https://github.com/vasugi2003/news_article_classification_using_llm_bert

News Category Classification with BERT - This repository contains implementation of a BERT-based model for classifying news articles into various categories. The model is fine-tuned on a dataset of news headlines and categories.
https://github.com/vasugi2003/news_article_classification_using_llm_bert

bert-model classification colab-notebook huggingface kaggle llm

Last synced: 7 months ago
JSON representation

News Category Classification with BERT - This repository contains implementation of a BERT-based model for classifying news articles into various categories. The model is fine-tuned on a dataset of news headlines and categories.

• Host: GitHub
• URL: https://github.com/vasugi2003/news_article_classification_using_llm_bert
• Owner: Vasugi2003
• Created: 2024-08-21T08:22:17.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-08-30T13:22:50.000Z (about 1 year ago)
• Last Synced: 2025-01-11T19:44:25.770Z (9 months ago)
• Topics: bert-model, classification, colab-notebook, huggingface, kaggle, llm
• Language: Jupyter Notebook
• Homepage:
• Size: 651 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: NEWS_Classification_and_generation.ipynb

Awesome Lists containing this project

README

          
**News Category Classification with BERT**

This repository contains a PyTorch implementation of a BERT-based model for classifying news articles into various categories. The model is fine-tuned on a dataset of news headlines and categories and includes TPU/GPU support for efficient training.

**Objective**

The goal of this project is to build a news category classifier using the BERT (Bidirectional Encoder Representations from Transformers) model. BERT is fine-tuned on a news dataset to predict the category of a news headline.

**Dataset**

The dataset used is the "News Category Dataset" from Kaggle, which consists of various news headlines and their corresponding categories. The dataset can be downloaded using the Kaggle API.

**Key Features**

TPU/GPU Support: The code is optimized to run on TPU or GPU using TensorFlow and PyTorch.

Fine-Tuning BERT: The BERT model (bert-base-uncased) is fine-tuned on the news dataset.

Early Stopping: The model training includes an early stopping mechanism to prevent overfitting.

Evaluation: The model is evaluated on a test set with metrics including accuracy, a classification report, and a confusion matrix.

Getting Started

Prerequisites

Python 3.x

TensorFlow

PyTorch

Hugging Face Transformers

Kaggle API

Other dependencies as specified in requirements.txt