In the current code, as per my understanding - the default Spacy model (en_core_web_lg) is utilised when the AnalyzerEngine() is instantiated.
It would be helpful if we could pass in parameters which would indicate the model to be used in the Analyser. Accordingly, based on the compute and accuracy requirements, folks could toggle the model used. This would also help in loading models for different languages.
Thanks for bringing this to our attention. we have not benchmarked this particular validator for latency yet. we will pick up benchmarking as a task soon and let you know how it goes.
In the current code, as per my understanding - the default Spacy model (en_core_web_lg) is utilised when the AnalyzerEngine() is instantiated.
It would be helpful if we could pass in parameters which would indicate the model to be used in the Analyser. Accordingly, based on the compute and accuracy requirements, folks could toggle the model used. This would also help in loading models for different languages.
Ref; https://github.com/microsoft/presidio/blob/main/docs/samples//python/customizing_presidio_analyzer.ipynb