, 1)</span>
|
|
<span class="n">activation</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">conv_out</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">3</span><span class="p">))</span> <span class="c1"># activation.size() = (batch_size, out_channels, dim1)</span>
|
|
<span class="n">max_out</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">max_pool1d</span><span class="p">(</span><span class="n">activation</span><span class="p">,</span> <span class="n">activation</span><span class="o">.</span><span class="n">size</span><span class="p">()[</span><span class="mi">2</span><span class="p">])</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span> <span class="c1"># maxpool_out.size() = (batch_size, out_channels)</span>
|
|
<span class="k">return</span> <span class="n">max_out</span>
|
|
|
|
<div class="viewcode-block" id="CNNnet.document_embedding">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.CNNnet.document_embedding">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">document_embedding</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""Embeds documents (i.e., performs the forward pass up to the</span>
|
|
<span class="sd"> next-to-last layer).</span>
|
|
|
|
<span class="sd"> :param input: a batch of instances, typically generated by a torch's `DataLoader`</span>
|
|
<span class="sd"> instance (see :class:`quapy.classification.neural.TorchDataset`)</span>
|
|
<span class="sd"> :return: a torch tensor of shape `(n_samples, n_dimensions)`, where</span>
|
|
<span class="sd"> `n_samples` is the number of documents, and `n_dimensions` is the</span>
|
|
<span class="sd"> dimensionality of the embedding</span>
|
|
<span class="sd"> """</span>
|
|
<span class="nb">input</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">word_embedding</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
|
|
<span class="nb">input</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># input.size() = (batch_size, 1, num_seq, embedding_length)</span>
|
|
|
|
<span class="n">max_out1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__conv_block</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">)</span>
|
|
<span class="n">max_out2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__conv_block</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">)</span>
|
|
<span class="n">max_out3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__conv_block</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">)</span>
|
|
|
|
<span class="n">all_out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">max_out1</span><span class="p">,</span> <span class="n">max_out2</span><span class="p">,</span> <span class="n">max_out3</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span> <span class="c1"># all_out.size() = (batch_size, num_kernels*out_channels)</span>
|
|
<span class="n">abstracted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">all_out</span><span class="p">))</span> <span class="c1"># (batch_size, num_kernels*out_channels)</span>
|
|
<span class="n">abstracted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">doc_embedder</span><span class="p">(</span><span class="n">abstracted</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">abstracted</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="CNNnet.get_params">
|
|
<a class="viewcode-back" href="../../../quapy.classification.html#quapy.classification.neural.CNNnet.get_params">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Get hyper-parameters for this estimator</span>
|
|
|
|
<span class="sd"> :return: a dictionary with parameter names mapped to their values</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">hyperparams</span></div>
|
|
|
|
|
|
<span class="nd">@property</span>
|
|
<span class="k">def</span> <span class="nf">vocabulary_size</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Return the size of the vocabulary</span>
|
|
|
|
<span class="sd"> :return: integer</span>
|
|
<span class="sd"> """</span>
|
|
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">vocabulary_size_</span></div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
</pre></div>
|
|
|
|
</div>
|
|
</div>
|
|
<footer>
|
|
|
|
<hr/>
|
|
|
|
<div role="contentinfo">
|
|
<p>© Copyright 2024, Alejandro Moreo.</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> |