GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → astha-chem → mvts-ano-eval

astha-chem / mvts-ano-eval

Licence: MIT License
A repository for code accompanying the manuscript 'An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series' (published at TNNLS)

Programming Languages

python
139335 projects - #7 most used programming language
shell
77523 projects

Labels

deep-learningtime-seriesanomaly-detection

Projects that are alternatives of or similar to mvts-ano-eval

Ad examples
A collection of anomaly detection methods (iid/point-based, graph and time series) including active learning for anomaly detection/discovery, bayesian rule-mining, description for diversity/explanation/interpretability. Analysis of incorporating label feedback with ensemble and tree-based detectors. Includes adversarial attacks with Graph Convolutional Network.
Stars: ✭ 641 (+2365.38%)
Mutual labels:  time-series, anomaly-detection
Pyodds
An End-to-end Outlier Detection System
Stars: ✭ 141 (+442.31%)
Mutual labels:  time-series, anomaly-detection
Rnn Time Series Anomaly Detection
RNN based Time-series Anomaly detector model implemented in Pytorch.
Stars: ✭ 718 (+2661.54%)
Mutual labels:  time-series, anomaly-detection
Deepadots
Repository of the paper "A Systematic Evaluation of Deep Anomaly Detection Methods for Time Series".
Stars: ✭ 335 (+1188.46%)
Mutual labels:  time-series, anomaly-detection
awesome-time-series
Resources for working with time series and sequence data
Stars: ✭ 178 (+584.62%)
Mutual labels:  time-series, anomaly-detection
Telemanom
A framework for using LSTMs to detect anomalies in multivariate time series data. Includes spacecraft anomaly data and experiments from the Mars Science Laboratory and SMAP missions.
Stars: ✭ 589 (+2165.38%)
Mutual labels:  time-series, anomaly-detection
Awesome Ts Anomaly Detection
List of tools & datasets for anomaly detection on time-series data.
Stars: ✭ 2,027 (+7696.15%)
Mutual labels:  time-series, anomaly-detection
Merlion
Merlion: A Machine Learning Framework for Time Series Intelligence
Stars: ✭ 2,368 (+9007.69%)
Mutual labels:  time-series, anomaly-detection
Lstm anomaly thesis
Anomaly detection for temporal data using LSTMs
Stars: ✭ 178 (+584.62%)
Mutual labels:  time-series, anomaly-detection
Adaptive Alerting
Anomaly detection for streaming time series, featuring automated model selection.
Stars: ✭ 152 (+484.62%)
Mutual labels:  time-series, anomaly-detection
Luminaire
Luminaire is a python package that provides ML driven solutions for monitoring time series data.
Stars: ✭ 316 (+1115.38%)
Mutual labels:  time-series, anomaly-detection
msda
Library for multi-dimensional, multi-sensor, uni/multivariate time series data analysis, unsupervised feature selection, unsupervised deep anomaly detection, and prototype of explainable AI for anomaly detector
Stars: ✭ 80 (+207.69%)
Mutual labels:  time-series, anomaly-detection
Anomalize
Tidy anomaly detection
Stars: ✭ 263 (+911.54%)
Mutual labels:  time-series, anomaly-detection
Adtk
A Python toolkit for rule-based/unsupervised anomaly detection in time series
Stars: ✭ 615 (+2265.38%)
Mutual labels:  time-series, anomaly-detection
Pycaret
An open-source, low-code machine learning library in Python
Stars: ✭ 4,594 (+17569.23%)
Mutual labels:  time-series, anomaly-detection
Getting Things Done With Pytorch
Jupyter Notebook tutorials on solving real-world problems with Machine Learning & Deep Learning using PyTorch. Topics: Face detection with Detectron 2, Time Series anomaly detection with LSTM Autoencoders, Object Detection with YOLO v5, Build your first Neural Network, Time Series forecasting for Coronavirus daily cases, Sentiment Analysis with BERT.
Stars: ✭ 738 (+2738.46%)
Mutual labels:  time-series, anomaly-detection
mtad-gat-pytorch
PyTorch implementation of MTAD-GAT (Multivariate Time-Series Anomaly Detection via Graph Attention Networks) by Zhao et. al (2020, https://arxiv.org/abs/2009.02040).
Stars: ✭ 85 (+226.92%)
Mutual labels:  time-series, anomaly-detection
khiva-ruby
High-performance time series algorithms for Ruby
Stars: ✭ 27 (+3.85%)
Mutual labels:  time-series, anomaly-detection
Matrixprofile
A Python 3 library making time series data mining tasks, utilizing matrix profile algorithms, accessible to everyone.
Stars: ✭ 141 (+442.31%)
Mutual labels:  time-series, anomaly-detection
Taganomaly
Anomaly detection analysis and labeling tool, specifically for multiple time series (one time series per category)
Stars: ✭ 200 (+669.23%)
Mutual labels:  time-series, anomaly-detection
View All Similar Projects ➔

Anomaly Detection on Time Series: An Evaluation of Deep Learning Methods

The goal of this repository is to evaluate multivariate time-series anomaly detection algorithms on a common set of datasets discussed in the paper:

A. Garg, W. Zhang, J. Samaran, R. Savitha and C. -S. Foo, "An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series," in IEEE Transactions on Neural Networks and Learning Systems, doi: 10.1109/TNNLS.2021.3105827. https://ieeexplore.ieee.org/document/9525836

Arxiv: https://arxiv.org/abs/2109.11428

Implemented Algorithms

Raw Signal

Raw signal passes as is. Base method for benchmarking.

