Implement an environment for testing out the extractor configurations

[eriknovak]

Connected to a problem?

Currently, to validate the configuration and performance of the extractors, the user has to hardcode the configuration, run the anonymization process, manually check if the extracted values are ok, and repeat the process until the configuration is done correctly.

It would be nice to have a way to streamline this process.

Solution?

Implement a StreamLit solution, that can be run locally from the terminal, and would allow testing of the configuration on some of the uploaded texts. Perhaps extend the implementation and testing to the whole pipeline (i.e. extractors, generators, and anonymization strategies).

Alternatives?

No response