docs/text/learner.html

<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1" />
<meta name="generator" content="pdoc 0.9.2" />
<title>ktrain.text.learner API documentation</title>
<meta name="description" content="" />
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/sanitize.min.css" integrity="sha256-PK9q560IAAa6WVRRh76LtCaI8pjTJ2z11v0miyNNjrs=" crossorigin>
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/typography.min.css" integrity="sha256-7l/o7C8jubJiy74VsKTidCy1yBkRtiUGbVkYBylBqUg=" crossorigin>
<link rel="stylesheet preload" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/styles/github.min.css" crossorigin>
<style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:30px;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:1em 0 .50em 0}h3{font-size:1.4em;margin:25px 0 10px 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .3s ease-in-out}a:hover{color:#e82}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900}pre code{background:#f8f8f8;font-size:.8em;line-height:1.4em}code{background:#f2f2f1;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{background:#f8f8f8;border:0;border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0;padding:1ex}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-weight:bold;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em .5em;margin-bottom:1em}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
<style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.item .name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul{padding-left:1.5em}.toc > ul > li{margin-top:.5em}}</style>
<style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
<script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/highlight.min.js" integrity="sha256-Uv3H6lx7dJmRfRvH8TH6kJD1TSK1aFcwgx+mdg3epi8=" crossorigin></script>
<script>window.addEventListener('DOMContentLoaded', () => hljs.initHighlighting())</script>
</head>
<body>
<main>
<article id="content">
<header>
<h1 class="title">Module <code>ktrain.text.learner</code></h1>
</header>
<section id="section-intro">
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">from ..imports import *
from .. import utils as U
from ..core import ArrayLearner, GenLearner, _load_model
from .preprocessor import TransformersPreprocessor


class BERTTextClassLearner(ArrayLearner):
    &#34;&#34;&#34;
    ```
    Main class used to tune and train Keras models for text classification using Array data.
    ```
    &#34;&#34;&#34;


    def __init__(self, model, train_data=None, val_data=None, 
                 batch_size=U.DEFAULT_BS, eval_batch_size=U.DEFAULT_BS,
                 workers=1, use_multiprocessing=False):
        super().__init__(model, train_data=train_data, val_data=val_data,
                         batch_size=batch_size, eval_batch_size=eval_batch_size,
                         workers=workers, use_multiprocessing=use_multiprocessing)
        return


    def view_top_losses(self, n=4, preproc=None, val_data=None):
        &#34;&#34;&#34;
        ```
        Views observations with top losses in validation set.
        Args:
         n(int or tuple): a range to select in form of int or tuple
                          e.g., n=8 is treated as n=(0,8)
         preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                                 For some data like text data, a preprocessor
                                 is required to undo the pre-processing
                                 to correctly view raw data.
          val_data:  optional val_data to use instead of self.val_data
        Returns:
            list of n tuples where first element is either 
            filepath or id of validation example and second element
            is loss.
        ```
        &#34;&#34;&#34;
        val = self._check_val(val_data)


        # get top losses and associated data
        tups = self.top_losses(n=n, val_data=val, preproc=preproc)

        # get multilabel status and class names
        classes = preproc.get_classes() if preproc is not None else None

        # iterate through losses
        for tup in tups:

            # get data
            idx = tup[0]
            loss = tup[1]
            truth = tup[2]
            pred = tup[3]

            # BERT-style tuple
            join_char = &#39; &#39;
            obs = val[0][0][idx]
            if preproc is not None: 
                obs = preproc.undo(obs)
                if preproc.is_nospace_lang(): join_char = &#39;&#39;
            if type(obs) == str:
                obs = join_char.join(obs.split()[:512])
            print(&#39;----------&#39;)
            print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
            print(obs)
        return


class TransformerTextClassLearner(GenLearner):
    &#34;&#34;&#34;
    ```
    Main class used to tune and train Keras models for text classification using Array data.
    ```
    &#34;&#34;&#34;


