- From: Thad Guidry <thadguidry@gmail.com>
- Date: Wed, 30 Jun 2021 11:20:06 -0500
- To: public-reconciliation@w3c.org, Jeff Mixter <jeffmixter@gmail.com>
Received on Wednesday, 30 June 2021 16:20:30 UTC
Here's a recent tool that USC's ISI department has been working on that gives various aspects of some of our Reconciliation API and Recon services, but not using it directly. I've met Pedro and his team and students before. They do some interesting work at USC's ISI department. I thought it might be interesting for folks on here to browse through, especially around areas of Ranking, Ground Truth, Pipelining, Parallelization, TF-IDF, and even GPU resourcing, to name a few. It can be used against their other project, KGTK (Knowledge Graph Toolkit) which is Elasticsearch based, and I know a few of you are using that as a Recon service and cache. https://github.com/usc-isi-i2/table-linker @Jeff Mixter <jeffmixter@gmail.com> your team might also be interested in some of the candidate ranking functions like this one that use Torch's tensors for example: https://github.com/usc-isi-i2/table-linker/blob/master/tl/candidate_ranking/predict_using_model.py Thad https://www.linkedin.com/in/thadguidry/ https://calendly.com/thadguidry/
Received on Wednesday, 30 June 2021 16:20:30 UTC