Skip to content

Latest commit

 

History

History
21 lines (15 loc) · 662 Bytes

readme.md

File metadata and controls

21 lines (15 loc) · 662 Bytes

✨ Am implement for Donut, a univariate time series anomaly detection algorithm, with pytorch .✨ Reference to flower-kyo/pytorch-donut.

Introduction

We use the cpu4.csv in NetManAIOps/donut as a test dataset. The result in different epochs are as follows:

1 epoch:

10 epochs:

100 epochs:

250 epochs:

Installation and runing

  1. Clone this repository
  2. cd pytorch-donut
  3. python main.py