    def __init__(self, model, train_data=None, val_data=None, 
                 batch_size=U.DEFAULT_BS, eval_batch_size=U.DEFAULT_BS,
                 workers=1, use_multiprocessing=False):
        super().__init__(model, train_data=train_data, val_data=val_data,
                         batch_size=batch_size, eval_batch_size=eval_batch_size,
                         workers=workers, use_multiprocessing=use_multiprocessing)
        return


    def view_top_losses(self, n=4, preproc=None, val_data=None):
        &#34;&#34;&#34;
        ```
        Views observations with top losses in validation set.
        Args:
         n(int or tuple): a range to select in form of int or tuple
                          e.g., n=8 is treated as n=(0,8)
         preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                                 For some data like text data, a preprocessor
                                 is required to undo the pre-processing
                                 to correctly view raw data.
          val_data:  optional val_data to use instead of self.val_data
        Returns:
            list of n tuples where first element is either 
            filepath or id of validation example and second element
            is loss.
        ```
        &#34;&#34;&#34;
        val = self._check_val(val_data)


        # get top losses and associated data
        tups = self.top_losses(n=n, val_data=val, preproc=preproc)

        # get multilabel status and class names
        classes = preproc.get_classes() if preproc is not None else None

        # iterate through losses
        for tup in tups:

            # get data
            idx = tup[0]
            loss = tup[1]
            truth = tup[2]
            pred = tup[3]

            join_char = &#39; &#39;
            #obs = val.x[idx][0]
            print(&#39;----------&#39;)
            print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
        return


    def _prepare(self, data, train=True):
        &#34;&#34;&#34;
        ```
        prepare data as tf.Dataset
        ```
        &#34;&#34;&#34;
        # HF_EXCEPTION
        # convert arrays to TF dataset (iterator) on-the-fly
        # to work around issues with transformers and tf.Datasets
        if data is None: return None
        return data.to_tfdataset(train=train)


    def predict(self, val_data=None):
        &#34;&#34;&#34;
        ```
        Makes predictions on validation set
        ```
        &#34;&#34;&#34;
        if val_data is not None:
            val = val_data
        else:
            val = self.val_data
        if val is None: raise Exception(&#39;val_data must be supplied to get_learner or predict&#39;)
        if hasattr(val, &#39;reset&#39;): val.reset()
        classification, multilabel = U.is_classifier(self.model)
        preds = self.model.predict(self._prepare(val, train=False))
        if hasattr(preds, &#39;logits&#39;): # dep_fix: breaking change in transformers==4.0.0 - also needed for Longformer
        #if type(preds).__name__ == &#39;TFSequenceClassifierOutput&#39;: # dep_fix: undocumented breaking change in transformers==4.0.0
            # REFERENCE: https://discuss.huggingface.co/t/new-model-output-types/195
            preds = preds.logits

        # dep_fix: transformers in TF 2.2.0 returns a tuple insead of NumPy array for some reason
        if isinstance(preds, tuple) and len(preds) == 1: preds = preds[0] 

        if classification:
            if multilabel:
                return activations.sigmoid(tf.convert_to_tensor(preds)).numpy()
            else:
                return activations.softmax(tf.convert_to_tensor(preds)).numpy()
        else:
            return preds


    def save_model(self, fpath):
        &#34;&#34;&#34;
        ```
        save Transformers model
        ```
        &#34;&#34;&#34;
        self._make_model_folder(fpath)
        self.model.save_pretrained(fpath)
        return


    # 2020-07-07: removed, as core.Learner.load_model calls TransformerPreprocessor.load_model_and_configure
    #def load_model(self, fpath, preproc=None):
    #    &#34;&#34;&#34;
    #    load Transformers model
    #    Args:
    #      fpath(str): path to folder containing model files
    #      preproc(TransformerPreprocessor): a TransformerPreprocessor instance.
    #    &#34;&#34;&#34;
    #    if preproc is None or not isinstance(preproc, TransformersPreprocessor):
    #        raise ValueError(&#39;preproc arg is required to load Transformer models from disk. &#39; +\
    #                          &#39;Supply a TransformersPreprocessor instance. This is &#39; +\
    #                          &#39;either the third return value from texts_from* function or &#39;+\
    #                          &#39;the result of calling ktrain.text.Transformer&#39;)


