This repository documents my Springboard Machine Learning Engineer capstone project. This project involved the full lifecycle of data collection, analysis, wrangling, and modeling. The project adapts a Bert-based model to be used for classification of text from dozens of publicans. It was built using both AWS and Google Cloud.
-
data-collection contains code used to crawl dozens of news and opinion websites.
-
data-wrangling contains Jupyter notebooks outlining evaluation of the collected data and analysis use to wrangle the data into a usable format.
-
modeling contains code and notebooks used to fine tune a BERT model to classify news stories.
-
api contains code used to host an API based on my model.
-
mini-projects contains Jupyter notebooks submitted as part of classwork for the Springboard Machine Learning Engineer Bootcamp. This code is not directly related to the capstone project.