`EN <https://docs.texta.ee/embedding.html>`_
`ET <https://docs.texta.ee/et/embedding.html>`_


#########
Embedding
#########

.. _create_embedding:

Create
*******

GUI
=====

Navigate to **Models -> Embeddings** and click on the **CREATE** button on top-left. Choose the name for you :ref:`embedding <embedding_concept>` (*Description*).
Define the :ref:`query <query_concept>` and select indices on which the :ref:`query <query_concept>` will be executed. If you leave *Query* empty, it will take all documents from the selected indices.
If you have any searches defined in your :ref:`project <project_concept>`, they will appear in a dropdown menu if you click on the field *Query* - you can use existing searches as queries.

Choose *fields* on which the :ref:`embedding <embedding_concept>` will be trained. The selected fields should contain textual data.

.. note::
	It is recommended to use lemmatized or tokenized data. Lemmatization is especially useful with morphologically complex languages. You can tokenize and lemmatize the data with :ref:`MLP <mlp>`.

Field *Number of dimensions* defines the length of the word vectors. 
100-200 dimensions is usually a good place to start with. 
Field *Minimum frequency* sets how many times a word must occur in the data in minimum in order to 
get included into the :ref:`embedding <embedding_concept>`. Again, you can leave it with the default value *5* if you are unsure which value to pick. 

.. note::
    The quality of the embedding depends on the size of the dataset. The larger the better.

API
===

Endpoint: **/projects/{project_pk}/embeddings/**

Example:

.. code-block:: bash

        curl -X POST "http://localhost:8000/api/v1/projects/11/embeddings/" \
        -H "accept: application/json" \
        -H "Content-Type: application/json" \
        -H "Authorization: Token 8229898dccf960714a9fa22662b214005aa2b049" \
        -d '{
                "description": "My embedding",
                "indices": [{"name": "texta_test_index"}],
                "fields": ["comment_content_lemmas"],
                "num_dimensions": 100,
                "max_documents": 10000,
                "min_freq": 5
            }'


View
*******

GUI
=====

Navigate to **Models -> Embeddings** to view existing :ref:`embedding <embedding_concept>`. 
If any of your embeddings is still training, the view will show you the progress of the training (:numref:`embedding_view`).
Besides than that, the view shows you general information about your embeddings.


.. _embedding_view:
.. figure:: images/embedding/embedding_view.png

	*Embedding view*

API
===

Endpoint: **/projects/{project_pk}/embeddings/**

Example:

.. code-block:: bash

    curl -X GET "http://localhost:8000/api/v1/projects/9/embeddings/" \
         -H "Authorization: Token 8229898dccf960714a9fa22662b214005aa2b049"


Delete
*******

GUI
=====

Navigate to **Models -> Embeddings** and click on the three dots under **Actions** column and choose **Delete** (:numref:`embedding_actions`).


.. _embedding_actions:
.. figure:: images/embedding/embedding_actions.png

	*Embedding actions*

API
===

Endpoint: **/projects/{project_pk}/embeddings/{embedding_id}**

Example:

.. code-block:: bash

    curl -X DELETE "http://localhost:8000/api/v1/projects/9/embeddings/9/" \
         -H "Authorization: Token 8229898dccf960714a9fa22662b214005aa2b049"

Edit
*******

GUI
=====

Navigate to **Models -> Embeddings** and click on the three dots under **Actions** column and choose **Edit** (:numref:`embedding_actions`).


API
===

Endpoint: **/projects/{project_pk}/embeddings/{embedding_id}**

.. code-block:: bash

    curl -X PATCH "http://localhost:8000/api/v1/projects/9/embeddings/8/" \
         -H "accept: application/json" \
         -H "Content-Type: application/json" \
         -H "Authorization: Token 8229898dccf960714a9fa22662b214005aa2b049" \
         -d '{"description":"changed"}'


Apply phraser
*************

GUI
=====

Navigate to **Models -> Embeddings**, click on the three dots under **Actions** and choose **Phrase** (:numref:`embedding_actions`).
Insert text that you want to phrase and click **Post**. You should see phrased version of the text (:numref:`apply_phraser`).

.. _apply_phraser:
.. figure:: images/embedding/phraser_gui.png

	*Apply phraser*

API
===

Endpoint: **/projects/{project_pk}/embeddings/**

Example:

.. code-block:: bash

        curl -X POST "http://localhost:8000/api/v1/projects/9/embeddings/8/phrase_text/" \
        -H "accept: application/json" \
        -H "Content-Type: application/json" \
        -H "Authorization: Token 8229898dccf960714a9fa22662b214005aa2b049" \
        -d '{
                "text": "Venus is the second planet from the Sun."
            }'