    #    self.model = _load_model(fpath, preproc=preproc)
    #    return</code></pre>
</details>
</section>
<section>
</section>
<section>
</section>
<section>
</section>
<section>
<h2 class="section-title" id="header-classes">Classes</h2>
<dl>
<dt id="ktrain.text.learner.BERTTextClassLearner"><code class="flex name class">
<span>class <span class="ident">BERTTextClassLearner</span></span>
<span>(</span><span>model, train_data=None, val_data=None, batch_size=32, eval_batch_size=32, workers=1, use_multiprocessing=False)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>Main class used to tune and train Keras models for text classification using Array data.
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">class BERTTextClassLearner(ArrayLearner):
    &#34;&#34;&#34;
    ```
    Main class used to tune and train Keras models for text classification using Array data.
    ```
    &#34;&#34;&#34;


    def __init__(self, model, train_data=None, val_data=None, 
                 batch_size=U.DEFAULT_BS, eval_batch_size=U.DEFAULT_BS,
                 workers=1, use_multiprocessing=False):
        super().__init__(model, train_data=train_data, val_data=val_data,
                         batch_size=batch_size, eval_batch_size=eval_batch_size,
                         workers=workers, use_multiprocessing=use_multiprocessing)
        return


    def view_top_losses(self, n=4, preproc=None, val_data=None):
        &#34;&#34;&#34;
        ```
        Views observations with top losses in validation set.
        Args:
         n(int or tuple): a range to select in form of int or tuple
                          e.g., n=8 is treated as n=(0,8)
         preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                                 For some data like text data, a preprocessor
                                 is required to undo the pre-processing
                                 to correctly view raw data.
          val_data:  optional val_data to use instead of self.val_data
        Returns:
            list of n tuples where first element is either 
            filepath or id of validation example and second element
            is loss.
        ```
        &#34;&#34;&#34;
        val = self._check_val(val_data)


        # get top losses and associated data
        tups = self.top_losses(n=n, val_data=val, preproc=preproc)

        # get multilabel status and class names
        classes = preproc.get_classes() if preproc is not None else None

        # iterate through losses
        for tup in tups:

            # get data
            idx = tup[0]
            loss = tup[1]
            truth = tup[2]
            pred = tup[3]

            # BERT-style tuple
            join_char = &#39; &#39;
            obs = val[0][0][idx]
            if preproc is not None: 
                obs = preproc.undo(obs)
                if preproc.is_nospace_lang(): join_char = &#39;&#39;
            if type(obs) == str:
                obs = join_char.join(obs.split()[:512])
            print(&#39;----------&#39;)
            print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
            print(obs)
        return</code></pre>
</details>
<h3>Ancestors</h3>
<ul class="hlist">
<li><a title="ktrain.core.ArrayLearner" href="../core.html#ktrain.core.ArrayLearner">ArrayLearner</a></li>
<li><a title="ktrain.core.Learner" href="../core.html#ktrain.core.Learner">Learner</a></li>
<li>abc.ABC</li>
</ul>
<h3>Methods</h3>
<dl>
<dt id="ktrain.text.learner.BERTTextClassLearner.view_top_losses"><code class="name flex">
<span>def <span class="ident">view_top_losses</span></span>(<span>self, n=4, preproc=None, val_data=None)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>Views observations with top losses in validation set.
Args:
 n(int or tuple): a range to select in form of int or tuple
                  e.g., n=8 is treated as n=(0,8)
 preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                         For some data like text data, a preprocessor
                         is required to undo the pre-processing
                         to correctly view raw data.
  val_data:  optional val_data to use instead of self.val_data
Returns:
    list of n tuples where first element is either 
    filepath or id of validation example and second element
    is loss.
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def view_top_losses(self, n=4, preproc=None, val_data=None):
    &#34;&#34;&#34;
    ```
    Views observations with top losses in validation set.
    Args:
     n(int or tuple): a range to select in form of int or tuple
                      e.g., n=8 is treated as n=(0,8)
     preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                             For some data like text data, a preprocessor
                             is required to undo the pre-processing
                             to correctly view raw data.
      val_data:  optional val_data to use instead of self.val_data
    Returns:
        list of n tuples where first element is either 
        filepath or id of validation example and second element
        is loss.
    ```
    &#34;&#34;&#34;
    val = self._check_val(val_data)


