text.html

---

title: text
keywords: fastai
sidebar: home_sidebar

summary: "Application to NLP, including ULMFiT fine-tuning"
---

<div class="container" id="notebook-container">
    
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Text-models,-data,-and-training">Text models, data, and training<a class="anchor-link" href="#Text-models,-data,-and-training">&#182;</a></h2>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>The <a href="/text.html#text"><code>text</code></a> module of the fastai library contains all the necessary functions to define a Dataset suitable for the various NLP (Natural Language Processing) tasks and quickly generate models you can use for them. Specifically:</p>
<ul>
<li><a href="/text.transform.html#text.transform"><code>text.transform</code></a> contains all the scripts to preprocess your data, from raw text to token ids,</li>
<li><a href="/text.data.html#text.data"><code>text.data</code></a> contains the definition of <a href="/text.data.html#TextDataBunch"><code>TextDataBunch</code></a>, which the main class you'll need in NLP,</li>
<li><a href="/text.learner.html#text.learner"><code>text.learner</code></a> contains helper functions to quickly create a language model or an RNN classifier.</li>
</ul>
<p>Have a look at the links above for full details of the API of each module, of read on for a quick overview.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Quick-Start:-Training-an-IMDb-sentiment-model-with-ULMFiT">Quick Start: Training an IMDb sentiment model with <em>ULMFiT</em><a class="anchor-link" href="#Quick-Start:-Training-an-IMDb-sentiment-model-with-ULMFiT">&#182;</a></h2>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Let's start with a quick end-to-end example of training a model. We'll train a sentiment classifier on a sample of the popular IMDb data, showing 4 steps:</p>
<ol>
<li>Reading and viewing the IMDb data</li>
<li>Getting your data ready for modeling</li>
<li>Fine-tuning a language model</li>
<li>Building a classifier</li>
</ol>

</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h3 id="Reading-and-viewing-the-IMDb-data">Reading and viewing the IMDb data<a class="anchor-link" href="#Reading-and-viewing-the-IMDb-data">&#182;</a></h3>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>First let's import everything we need for text.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="kn">from</span> <span class="nn">fastai.text</span> <span class="k">import</span> <span class="o">*</span> 
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Contrary to images in Computer Vision, text can't directly be transformed into numbers to be fed into a model. The first thing we need to do is to preprocess our data so that we change the raw texts to lists of words, or tokens (a step that is called tokenization) then transform these tokens into numbers (a step that is called numericalization). These numbers are then passed to embedding layers that will convert them in arrays of floats before passing them through a model.</p>
<p>You can find on the web plenty of <a href="https://en.wikipedia.org/wiki/Word_embedding">Word Embeddings</a> to directly convert your tokens into floats. Those word embeddings have generally be trained on a large corpus such as wikipedia. Following the work of <a href="https://arxiv.org/abs/1801.06146">ULMFiT</a>, the fastai library is more focused on using pre-trained Language Models and fine-tuning them. Word embeddings are just vectors of 300 or 400 floats that represent different words, but a pretrained language model not only has those, but has also been trained to get a representation of full sentences and documents.</p>
<p>That's why the library is structured around three steps:</p>
<ol>
<li>Get your data preprocessed and ready to use in a minimum amount of code,</li>
<li>Create a language model with pretrained weights that you can fine-tune to your dataset,</li>
<li>Create other models such as classifiers on top of the encoder of the language model.</li>
</ol>
<p>To show examples, we have provided a small sample of the <a href="https://www.imdb.com/interfaces/">IMDB dataset</a> which contains 1,000 reviews of movies with labels (positive or negative).</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">path</span> <span class="o">=</span> <span class="n">untar_data</span><span class="p">(</span><span class="n">URLs</span><span class="o">.</span><span class="n">IMDB_SAMPLE</span><span class="p">)</span>
<span class="n">path</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">



