/Multi-Modal-Sentiment-Analysis-with-Transformers

This project leverages the power of transformer models to perform sentiment analysis on both text and images. It uses BERT for text sentiment analysis and a pre-trained vision transformer (ViT) for image sentiment analysis.

Primary LanguageJupyter NotebookMIT LicenseMIT

Multi-Modal-Sentiment-Analysis-with-Transformers

In this project, I have performed the sentiment analysis on both text and images using different tranformer models. It uses BERT for text sentiment analysis and a pre-trained vision transformer (ViT) for image sentiment analysis. The goal is to combine these modalities to predict the overall sentiment of a given multi-modal input.