    # get top losses and associated data
    tups = self.top_losses(n=n, val_data=val, preproc=preproc)

    # get multilabel status and class names
    classes = preproc.get_classes() if preproc is not None else None

    # iterate through losses
    for tup in tups:

        # get data
        idx = tup[0]
        loss = tup[1]
        truth = tup[2]
        pred = tup[3]

        # BERT-style tuple
        join_char = &#39; &#39;
        obs = val[0][0][idx]
        if preproc is not None: 
            obs = preproc.undo(obs)
            if preproc.is_nospace_lang(): join_char = &#39;&#39;
        if type(obs) == str:
            obs = join_char.join(obs.split()[:512])
        print(&#39;----------&#39;)
        print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
        print(obs)
    return</code></pre>
</details>
</dd>
</dl>
<h3>Inherited members</h3>
<ul class="hlist">
<li><code><b><a title="ktrain.core.ArrayLearner" href="../core.html#ktrain.core.ArrayLearner">ArrayLearner</a></b></code>:
<ul class="hlist">
<li><code><a title="ktrain.core.ArrayLearner.autofit" href="../core.html#ktrain.core.Learner.autofit">autofit</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.evaluate" href="../core.html#ktrain.core.Learner.evaluate">evaluate</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.fit" href="../core.html#ktrain.core.ArrayLearner.fit">fit</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.fit_onecycle" href="../core.html#ktrain.core.Learner.fit_onecycle">fit_onecycle</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.freeze" href="../core.html#ktrain.core.Learner.freeze">freeze</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.get_weight_decay" href="../core.html#ktrain.core.Learner.get_weight_decay">get_weight_decay</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.layer_output" href="../core.html#ktrain.core.ArrayLearner.layer_output">layer_output</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.load_model" href="../core.html#ktrain.core.Learner.load_model">load_model</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.lr_estimate" href="../core.html#ktrain.core.Learner.lr_estimate">lr_estimate</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.lr_find" href="../core.html#ktrain.core.Learner.lr_find">lr_find</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.lr_plot" href="../core.html#ktrain.core.Learner.lr_plot">lr_plot</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.plot" href="../core.html#ktrain.core.Learner.plot">plot</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.predict" href="../core.html#ktrain.core.Learner.predict">predict</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.print_layers" href="../core.html#ktrain.core.Learner.print_layers">print_layers</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.reset_weights" href="../core.html#ktrain.core.Learner.reset_weights">reset_weights</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.save_model" href="../core.html#ktrain.core.Learner.save_model">save_model</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.set_model" href="../core.html#ktrain.core.Learner.set_model">set_model</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.set_weight_decay" href="../core.html#ktrain.core.Learner.set_weight_decay">set_weight_decay</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.top_losses" href="../core.html#ktrain.core.Learner.top_losses">top_losses</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.unfreeze" href="../core.html#ktrain.core.Learner.unfreeze">unfreeze</a></code></li>
<li><code><a title="ktrain.core.ArrayLearner.validate" href="../core.html#ktrain.core.Learner.validate">validate</a></code></li>
</ul>
</li>
</ul>
</dd>
<dt id="ktrain.text.learner.TransformerTextClassLearner"><code class="flex name class">
<span>class <span class="ident">TransformerTextClassLearner</span></span>
<span>(</span><span>model, train_data=None, val_data=None, batch_size=32, eval_batch_size=32, workers=1, use_multiprocessing=False)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>Main class used to tune and train Keras models for text classification using Array data.
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">class TransformerTextClassLearner(GenLearner):
    &#34;&#34;&#34;
    ```
    Main class used to tune and train Keras models for text classification using Array data.
    ```
    &#34;&#34;&#34;


