smile-yan/vae-anomaly-detection-for-timeseries

Anomaly detection for timeseries basing on Variational AutoEncoder.

vae-anomaly-detection-for-timeseries

中文文档

Tensorflow 2.x for timeseries implementation of Variational AutoEncoder for anomaly detection following the paper 《Variational Autoencoder based Anomaly Detection using Reconstruction Probability》.

dependencies

• tensorflow 2.x
• numpy
• pandas
• sklearn

usage

Firstly, clone this repository into your local environment.

$ git clone git@github.com:smile-yan/vae-anomaly-detection-for-timeseries.git

Then make sure the dependencies are installed.

$ cd vae-anomaly-detection-for-timeseries
$ pip install -r requirement.txt

Lastly we can run this project as follows:

$ python main.py

Custom dataset

Make sure that your dataset is a time series set and then do as in main.py.

Q & Α

Any question please push issues here or comment on my blog. > smileyan9

Smileyan 2022.9.6 14:28