WeatherAnomalyDetectionClassification

This repository holds the scripts and reports for a project on time series anomaly detection, time series classification & dynamic time warping, performed on a dataset of Canadian weather measurements. The data was sourced from OpenML, shared by user Elif Ceren Gök.

Time series anomaly detection

Multivariate time series anomaly detection using PyOD algorithms & the Darts package: K-means clustering, Gaussian Mixture Models, ECOD, Isolation Forest and an Autoencoder with PyTorch Lightning. Visualizing & comparing the results with multiple plots, including 3D interactive Plotly scatterplots.
Full report
Scripts, Lightning classes, functions

Time series classification

Multivariate time series classification using sktime and pyts: kNN with DTW distance, ROCKET & Arsenal, WEASELMUSE and a PyTorch Lightning convolutional neural network trained on image transformed data. Visualizing & comparing the performances of all algorithms.
Full report
Scripts, Lightning classes, functions