PCA

PCA is used for lossy reconstruction.

Univar Auto-encoder

Channel-wise auto-encoders for each channel.

Autoencoder

Hawkins, Simon et al. "Outlier detection using replicator neural networks." DaWaK, 2002.

LSTM-ED

Malhotra, Pankaj et al. "LSTM-based encoder-decoder for multi-sensor anomaly detection." ICML, 2016.

TCN-ED

Based on the TCN benchmarked by Bai, Shaojie et al. "An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling." arxiv, 2018

LSTM VAE

Based on Park, Daehyung, Yuuna Hoshi, and Charles C. Kemp. "A multimodal anomaly detector for robot-assisted feeding using an LSTM-based variational autoencoder." IEEE Robotics and Automation Letters 3.3 (2018): 1544-1551.

Omni-anomaly

Su, Ya, et al. "Robust Anomaly Detection for Multivariate Time Series through Stochastic Recurrent Neural Network." Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019.

MSCRED

Zhang, Chuxu, et al. "A Deep Neural Network for Unsupervised Anomaly Detection and Diagnosis in Multivariate Time Series Data." Proceedings of the AAAI Conference on Artificial Intelligence. 2019.

Additional Algorithms Evaluated

These algorithms are also evaluated but could not be included in the repo at the moment as the original implementations lack a license.

DAGMM

Zong, Bo, et al. “Deep autoencoding gaussian mixture model for unsupervised anomaly detection.” ICLR, 2018

Telemanom (NASA LSTM)

Hundman, Kyle, et al. "Detecting spacecraft anomalies using LSTMs and nonparametric dynamic thresholding." Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining. 2018.

BeatGan

Zhou, Bin, et al. "BeatGAN: Anomalous Rhythm Detection using Adversarially Generated Time Series." IJCAI, 2019.

OCAN

Zheng, Panpan, et al. “One-class adversarial nets for fraud detection.” AAAI, 2019.

Data sets and how to get them

Swat

This data set contains 11 days of continuous operation in a water treatment testbed. 7 days’ worth of data was collected under normal operation while 4 days’ worth of data was collected with attack scenarios. During the data collection, all network traffic, sensor and actuator data were collected. To obtain the data one must make a request by filling out this form Download the files "SWaT_Dataset_Normal_v1.xlsx" and "SWaT_Dataset_Attack_v0.xlsx" and put it in the location /data/raw/swat/raw

Wadi

Similar to Swat this data set consists of 16 days of continuous operation, of which 14 days’ worth of data from a water distribution testbed was collected under normal operation and 2 days with attack scenarios. After submitting the same form as for swat, you will receive a link to access a google drive. Then both csv tables that can be found in the 02_WADI Dataset_19 Nov 2017 folder must be placed at the following location: /data/raw/wadi/raw

Damadics

This data set contains data from actuators collected over several days at the Lublin Sugar Factory. Data can be downloaded from this website, each file contains the data acquired during one day and the data from October 29th to November 22, 2001 should be downloaded and placed at the following location: /data/raw/damadics/raw

Skab

The Skoltech Anomaly Benchmark testbed consists of a water circulation system and its control system, along with a data-processing and storage system. Examples of the type of anomalies induced include partial valve closures, connecting shaft imbalance, reduced motor power, cavitation and flow disturbances. Train and test splits are provided by the authors. The dataset is available at https://www.kaggle.com/yuriykatser/skoltech-anomaly-benchmark-skab/version/1

Smap and Msl

Get data

You can get the public datasets (SMAP and MSL) using:

wget https://s3-us-west-2.amazonaws.com/telemanom/data.zip && unzip data.zip && rm data.zip

cd data && wget https://raw.githubusercontent.com/khundman/telemanom/master/labeled_anomalies.csv

Place the folder smap_msl under /data/raw/

Smd

SMD (Server Machine Dataset) is the folder here: https://github.com/NetManAIOps/OmniAnomaly/tree/master/ServerMachineDataset

Place the folder 'ServerMachineDataset' at the location /data/raw/

Usage

Put all the datasets in the data/raw/ directory.

git clone https://github.com/astha-chem/mvts-ano-eval.git
conda create -n mvtsenv python=3.6
source activate mvtsenv
python3 -m pip3 install --user --upgrade pip
pip install -r requirements.txt --user
# based on your cuda version or use the cpu only version
conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch 
# cpu only
# conda install pytorch==1.2.0 torchvision==0.4.0 cpuonly -c pytorch
python3 setup.py install
get_datasets.sh
python3 main.py
python3 metrics_expts.py

Authors/Contributors

Credits

This repository is forked from https://github.com/KDD-OpenSource/DeepADoTS Base implementation for AutoEncoder, LSTM-ED Base implementation for Omni-anomaly
Base implementation for VAE-LSTM Base implementation for MSCRED 1 Base implementation for MSCRED 2 TCN module

Note

The following algorithms have also been evaluated but could not be included in the repo at the moment as the original implementations lack a license:

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].