General Document Ingestion

See guide to Solr for more information about Solr setup with ClarityNLP.

Solr has built-in APIs for ingesting documents, which are documented here. The simplest way is generally to use curl to upload JSON, CSV, or XML. Documents need to be pre-processed as plain text before they are uploaded into ClarityNLP.

Sample JSON upload for ClarityNLP:

curl -X POST -H 'Content-Type: application/json' 'http://localhost:8983/solr/report_core/update/json/docs' --data-binary '
        "report_type":"Report Type",
        "source":"My Institution",
        "report_text":"Report text here"

Upload Scripts

A collection of scripts for ingesting popular datasets (MIMIC, AACT Clinical Trials, Gleason Pathology Documents, etc.) is available here.