Creating a revised annotation dataset, from the output of another NER model

I have output from an NER model which I want to use as a starting point to create a revised gold dataset, and use this gold dataset for addtional training with prodigy.

Since I already have annotations from the NER model, I want to be able to display these annotations on the annotation interface, and allow the user to make revised annotations.

How can i accomplish this please?

Hi! Prodigy's input and output formats are the same, so you can load in pre-labelled data in Prodigy's JSON format and the annotated spans will be pre-highlighted in the UI. See here for an example of the data format, which is pretty straightforward: https://prodi.gy/docs/api-interfaces#ner_manual

Here's an example of how to plug in a custom NER model and correct it's predictions manually: https://prodi.gy/docs/named-entity-recognition#custom-model