Attendees:
Updates from contributors
-
Abdelrahman Jamal
- Completed project documentation for GSoC model.
- Looking for low hanging fruits.
- Found specific cases like
copyright <name>
which is missidentified. NER can help.
- Will be working for edge cases using NER. But pretrained models are not accurate enough in detecting name and orgs.
- Looking for smaller pre-trained models to use.
-
Soham Banerjee
- Working on Obligation endpoints. Some changes suggested by mentors on endpoint name.
GET
api is working, working on DELETE
. Will continue work on create and edit endpoints.
- Next endpoints to work on will be IPRA/Keyword, etc.
- Following the v2 guidelines for new endpoints.
-
Sushant Kumar
- Raised PR for ScanCode improvements #2569.
- The idea is to invoke ScanCode API only once for the entire upload.
- Working for final report available here.
-
Kavya Shukla
- Made changes on Audit, continuing work on Obligations.
- Showed demo of updated the audit and changes endpoint.
- Created data structures for obligations, had questions about the endpoints
- Following endpoints will be required for obligation:
- Create Obligation (with optional list of associated licenses).
- Edit the obligation
- Delete the obligation
- Add new licenses to associated license list of the obligation.
- Remove licenses from associated license list of the obligation.