Closed
Description
Description
Add a flag in the api config to add transformed feature values to the prediction response. Or respond with these values automatically (without adding a flag to enable/disable it).
Motivation
Clients may need to perform post-processing on the prediction response based on the values of the transformed features (e.g. entity extraction requires the tokenized text).
Notes
- Add e.g.
-r/--raw
flag to CLI to show full json response?