Pawan Kumar Naveen

@PawanKumar18 · User

GitHub profile ↗ · Compare

CS Grad Student at Northeastern University

Boston0 followers40 repositories

Repositories

PawanKumar18/U.S.-Patent-Phrase-to-Phrase-Matching-NLP

This project utilizes natural language processing techniques for matching phrases in US patents. It involves data loading, preprocessing, and tokenization with the DeBERTa V3 tokenizer, along with finetuning using the Hugging Face Transformers library. The task's evaluation metric is the Pearson correlation coefficient.

★ 1Jupyter NotebookForks 0