Initial commit

pretrained/

results/

logs/

runs/

temp/

run_copy.sh

__pycache__

 No newline at end of file

MIT License

Copyright (c) 2021 d-ailin

Permission is hereby granted, free of charge, to any person obtaining a copy

of this software and associated documentation files (the "Software"), to deal

in the Software without restriction, including without limitation the rights

to use, copy, modify, merge, publish, distribute, sublicense, and/or sell

copies of the Software, and to permit persons to whom the Software is

furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all

copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE

AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE

SOFTWARE.

# GDN

Code implementation for : [Graph Neural Network-Based Anomaly Detection in Multivariate Time Series(AAAI'21)](https://arxiv.org/pdf/2106.06947.pdf)

# Installation

### Requirements

* Python >= 3.6

* cuda == 10.2

* [Pytorch==1.5.1](https://pytorch.org/)

* [PyG: torch-geometric==1.5.0](https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html)

### Install packages

```

    # run after installing correct Pytorch package

    bash install.sh

```

### Quick Start

Run to check if the environment is ready

```

    bash run.sh cpu msl

    # or with gpu

    bash run.sh <gpu_id> msl    # e.g. bash run.sh 1 msl

```

# Usage

We use part of msl dataset(refer to [telemanom](https://github.com/khundman/telemanom)) as demo example. 

## Data Preparation

```

# put your dataset under data/ directory with the same structure shown in the data/msl/

data

 |-msl

 | |-list.txt    # the feature names, one feature per line

 | |-train.csv   # training data

 | |-test.csv    # test data

 |-your_dataset

 | |-list.txt

 | |-train.csv

 | |-test.csv

 | ...

```

### Notices:

* The first column in .csv will be regarded as index column. 

* The column sequence in .csv don't need to match the sequence in list.txt, we will rearrange the data columns according to the sequence in list.txt.

* test.csv should have a column named "attack" which contains ground truth label(0/1) of being attacked or not(0: normal, 1: attacked)

## Run

```

    # using gpu

    bash run.sh <gpu_id> <dataset>

    # or using cpu

    bash run.sh cpu <dataset>

```

You can change running parameters in the run.sh.

# Others

SWaT and WADI datasets can be requested from [iTrust](https://itrust.sutd.edu.sg/)

# Citation

If you find this repo or our work useful for your research, please consider citing the paper

```

@inproceedings{deng2021graph,

  title={Graph neural network-based anomaly detection in multivariate time series},

  author={Deng, Ailin and Hooi, Bryan},

  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},

  volume={35},

  number={5},

  pages={4027--4035},

  year={2021}

}

```

M-6

M-1

M-2

S-2

P-10

T-4

T-5

F-7

M-3

M-4

M-5

P-15

C-1

C-2

T-12

T-13

F-4

F-5

D-14

T-9

P-14

T-8

P-11

D-15

D-16

M-7

F-8

 No newline at end of file