Pre-Built Model

What is a Pre-trained model?

Nanonets pre-trained or prebuilt models help you extract key information from documents without having to gather data and then build and train your own models.

For example: Use our Receipts pre-built model to upload your own receipts and see data extracted instantly. You can integrate the model or download results instantly.

Watch the tutorial below to see how to extract data from a pre-trained model on the free plan.

Create a Pre-trained model (Video)

Create a Pre-trained model (Steps)

Choose a pre-trained model based on your document type.
Click on the model card to add it to your account.
You will be redirected to the Extract data screen for your model.
Click on 'Add files' to upload your document(s) to extract data from.
The documents may take 3-5 seconds to process, depending on their size.
Click on a document in the list to see the extracted and structured data.

How to Edit, Add or Delete data (Video)

How to Edit, Add or Delete data (Steps)

Edit extracted data:

Click on the box or field containing the wrongly extracted data.
You will see a modal on the image showing the extracted text and the label auto-assigned by the AI.
Drag the corners of the box to resize and re-capture the data, OR
Move the box to where the data is present anywhere else on the image.
Once you're happy with the extracted data, click Save.

Add missing data:

From the image, locate the text you want to extract.
Click and drag your cursor over the text to draw a box around it.
You will see a popup with the extracted text and a dropdown to assign a label.
Click on the 'Label' dropdown.
Select the label you want to assign it.
Click Save.
Click on the check icon next to the entry so that the model learns from this correction.

Delete an extra entry:

Click on the box or field containing the wrongly extracted data.
You will see a modal on the image showing the extracted text and the label auto-assigned by the AI.
You can delete this entry by clicking on the Delete icon on the bottom left corner of the popup.

Supported file formats

You can import files in PNG, JPEG or PDF format.