<div class="output_text output_subarea output_execute_result">
<pre>PosixPath(&#39;/home/ubuntu/.fastai/data/imdb_sample&#39;)</pre>
</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Creating a dataset from your raw texts is very simple if you have it in one of those ways</p>
<ul>
<li>organized it in folders in an ImageNet style</li>
<li>organized in a csv file with labels columns and a text columns</li>
</ul>
<p>Here, the sample from imdb is in a texts csv files that looks like this:</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="n">path</span><span class="o">/</span><span class="s1">&#39;texts.csv&#39;</span><span class="p">)</span>
<span class="n">df</span><span class="o">.</span><span class="n">head</span><span class="p">()</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea output_execute_result">
<div>
<style scoped>
    .dataframe tbody tr th:only-of-type {
        vertical-align: middle;
    }

    .dataframe tbody tr th {
        vertical-align: top;
    }

    .dataframe thead th {
        text-align: right;
    }
</style>
<table border="1" class="dataframe">
  <thead>
    <tr style="text-align: right;">
      <th></th>
      <th>label</th>
      <th>text</th>
      <th>is_valid</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <th>0</th>
      <td>negative</td>
      <td>Un-bleeping-believable! Meg Ryan doesn't even ...</td>
      <td>False</td>
    </tr>
    <tr>
      <th>1</th>
      <td>positive</td>
      <td>This is a extremely well-made film. The acting...</td>
      <td>False</td>
    </tr>
    <tr>
      <th>2</th>
      <td>negative</td>
      <td>Every once in a long while a movie will come a...</td>
      <td>False</td>
    </tr>
    <tr>
      <th>3</th>
      <td>positive</td>
      <td>Name just says it all. I watched this movie wi...</td>
      <td>False</td>
    </tr>
    <tr>
      <th>4</th>
      <td>negative</td>
      <td>This movie succeeds at being one of the most u...</td>
      <td>False</td>
    </tr>
  </tbody>
</table>
</div>
</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h3 id="Getting-your-data-ready-for-modeling">Getting your data ready for modeling<a class="anchor-link" href="#Getting-your-data-ready-for-modeling">&#182;</a></h3>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>To get a <a href="/basic_data.html#DataBunch"><code>DataBunch</code></a> quickly, there are also several factory methods depending on how our data is structured. They are all detailed in <a href="/text.data.html#text.data"><code>text.data</code></a>, here we'll use the method <code>from_csv</code> of the <a href="/text.data.html#TextLMDataBunch"><code>TextLMDataBunch</code></a> (to get the data ready for a language model) and <a href="/text.data.html#TextClasDataBunch"><code>TextClasDataBunch</code></a> (to get the data ready for a text classifier) classes.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="c1"># Language model data</span>
<span class="n">data_lm</span> <span class="o">=</span> <span class="n">TextLMDataBunch</span><span class="o">.</span><span class="n">from_csv</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s1">&#39;texts.csv&#39;</span><span class="p">)</span>
<span class="c1"># Classifier model data</span>
<span class="n">data_clas</span> <span class="o">=</span> <span class="n">TextClasDataBunch</span><span class="o">.</span><span class="n">from_csv</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s1">&#39;texts.csv&#39;</span><span class="p">,</span> <span class="n">vocab</span><span class="o">=</span><span class="n">data_lm</span><span class="o">.</span><span class="n">train_ds</span><span class="o">.</span><span class="n">vocab</span><span class="p">,</span> <span class="n">bs</span><span class="o">=</span><span class="mi">32</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>This does all the necessary preprocessing behind the scene. For the classifier, we also pass the vocabulary (mapping from ids to words) that we want to use: this is to ensure that <code>data_clas</code> will use the same dictionary as <code>data_lm</code>.</p>
<p>Since this step can be a bit time-consuming, it's best to save the result with:</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">data_lm</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="s1">&#39;data_lm_export.pkl&#39;</span><span class="p">)</span>
<span class="n">data_clas</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="s1">&#39;data_clas_export.pkl&#39;</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>This will create a 'tmp' directory where all the computed stuff will be stored. You can then reload those results with:</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">data_lm</span> <span class="o">=</span> <span class="n">load_data</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">fname</span><span class="o">=</span><span class="s1">&#39;data_lm_export.pkl&#39;</span><span class="p">)</span>
<span class="n">data_clas</span> <span class="o">=</span> <span class="n">load_data</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">fname</span><span class="o">=</span><span class="s1">&#39;data_clas_export.pkl&#39;</span><span class="p">,</span> <span class="n">bs</span><span class="o">=</span><span class="mi">16</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Note that you can load the data with different <a href="/basic_data.html#DataBunch"><code>DataBunch</code></a> parameters (batch size, <code>bptt</code>,...)</p>