    def __init__(self, model, train_data=None, val_data=None, 
                 batch_size=U.DEFAULT_BS, eval_batch_size=U.DEFAULT_BS,
                 workers=1, use_multiprocessing=False):
        super().__init__(model, train_data=train_data, val_data=val_data,
                         batch_size=batch_size, eval_batch_size=eval_batch_size,
                         workers=workers, use_multiprocessing=use_multiprocessing)
        return


    def view_top_losses(self, n=4, preproc=None, val_data=None):
        &#34;&#34;&#34;
        ```
        Views observations with top losses in validation set.
        Args:
         n(int or tuple): a range to select in form of int or tuple
                          e.g., n=8 is treated as n=(0,8)
         preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                                 For some data like text data, a preprocessor
                                 is required to undo the pre-processing
                                 to correctly view raw data.
          val_data:  optional val_data to use instead of self.val_data
        Returns:
            list of n tuples where first element is either 
            filepath or id of validation example and second element
            is loss.
        ```
        &#34;&#34;&#34;
        val = self._check_val(val_data)


        # get top losses and associated data
        tups = self.top_losses(n=n, val_data=val, preproc=preproc)

        # get multilabel status and class names
        classes = preproc.get_classes() if preproc is not None else None

        # iterate through losses
        for tup in tups:

            # get data
            idx = tup[0]
            loss = tup[1]
            truth = tup[2]
            pred = tup[3]

            join_char = &#39; &#39;
            #obs = val.x[idx][0]
            print(&#39;----------&#39;)
            print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
        return


    def _prepare(self, data, train=True):
        &#34;&#34;&#34;
        ```
        prepare data as tf.Dataset
        ```
        &#34;&#34;&#34;
        # HF_EXCEPTION
        # convert arrays to TF dataset (iterator) on-the-fly
        # to work around issues with transformers and tf.Datasets
        if data is None: return None
        return data.to_tfdataset(train=train)


    def predict(self, val_data=None):
        &#34;&#34;&#34;
        ```
        Makes predictions on validation set
        ```
        &#34;&#34;&#34;
        if val_data is not None:
            val = val_data
        else:
            val = self.val_data
        if val is None: raise Exception(&#39;val_data must be supplied to get_learner or predict&#39;)
        if hasattr(val, &#39;reset&#39;): val.reset()
        classification, multilabel = U.is_classifier(self.model)
        preds = self.model.predict(self._prepare(val, train=False))
        if hasattr(preds, &#39;logits&#39;): # dep_fix: breaking change in transformers==4.0.0 - also needed for Longformer
        #if type(preds).__name__ == &#39;TFSequenceClassifierOutput&#39;: # dep_fix: undocumented breaking change in transformers==4.0.0
            # REFERENCE: https://discuss.huggingface.co/t/new-model-output-types/195
            preds = preds.logits

        # dep_fix: transformers in TF 2.2.0 returns a tuple insead of NumPy array for some reason
        if isinstance(preds, tuple) and len(preds) == 1: preds = preds[0] 

        if classification:
            if multilabel:
                return activations.sigmoid(tf.convert_to_tensor(preds)).numpy()
            else:
                return activations.softmax(tf.convert_to_tensor(preds)).numpy()
        else:
            return preds


