`OpenNRE` implement SentenceREDataset based on SQLite storage

nicolay-r / ARElight

Granular Viewer of Sentiments Between Entities in Massively Large Documents and Collections of Texts, powered by AREkit

https://link.springer.com/chapter/10.1007/978-3-031-56069-9_23

MIT License

37 stars 2 forks source link

`OpenNRE` implement SentenceREDataset based on SQLite storage #132

Closed nicolay-r closed 11 months ago

nicolay-r commented 11 months ago

Quick serialization test:

python3 -m arelight.run.infer      --sampling-framework "arekit"     --ner-framework "deeppavlov"     --ner-model-name "ner_ontonotes_bert"     --ner-types "ORG|PERSON|LOC|GPE"     --terms-per-context 50     --sentence-parser "nltk:english"     --tokens-per-context 128    --from-files ~/Desktop/TwExportly_JeffBezos_tweets_2023_10_11.csv

You may see that the significant amout of columns were missed: Before in jsonl:

AREkit bug, fixed in 0.25.0 https://github.com/nicolay-r/AREkit/issues/536