This repository contains code for denoising seismic sections using a Vision Transformer(VIT). This is a base model trained to remove seismic noise from seismic images while preserving structural details, making it usefil for seismic interpretation and analysis.