Classifying New Testament Authors Using Huggingface Transformers
This is my first post in a while! A lot has changed in the last year. I’ve been coding almost entirely in Python for work, and my next posts will reflect that switch in emphasis.
This project classifies New Testament authorship using Huggingface Transformers (specifically distilBERT) and achieves ~85% F1 score on the test data. You can find the notebook itself at this Colab link!
| eval_loss | eval_Accuracy | eval_F1 | eval_Precision | eval_Recall | |
|---|---|---|---|---|---|
| train | 0.168601 | 0.947090 | 0.947000 | 0.948010 | 0.947224 |
| val | 0.485929 | 0.856073 | 0.854586 | 0.856452 | 0.855263 |
| test | 0.481824 | 0.853398 | 0.848969 | 0.852319 | 0.850766 |
Chad Peltier
My name is Chad Peltier and I the Head of Data & Integration for the US at Janes. I am interested in data science for social good, NLP, and GEOINT data.