Hi Prodigy Team,
I do no have a good GPU on my computer to train my models. I use a GPU node in a cluster. How do you submit the jobs to train in a cluster?It is like Spacy? I'm sorry if is in the documentation but I have not been able to find it.
Also, is there any documentation on how to visualize and extract the entity relations?
Prodigy uses spaCy under the hood and you can always run
prodigy data-to-spacy to export your data and train with spaCy directly: https://spacy.io/api/cli#train That's also what we'd recommend if you're training on a remote machine.
The easiest way would probably be to load the data back into Prodigy using the
relations UI. You can run
db-out to save the annotations to a file and then load them back in as the input source.