TransFusion: Generating Long, High Fidelity Time Series using Diffusion Models with Transformers

Md Fahim Sikder, Resmi Ramachandranpillai, Fredrik Heintz
7/24/2023
cs.LG

Official code: fahim-sikder/TransFusion (25 stars, Apache-2.0, last commit 2025-04-15).

Abstract

The generation of high-quality, long-sequenced time-series data is essential due to its wide range of applications. In the past, standalone Recurrent and Convolutional Neural Network-based Generative Adversarial Networks (GAN) were used to synthesize time-series data. However, they are inadequate for generating long sequences of time-series data due to limitations in the architecture. Furthermore, GANs are well known for their training instability and mode collapse problem. To address this, we propose TransFusion, a diffusion, and transformers-based generative model to generate high-quality long-sequence time-series data. We have stretched the sequence length to 384, and generated high-quality synthetic data. Also, we introduce two evaluation metrics to evaluate the quality of the synthetic data as well as its predictive characteristics. We evaluate TransFusion with a wide variety of visual and empirical metrics, and TransFusion outperforms the previous state-of-the-art by a significant margin.

View on arXivView PDF

Code Implementations(1)

fahim-sikder/TransFusionOfficial
Apache-2.0

This is the official repository for the paper "TransFusion: Generating Long, High Fidelity Time Series using Diffusion Models with Transformers"

259Shell, PythonJul 28, 20231 years agoApache-2.0

Similar Papers

A Time-Series Data Augmentation Model through Diffusion and Transformer Integration

85% match2025

Utilizing Image Transforms and Diffusion Models for Generative Modeling of Short and Long Time Series

84% match1 repo652024

A Survey of Transformer Enabled Time Series Synthesis

82% match2024

ChronoGAN: Supervised and Embedded Generative Adversarial Networks for Time Series Generation

81% match1 repo102024

Generative Modeling of Networked Time-Series via Transformer Architectures

81% match2025

Rough Transformers for Continuous and Efficient Time-Series Modelling

80% match2024

Cite this paper

@article{sikder2023transfusion,
  title  = {TransFusion: Generating Long, High Fidelity Time Series using Diffusion Models with Transformers},
  author = {Md Fahim Sikder and Resmi Ramachandranpillai and Fredrik Heintz},
  year   = {2023},
  eprint = {2307.12667},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.12667v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.12667.svg)](https://codeofpaper.com/paper/2307.12667)

Discussion