--- title: "Training a custom extraction model with Document Intelligence Studio | Twyzer" description: "How to extract fields, tables, checkmarks and signatures from your own documents with a custom extraction model in Azure Document Intelligence Studio." url: "https://twyzer.nl/blogs/training-a-custom-extraction-model-with-document-intelligence-studio/" language: "nl-NL" --- Blog # Training a custom extraction model with Document Intelligence Studio Technologie13 juni 2024 ![Portret van Michelle Barlo, Engineer bij Twyzer](https://twyzer.nl/media/uploads/2026/07/1Y1A6408-e1786612601175.jpg?width=200 "Michelle Barlo") Geschreven door Michelle Barlo - Engineer In this blog, I will show you how you can build your own custom extraction model using the Document Intelligence Studio. If this is your first time looking at the Document Intelligence Studio I recommend looking at this blog first: [First look at Document Intelligence Studio](https://twyzer.nl/blogs/first-look-at-document-intelligence-studio/) ![Illustration of three documents next to a circuit icon with a gear](https://twyzer.nl/media/uploads/2026/08/ThumbnailCustomExtraction.png?width=1200 "Custom extraction thumbnail") ![](https://twyzer.nl/static/gradient.jpg) Link: [Document Intelligence Studio](https://documentintelligence.ai.azure.com/studio) ## Blogs This blog is part in a series of blogs about the capabilities of the Azure Document Intelligence resource. - [First look at Document Intelligence Studio](https://twyzer.nl/blogs/first-look-at-document-intelligence-studio/) - [Training a custom extraction model with Document Intelligence Studio](https://twyzer.nl/blogs/training-a-custom-extraction-model-with-document-intelligence-studio/) - [Training a custom classification model with Document Intelligence Studio](https://twyzer.nl/blogs/training-a-custom-classification-model-with-document-intelligence-studio/) - [Training a custom classification model with Word or Excel Documents](https://twyzer.nl/blogs/training-a-custom-classification-model-with-with-word-or-excel-documents/) ## Requirements - Azure ‘Document Intelligence’ or ‘Azure AI services multi-service account’ resource - Azure ‘Storage account’ - Minimal of 6 Documents (5 training documents and 1 test document) ## Project setup Open the custom extraction model page in Document Intelligence Studio. Scroll down the page and select Create a project in the My projects list. ![Document Intelligence Studio tile Custom extraction model](https://twyzer.nl/media/uploads/2026/08/ExtractionModel.png?width=1200 "Custom extraction model") ![Document Intelligence Studio My Projects with Create a project highlighted](https://twyzer.nl/media/uploads/2026/08/My-projects-1.png?width=1200 "Create an extraction project") This will open a wizard with four steps for setting up your project. 1. Provide a name and description for the project. 2. Select the Document Intelligence resource and API version you want to use. 3. Select the storage account, container, and folder to store the training data. 4. Review and create your project. ## Analyze documents Upload the files you want to use to train your model. (1) The first time you upload documents you get the option to _run layout_ or _auto label_ for all documents. Select Run layout on this screen or use the _Run layout_ button for each document you uploaded (2). When you use the Free tier (F0) of the Document Intelligence service, it can happen that not all your documents can be analyzed at once because of the rate limit. In that case wait for the appointed time and try again, or upgrade to a paid tier. In the middle of the screen you can view the selected document. (3) You can see all the pages, search for text in the document, zoom in and out and rotate it accordingly. ![Document Intelligence Studio label screen with an unlabeled packing list](https://twyzer.nl/media/uploads/2026/08/ScreenUnlabeled-1024x469-1.png?width=1200 "Unlabeled document") ## Fields With all files uploaded and analyzed we can start defining the fields we want to extract. Press the ‘Add a field’ button to create your first field. (4) There are 4 different types of fields. - Field: A text field - Selection mark: A true or false value based on a checkmark. - Signature: An image from within a bounding box - Table : A dynamic or fixed table ![Add a field menu with Field, Selection mark, Signature and Table](https://twyzer.nl/media/uploads/2026/08/fields.png?width=1200 "Add a field") A Table can be customized by selecting the table field after creation on the right side of the screen. You can rename, insert, and delete a column or row by using the menu under the chevron. ![Table labeling with a column menu to rename, insert or delete a column](https://twyzer.nl/media/uploads/2026/08/editTable.png?width=1200 "Edit a table") ## Training When you are happy with the fields you created you can assign values to your fields. Analyzed documents have yellow highlighted texts. You can select these texts and specify