# Document classification

In order to label documents to train the document classification model, create a document classification task (see [tasks](https://docs-old.app.metamaze.eu/project/training/tasks "mention")). Once you start the task, you will arrive upon the following screen:

![](https://382487820-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LuWehHupT-Y9kwe-PJb%2Fuploads%2FjoMSO5oYcVi2IypnRZYX%2Fimage.png?alt=media\&token=f5fc7e77-8059-418b-96d5-3e53c8529f40)

Select the correct document type and language in the drop-down list, and click "Done". If the document should not be included in the training data, for instance because it is irrelevant, click "Reject". A pop-up window will open, where you have to specify the reson for rejecting the document. 

After clicking either "Done" or "Reject", the next document to be labeled will automatically be loaded.