What data was this model trained with? Specifically the NER?

[owquresh]

Want to see what dataset the model was trained on, mainly the named entity recognition that's going on, is there a repo for that?

[omri374]

Presidio uses open-source models. The default one used is en_core_web_lg which was trained on the OntoNotes 5 corpus. Most common datasets are OntoNotes 5, CONLL2003, I2B2 2006, I2B2 2014, and a combination of those with synthetic data.