Hi,
Currently, we run a serverless text classification endpoint in sagemaker that returns predictions as:
[{'label': 'POSITIVE', 'score': 0.9998743534088135}]
Since we’d like to store the prediction in a database it would be very beneficial if the model could also return the label_id
as configured in the id2label
and label2id
part.
For example:
[{'label_id': 1, 'label': 'POSITIVE', 'score': 0.9998743534088135}]
Any ideas how to configure that? I checked the pipeline docs but wasn’t successful…
Regards,
David