> ## Documentation Index
> Fetch the complete documentation index at: https://docs.abbyy.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Set up a Segmentation activity

> Add a Segmentation activity, select segment fields, train it in Fast or Thorough mode, and review extraction results in Advanced Designer.

You can use a separate document set to train your Segmentation activity. To do so, select the Segmentation activity from the drop-down list next to the skill name. Then, in the drop-down list to the left of the **Upload** button, select the necessary document set or click **Create Set...** to create a new one. You can upload, delete, and rotate documents on this tab as described in the [Documents](/vantage/documentation/advanced-designer/document-skills/documents) section.

<Steps>
  <Step title="Add the activity">
    On the **Activities** tab, add a Segmentation activity to the document processing flow. Please note that your Segmentation activity should precede the activity that will extract the fields from the text segments.
  </Step>

  <Step title="Select fields">
    On the **Activity Properties** pane, select all the fields that correspond to the segments to be extracted.

    <Note>
      Only fields of type Text that have data type set to Text are supported.
    </Note>
  </Step>

  <Step title="Open Activity Editor">
    Click **Activity Editor**. Make further modifications to the document labeling on the **Fields** tab if required.
  </Step>

  <Step title="Train the activity">
    Click **Train Activity**. Training can be performed in **Fast** or **Thorough** mode.

    * **Fast** mode is selected by default. This mode will work even on small document sets, and the activity will be trained quickly.
    * If you are not satisfied with the results obtained in **Fast** mode, consider switching to **Thorough** mode that trains a Deep Learning model. This mode requires more documents in the training set and takes longer to train, but it will be able to perform better on a wide variety of documents. The document set must contain at least 50 labeled documents, but we recommend having at least 150 labeled documents. To switch to **Thorough** mode, use a drop-down menu next to the **Train Activity** button.
    * You may want to test both modes and choose the one that works best for your documents.

    <Note>
      **Thorough** mode will only work with English-language documents.
    </Note>
  </Step>

  <Step title="Review results">
    Once the activity has been trained, activity testing will start automatically. After testing has completed, navigate to the **Results** tab and analyze the field extraction results for your activity. Statistics displayed on the **Results** tab are identical to the general statistics for the skill displayed on the Results tab. If required, make any necessary changes to your labeling and train the activity again.
  </Step>
</Steps>

<Note>
  The activity can only be trained and tested using documents with confirmed labeling. Documents have unconfirmed labeling if the reference labeling was generated automatically based on the predicted labeling, unless you copy predicted labeling to reference using the corresponding option in the document context menu. You can check the labeling status for each document on the **Documents** tab. To confirm labeling for a document, you should review it on the **Fields** tab.
</Note>

Supported languages: English, Russian, German, French, Spanish, Italian, Portuguese (Standard), Japanese, and Dutch.