the field. Do this for each field of each document. ![Selected address text on an invoice being labeled as Shipping address](https://twyzer.nl/media/uploads/2026/08/FieldSelection.png?width=1200 "Select a field") When you want to train a signature or a selection mark field you can use the Draw region option and draw the bounding box in the correct location on the document. ![Drawing a bounding box with the mouse cursor](https://twyzer.nl/media/uploads/2026/08/bounding-box.png?width=1200 "Draw a bounding box") Tables can be trained in 2 different ways. **Manually:** by selecting the text that you want to store in a specific cell of your table. Then select the cell you want to store it in. ![Table rows labeled manually on a packing list](https://twyzer.nl/media/uploads/2026/08/TableManual-1024x260-1.png?width=1200 "Manual table labeling") **Automatically:** if the layout has detected a table in your document you can use the ![](https://twyzer.nl/media/uploads/2026/08/oLuFKHAAAAAElFTkSuQmCC.png)table icon to open a wizard that helps you auto label your table using a wizard. ![Table Labeling dialog with automatically labeled items](https://twyzer.nl/media/uploads/2026/08/TableAuto-1024x477-1.png?width=1200 "Automatic table labeling") When all fields of all the training documents are assigned, you can train your model by selecting the train button in the top right corner of the screen. Be patient, training a model can take a little while. You can see the status of your model by navigating to the _Models_ tab from the menu on the left of the screen. ## Testing When the model is ready you can test it using the _Test_ (1) option in the menu. Upload the file you want to use to test your model (2), Do not use the same files you used to train the model. Select _Run analysis_ (3) on the file. On the right side of the screen you will see the results of the analysis. (4). On the first tab _Fields_ you will see your fields and the values it extracted. It also provides you with a percentage. This represents the probability that this is correct. The more documents you use to train a model, the more accurate it will be. On the _Result_ tab you will see the same values but in Json format. And _Code_ will give you examples of how to call the model using Python, C# or JavaScript. ![Document Intelligence Studio test screen with the extracted fields of a packing list](https://twyzer.nl/media/uploads/2026/08/TestScreen-1024x486-1.png?width=1200 "Test the extraction model") ## Composed models You can combine multiple custom models in one endpoint by creating a composed model. This can be helpful if you have versions of the same document that are too different or if you do not know what type of document will be provided. The composed model will first classify the document to determine which models to use before analyzing the document. To do this you first need to train all your custom models individually. For easy maintenance make a separate project per model. From the Models tab select all the models you want the composed model to contain. Then press the Compose button and give your composed model a name and description. When you test your composed model you can see what model was used to extract the data by looking at the DocType in the results pane. ## Bouw verder op een fundament dat klopt Je wilt vooruit met data en AI. Maar wel vanuit een omgeving waarop je kunt vertrouwen. Wij helpen je grip te houden, juist wanneer keuzes complex worden. Laten we samen kijken welke stap jouw organisatie nu verder helpt. [Neem contact met ons op](https://twyzer.nl/contact/) [Neem contact met ons op](https://twyzer.nl/contact/) ![](https://twyzer.nl/static/pattern.png)![](https://twyzer.nl/static/gradient.jpg) ![Illustratie van servers die verbonden zijn met de cloud](https://twyzer.nl/media/uploads/2026/05/data.png?width=800 "Illustratie data-infrastructuur") Blogs ## Ook interessant Gebruik de pijltjestoetsen om door de carousel te navigeren. Gebruik Home en End om naar het begin of einde te gaan. [![Twee lachende collega's aan een bureau met meerdere schermen](https://twyzer.nl/media/uploads/2026/07/1Y1A6548.jpg?width=800 "Grip op retail begint bij data die je kunt vertrouwen")Actueel28 oktober 2026Grip op retail begint bij data die je kunt vertrouwen](https://twyzer.nl/blogs/betrouwbare-data-retail/) [![Lachende collega achter een sta-bureau met meerdere schermen vol code](https://twyzer.nl/media/uploads/2026/07/1Y1A6817.jpg?width=800 "Grip op Azure-kosten begint met inzicht en eigenaarschap")Actueel1 oktober 2026Grip op Azure-kosten begint met inzicht en eigenaarschap](https://twyzer.nl/blogs/azure-kosten-beheersen/) [![Over-de-schouderfoto van een ontwikkelaar die code schrijft op een laptop op een standaard](https://twyzer.nl/media/uploads/2026/07/1Y1A6649.jpg?width=800 "AutoMapper’s Licensing and Security Dilemma")Technologie7 september 2026AutoMapper’s Licensing and Security Dilemma](https://twyzer.nl/blogs/automappers-licensing-and-security-dilemma/)