Skip to content

Latest commit

 

History

History
40 lines (24 loc) · 991 Bytes

README.md

File metadata and controls

40 lines (24 loc) · 991 Bytes
Raw

BERT_News_Classifier

NLP to classify news articles into catagories. Model inputs a news article as raw text and outputs one of the following categories: "business", "entertainment", "sport", "tech", or "politics".

Demo Video

BERT_Classifier_Demo.mov

Table of Contents

How to Download

git clone https://github.com/iansnyder333/BERT_News_Classifier.git
cd BERT_News_Classifier
python3.11 -m venv venv
source venv/bin/activate
pip3.11 install -r requirements.txt

How to Run

python3.11 main.py

Notes

Python 3.11 recommended, the app was not tested on any other versions so performance is not known.

Model was trained with 5 EPOCHS using bbc-text csv file from kaggle. Model has 97% accuracy on testing data but is still in development for accuratley classifying new articles.