page_type: reference description: Class that performs dense feature vector extraction on text.

<link rel="stylesheet" href="/site-assets/css/style.css"> <!-- DO NOT EDIT! Automatically generated file. --> <div itemscope itemtype="http://developers.google.com/ReferenceObject"> <meta itemprop="name" content="tflite_support.task.text.TextEmbedder" /> <meta itemprop="path" content="Stable" /> <meta itemprop="property" content="__init__"/> <meta itemprop="property" content="cosine_similarity"/> <meta itemprop="property" content="create_from_file"/> <meta itemprop="property" content="create_from_options"/> <meta itemprop="property" content="embed"/> <meta itemprop="property" content="get_embedding_dimension"/> </div>

tflite_support.task.text.TextEmbedder

<!-- Insert buttons and diff --> <table class="tfo-notebook-buttons tfo-api nocontent" align="left"> <td> <a target="_blank" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L42-L128"> </a> </td> </table>

Class that performs dense feature vector extraction on text.

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>tflite_support.task.text.TextEmbedder( options: <a href="../../../tflite_support/task/text/TextEmbedderOptions"><code>tflite_support.task.text.TextEmbedderOptions</code></a>, cpp_embedder: _CppTextEmbedder ) -> None </code></pre> <!-- Placeholder for "Used in" --> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2"><h2 class="add-link">Attributes</h2></th></tr> <tr> <td> `number_of_output_layers`<a id="number_of_output_layers"></a> </td> <td> Gets the number of output layers of the model. </td> </tr><tr> <td> `options`<a id="options"></a> </td> <td> </td> </tr> </table>

Methods

<h3 id="cosine_similarity"><code>cosine_similarity</code></h3>

<a target="_blank" class="external" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L104-L107">View source</a>

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>cosine_similarity( u: <a href="../../../tflite_support/task/processor/FeatureVector"><code>tflite_support.task.processor.FeatureVector</code></a>, v: <a href="../../../tflite_support/task/processor/FeatureVector"><code>tflite_support.task.processor.FeatureVector</code></a> ) -> float </code></pre>

Computes cosine similarity [1] between two feature vectors.

<h3 id="create_from_file"><code>create_from_file</code></h3>

<a target="_blank" class="external" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L52-L68">View source</a>

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>@classmethod</code> <code>create_from_file( file_path: str ) -> 'TextEmbedder' </code></pre>

Creates the TextEmbedder object from a TensorFlow Lite model.

<!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Args</th></tr> <tr> <td> `file_path` </td> <td> Path to the model. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Returns</th></tr> <tr class="alt"> <td colspan="2"> `TextEmbedder` object that's created from the model file. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Raises</th></tr> <tr> <td> `ValueError` </td> <td> If failed to create `TextEmbedder` object from the provided file such as invalid file. </td> </tr><tr> <td> `RuntimeError` </td> <td> If other types of error occurred. </td> </tr> </table> <h3 id="create_from_options"><code>create_from_options</code></h3>

<a target="_blank" class="external" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L70-L86">View source</a>

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>@classmethod</code> <code>create_from_options( options: <a href="../../../tflite_support/task/text/TextEmbedderOptions"><code>tflite_support.task.text.TextEmbedderOptions</code></a> ) -> 'TextEmbedder' </code></pre>

Creates the TextEmbedder object from text embedder options.

<!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Args</th></tr> <tr> <td> `options` </td> <td> Options for the text embedder task. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Returns</th></tr> <tr class="alt"> <td colspan="2"> `TextEmbedder` object that's created from `options`. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Raises</th></tr> <tr> <td> `ValueError` </td> <td> If failed to create `TextEmbedder` object from `TextEmbedderOptions` such as missing the model. </td> </tr><tr> <td> `RuntimeError` </td> <td> If other types of error occurred. </td> </tr> </table> <h3 id="embed"><code>embed</code></h3>

<a target="_blank" class="external" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L88-L102">View source</a>

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>embed( text: str ) -> <a href="../../../tflite_support/task/processor/EmbeddingResult"><code>tflite_support.task.processor.EmbeddingResult</code></a> </code></pre>

Performs actual feature vector extraction on the provided text.

<!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Args</th></tr> <tr> <td> <a href="https://www.tensorflow.org/text/api_docs/python/text"><code>text</code></a> </td> <td> the input text, used to extract the feature vectors. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Returns</th></tr> <tr class="alt"> <td colspan="2"> embedding result. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Raises</th></tr> <tr> <td> `ValueError` </td> <td> If any of the input arguments is invalid. </td> </tr><tr> <td> `RuntimeError` </td> <td> If failed to calculate the embedding vector. </td> </tr> </table> <h3 id="get_embedding_dimension"><code>get_embedding_dimension</code></h3>

<a target="_blank" class="external" href="https://github.com/tensorflow/tflite-support/blob/v0.4.4/tensorflow_lite_support/python/task/text/text_embedder.py#L109-L119">View source</a>

<pre class="devsite-click-to-copy prettyprint lang-py tfo-signature-link"> <code>get_embedding_dimension( output_index: int ) -> int </code></pre>

Gets the dimensionality of the embedding output.

<!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Args</th></tr> <tr> <td> `output_index` </td> <td> The output index of output layer. </td> </tr> </table> <!-- Tabular view --> <table class="responsive fixed orange"> <colgroup><col width="214px"><col></colgroup> <tr><th colspan="2">Returns</th></tr> <tr class="alt"> <td colspan="2"> Dimensionality of the embedding output by the output_index'th output layer. Returns -1 if `output_index` is out of bounds. </td> </tr> </table>