    def save_model(self, fpath):
        &#34;&#34;&#34;
        ```
        save Transformers model
        ```
        &#34;&#34;&#34;
        self._make_model_folder(fpath)
        self.model.save_pretrained(fpath)
        return</code></pre>
</details>
<h3>Ancestors</h3>
<ul class="hlist">
<li><a title="ktrain.core.GenLearner" href="../core.html#ktrain.core.GenLearner">GenLearner</a></li>
<li><a title="ktrain.core.Learner" href="../core.html#ktrain.core.Learner">Learner</a></li>
<li>abc.ABC</li>
</ul>
<h3>Methods</h3>
<dl>
<dt id="ktrain.text.learner.TransformerTextClassLearner.save_model"><code class="name flex">
<span>def <span class="ident">save_model</span></span>(<span>self, fpath)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>save Transformers model
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def save_model(self, fpath):
    &#34;&#34;&#34;
    ```
    save Transformers model
    ```
    &#34;&#34;&#34;
    self._make_model_folder(fpath)
    self.model.save_pretrained(fpath)
    return</code></pre>
</details>
</dd>
<dt id="ktrain.text.learner.TransformerTextClassLearner.view_top_losses"><code class="name flex">
<span>def <span class="ident">view_top_losses</span></span>(<span>self, n=4, preproc=None, val_data=None)</span>
</code></dt>
<dd>
<div class="desc"><pre><code>Views observations with top losses in validation set.
Args:
 n(int or tuple): a range to select in form of int or tuple
                  e.g., n=8 is treated as n=(0,8)
 preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                         For some data like text data, a preprocessor
                         is required to undo the pre-processing
                         to correctly view raw data.
  val_data:  optional val_data to use instead of self.val_data
Returns:
    list of n tuples where first element is either 
    filepath or id of validation example and second element
    is loss.
</code></pre></div>
<details class="source">
<summary>
<span>Expand source code</span>
</summary>
<pre><code class="python">def view_top_losses(self, n=4, preproc=None, val_data=None):
    &#34;&#34;&#34;
    ```
    Views observations with top losses in validation set.
    Args:
     n(int or tuple): a range to select in form of int or tuple
                      e.g., n=8 is treated as n=(0,8)
     preproc (Preprocessor): A TextPreprocessor or ImagePreprocessor.
                             For some data like text data, a preprocessor
                             is required to undo the pre-processing
                             to correctly view raw data.
      val_data:  optional val_data to use instead of self.val_data
    Returns:
        list of n tuples where first element is either 
        filepath or id of validation example and second element
        is loss.
    ```
    &#34;&#34;&#34;
    val = self._check_val(val_data)