</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h3 id="Fine-tuning-a-language-model">Fine-tuning a language model<a class="anchor-link" href="#Fine-tuning-a-language-model">&#182;</a></h3>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>We can use the <code>data_lm</code> object we created earlier to fine-tune a pretrained language model. <a href="http://www.fast.ai/">fast.ai</a> has an English model with an AWD-LSTM architecture available that we can download. We can create a learner object that will directly create a model, download the pretrained weights and be ready for fine-tuning.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span> <span class="o">=</span> <span class="n">language_model_learner</span><span class="p">(</span><span class="n">data_lm</span><span class="p">,</span> <span class="n">AWD_LSTM</span><span class="p">,</span> <span class="n">drop_mult</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
<span class="n">learn</span><span class="o">.</span><span class="n">fit_one_cycle</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mf">1e-2</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
Total time: 00:13 <p><table style='width:300px; margin-bottom:10px'>
  <tr>
    <th>epoch</th>
    <th>train_loss</th>
    <th>valid_loss</th>
    <th>accuracy</th>
  </tr>
  <tr>
    <th>1</th>
    <th>4.514897</th>
    <th>3.974741</th>
    <th>0.282455</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Like a computer vision model, we can then unfreeze the model and fine-tune it.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">unfreeze</span><span class="p">()</span>
<span class="n">learn</span><span class="o">.</span><span class="n">fit_one_cycle</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mf">1e-3</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
Total time: 00:17 <p><table style='width:300px; margin-bottom:10px'>
  <tr>
    <th>epoch</th>
    <th>train_loss</th>
    <th>valid_loss</th>
    <th>accuracy</th>
  </tr>
  <tr>
    <th>1</th>
    <th>4.159299</th>
    <th>3.886238</th>
    <th>0.289092</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>To evaluate your language model, you can run the <a href="/basic_train.html#Learner.predict"><code>Learner.predict</code></a> method and specify the number of words you want it to guess.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="s2">&quot;This is a review about&quot;</span><span class="p">,</span> <span class="n">n_words</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">



