Repository for "Evaluating the Efficacy of Foundational Models in Time Series Anomaly Detection and Prediction: A Critical Analysis" Paper

This repository contains derived datasets, analysis of methods experimented and results in the paper titled "Evaluating the Efficacy of Foundational Models in Time Series Anomaly Detection and Prediction: A Critical Analysis"

1. Data Preprocessing

The datasets used in this study are available at: https://drive.google.com/drive/folders/1vYpNA1VgcAs6Aan0jO14DlkAu5BUfMRD?usp=sharing

2. FPT

This folder includes the replicated codes of the FPT(Frozen Pretrained Transformer) model for the datasets used in this study.

3. TimeGPT

This folder includes the codes of the TimeGPT model for the datasets used in this study.

Acknowledgement

We appreciate the following github repos for their valuable code base or datasets:

https://github.com/DAMO-DI-ML/NeurIPS2023-One-Fits-All

https://github.com/smtmnfg/NSF-MAP

https://raw.githubusercontent.com/Nixtla/transfer-learning-time-series/main/datasets/

https://github.com/thuml/Time-Series-Library