If I use torchserve api, inference does not work.

[EvelinaAleksiutenko]

Hi! When I run torch-serve start and after it run streamlit with lit-ner.py, I set LOCAL = True to use pytorchserve API. But it runs endlessly without an error and without an inference.

[cceyda]

Sorry I haven't maintained this for a while so maybe something changed on the dependencies.

Did you also modify the API_URL and MODEL_NAME at https://github.com/cceyda/lit-NER/blob/master/lit_ner/lit_ner.py#L12
Do you see any errors on the torchserve logs?

Also try if this works;

curl -X POST http://[API_URL]/predictions/[MODEL_NAME]
   -H 'Content-Type: application/json'
   -d '{"data":"my test sentence"}'

& let me know of any error messages you see