<div class="output_text output_subarea output_execute_result">
<pre>&#39;This is a review about dog - twist credited , and that , along with&#39;</pre>
</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>It doesn't make much sense (we have a tiny vocabulary here and didn't train much on it) but note that it respects basic grammar (which comes from the pretrained model).</p>
<p>Finally we save the encoder to be able to use it for classification in the next section.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">save_encoder</span><span class="p">(</span><span class="s1">&#39;ft_enc&#39;</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h3 id="Building-a-classifier">Building a classifier<a class="anchor-link" href="#Building-a-classifier">&#182;</a></h3>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>We now use the <code>data_clas</code> object we created earlier to build a classifier with our fine-tuned encoder. The learner object can be done in a single line.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span> <span class="o">=</span> <span class="n">text_classifier_learner</span><span class="p">(</span><span class="n">data_clas</span><span class="p">,</span> <span class="n">AWD_LSTM</span><span class="p">,</span> <span class="n">drop_mult</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
<span class="n">learn</span><span class="o">.</span><span class="n">load_encoder</span><span class="p">(</span><span class="s1">&#39;ft_enc&#39;</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">data_clas</span><span class="o">.</span><span class="n">show_batch</span><span class="p">()</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
<table>  <col width='90%'>  <col width='10%'>  <tr>
    <th>text</th>
    <th>target</th>
  </tr>
  <tr>
    <th>xxbos xxmaj raising xxmaj victor xxmaj vargas : a xxmaj review \n \n  xxmaj you know , xxmaj raising xxmaj victor xxmaj vargas is like sticking your hands into a big , xxunk bowl of xxunk . xxmaj it 's warm and gooey , but you 're not sure if it feels right . xxmaj try as i might , no matter how warm and gooey xxmaj raising xxmaj</th>
    <th>negative</th>
  </tr>
  <tr>
    <th>xxbos xxmaj now that xxmaj che(2008 ) has finished its relatively short xxmaj australian cinema run ( extremely limited xxunk screen in xxmaj xxunk , after xxunk ) , i can xxunk join both xxunk of " xxmaj at xxmaj the xxmaj movies " in taking xxmaj steven xxmaj soderbergh to task . \n \n  xxmaj it 's usually satisfying to watch a film director change his style /</th>
    <th>negative</th>
  </tr>
  <tr>
    <th>xxbos i really wanted to love this show . i truly , honestly did . \n \n  xxmaj for the first time , gay viewers get their own version of the " xxmaj the xxmaj xxunk " . xxmaj with the help of his obligatory " hag " xxmaj xxunk , xxmaj james , a good looking , well - to - do thirty - something has the chance</th>
    <th>negative</th>
  </tr>
  <tr>
    <th>xxbos xxmaj to review this movie , i without any doubt would have to quote that memorable scene in xxmaj tarantino 's " xxmaj pulp xxmaj fiction " ( xxunk ) when xxmaj jules and xxmaj vincent are talking about xxmaj mia xxmaj wallace and what she does for a living . xxmaj jules tells xxmaj vincent that the " xxmaj only thing she did worthwhile was pilot " .</th>
    <th>negative</th>
  </tr>
  <tr>
    <th>xxbos xxmaj how viewers react to this new " adaption " of xxmaj shirley xxmaj jackson 's book , which was promoted as xxup not being a remake of the original 1963 movie ( true enough ) , will be based , i suspect , on the following : those who were big fans of either the book or original movie are not going to think much of this one</th>
    <th>negative</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">fit_one_cycle</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mf">1e-2</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
Total time: 00:33 <p><table style='width:300px; margin-bottom:10px'>
  <tr>
    <th>epoch</th>
    <th>train_loss</th>
    <th>valid_loss</th>
    <th>accuracy</th>
  </tr>
  <tr>
    <th>1</th>
    <th>0.650518</th>
    <th>0.599687</th>
    <th>0.691542</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Again, we can unfreeze the model and fine-tune it.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">freeze_to</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">)</span>
<span class="n">learn</span><span class="o">.</span><span class="n">fit_one_cycle</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="nb">slice</span><span class="p">(</span><span class="mf">5e-3</span><span class="o">/</span><span class="mf">2.</span><span class="p">,</span> <span class="mf">5e-3</span><span class="p">))</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
Total time: 00:38 <p><table style='width:300px; margin-bottom:10px'>
  <tr>
    <th>epoch</th>
    <th>train_loss</th>
    <th>valid_loss</th>
    <th>accuracy</th>
  </tr>
  <tr>
    <th>1</th>
    <th>0.628320</th>
    <th>0.563579</th>
    <th>0.716418</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">unfreeze</span><span class="p">()</span>
<span class="n">learn</span><span class="o">.</span><span class="n">fit_one_cycle</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="nb">slice</span><span class="p">(</span><span class="mf">2e-3</span><span class="o">/</span><span class="mi">100</span><span class="p">,</span> <span class="mf">2e-3</span><span class="p">))</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">


<div class="output_html rendered_html output_subarea ">
Total time: 00:56 <p><table style='width:300px; margin-bottom:10px'>
  <tr>
    <th>epoch</th>
    <th>train_loss</th>
    <th>valid_loss</th>
    <th>accuracy</th>
  </tr>
  <tr>
    <th>1</th>
    <th>0.533411</th>
    <th>0.539693</th>
    <th>0.721393</th>
  </tr>
</table>

</div>

</div>

</div>
</div>

</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Again, we can predict on a raw text by using the <a href="/basic_train.html#Learner.predict"><code>Learner.predict</code></a> method.</p>

</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">

<div class="inner_cell">
    <div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">learn</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="s2">&quot;This was a great movie!&quot;</span><span class="p">)</span>
</pre></div>

    </div>
</div>
</div>

<div class="output_wrapper">
<div class="output">

<div class="output_area">



<div class="output_text output_subarea output_execute_result">
<pre>(Category positive, tensor(1), tensor([0.0118, 0.9882]))</pre>
</div>

</div>

</div>
</div>

</div>
</div>