    # get top losses and associated data
    tups = self.top_losses(n=n, val_data=val, preproc=preproc)

    # get multilabel status and class names
    classes = preproc.get_classes() if preproc is not None else None

    # iterate through losses
    for tup in tups:

        # get data
        idx = tup[0]
        loss = tup[1]
        truth = tup[2]
        pred = tup[3]

        join_char = &#39; &#39;
        #obs = val.x[idx][0]
        print(&#39;----------&#39;)
        print(&#34;id:%s | loss:%s | true:%s | pred:%s)\n&#34; % (idx, round(loss,2), truth, pred))
    return</code></pre>
</details>
</dd>
</dl>
<h3>Inherited members</h3>
<ul class="hlist">
<li><code><b><a title="ktrain.core.GenLearner" href="../core.html#ktrain.core.GenLearner">GenLearner</a></b></code>:
<ul class="hlist">
<li><code><a title="ktrain.core.GenLearner.autofit" href="../core.html#ktrain.core.Learner.autofit">autofit</a></code></li>
<li><code><a title="ktrain.core.GenLearner.evaluate" href="../core.html#ktrain.core.Learner.evaluate">evaluate</a></code></li>
<li><code><a title="ktrain.core.GenLearner.fit" href="../core.html#ktrain.core.GenLearner.fit">fit</a></code></li>
<li><code><a title="ktrain.core.GenLearner.fit_onecycle" href="../core.html#ktrain.core.Learner.fit_onecycle">fit_onecycle</a></code></li>
<li><code><a title="ktrain.core.GenLearner.freeze" href="../core.html#ktrain.core.Learner.freeze">freeze</a></code></li>
<li><code><a title="ktrain.core.GenLearner.get_weight_decay" href="../core.html#ktrain.core.Learner.get_weight_decay">get_weight_decay</a></code></li>
<li><code><a title="ktrain.core.GenLearner.layer_output" href="../core.html#ktrain.core.GenLearner.layer_output">layer_output</a></code></li>
<li><code><a title="ktrain.core.GenLearner.load_model" href="../core.html#ktrain.core.Learner.load_model">load_model</a></code></li>
<li><code><a title="ktrain.core.GenLearner.lr_estimate" href="../core.html#ktrain.core.Learner.lr_estimate">lr_estimate</a></code></li>
<li><code><a title="ktrain.core.GenLearner.lr_find" href="../core.html#ktrain.core.Learner.lr_find">lr_find</a></code></li>
<li><code><a title="ktrain.core.GenLearner.lr_plot" href="../core.html#ktrain.core.Learner.lr_plot">lr_plot</a></code></li>
<li><code><a title="ktrain.core.GenLearner.plot" href="../core.html#ktrain.core.Learner.plot">plot</a></code></li>
<li><code><a title="ktrain.core.GenLearner.predict" href="../core.html#ktrain.core.Learner.predict">predict</a></code></li>
<li><code><a title="ktrain.core.GenLearner.print_layers" href="../core.html#ktrain.core.Learner.print_layers">print_layers</a></code></li>
<li><code><a title="ktrain.core.GenLearner.reset_weights" href="../core.html#ktrain.core.Learner.reset_weights">reset_weights</a></code></li>
<li><code><a title="ktrain.core.GenLearner.set_model" href="../core.html#ktrain.core.Learner.set_model">set_model</a></code></li>
<li><code><a title="ktrain.core.GenLearner.set_weight_decay" href="../core.html#ktrain.core.Learner.set_weight_decay">set_weight_decay</a></code></li>
<li><code><a title="ktrain.core.GenLearner.top_losses" href="../core.html#ktrain.core.Learner.top_losses">top_losses</a></code></li>
<li><code><a title="ktrain.core.GenLearner.unfreeze" href="../core.html#ktrain.core.Learner.unfreeze">unfreeze</a></code></li>
<li><code><a title="ktrain.core.GenLearner.validate" href="../core.html#ktrain.core.Learner.validate">validate</a></code></li>
</ul>
</li>
</ul>
</dd>
</dl>
</section>
</article>
<nav id="sidebar">
<h1>Index</h1>
<div class="toc">
<ul></ul>
</div>
<ul id="index">
<li><h3>Super-module</h3>
<ul>
<li><code><a title="ktrain.text" href="index.html">ktrain.text</a></code></li>
</ul>
</li>
<li><h3><a href="#header-classes">Classes</a></h3>
<ul>
<li>
<h4><code><a title="ktrain.text.learner.BERTTextClassLearner" href="#ktrain.text.learner.BERTTextClassLearner">BERTTextClassLearner</a></code></h4>
<ul class="">
<li><code><a title="ktrain.text.learner.BERTTextClassLearner.view_top_losses" href="#ktrain.text.learner.BERTTextClassLearner.view_top_losses">view_top_losses</a></code></li>
</ul>
</li>
<li>
<h4><code><a title="ktrain.text.learner.TransformerTextClassLearner" href="#ktrain.text.learner.TransformerTextClassLearner">TransformerTextClassLearner</a></code></h4>
<ul class="">
<li><code><a title="ktrain.text.learner.TransformerTextClassLearner.save_model" href="#ktrain.text.learner.TransformerTextClassLearner.save_model">save_model</a></code></li>
<li><code><a title="ktrain.text.learner.TransformerTextClassLearner.view_top_losses" href="#ktrain.text.learner.TransformerTextClassLearner.view_top_losses">view_top_losses</a></code></li>
</ul>
</li>
</ul>
</li>
</ul>
</nav>
</main>
<footer id="footer">
<p>Generated by <a href="https://pdoc3.github.io/pdoc"><cite>pdoc</cite> 0.9.2</a>.</p>
</footer>
</body>
</html>