Scripts to find reoccuring patterns in time series.
Project description
Time Series Pattern Finding
The TSPatternFinding package is a collection of Python scripts aimed at finding reoccuring patterns of various types in time series data.
This package is aimed at time series data extracted from network logs and packet captures. However, most functions are generic enough to be applied to any time series data. Though results are presented with a network event focus.
Approaches
Stomp
STOMP is a highly efficient approach to "time motifs" discovery in time series data and one of several of USR's Matrix Profile data mining approaches (SCRIMP). Time motifs are repeating patterns which indicate an underlying common cause. Details on these approaches can be found here.
Haar
The Haar approach is based on a full decomposition of the time series using haar wavelets. A detailed description of this approach can be found here..
Most of the configuration for a Python project is done in the setup.py
file,
an example of which is included in this project. You should edit this file
accordingly to adapt this sample project to your needs.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for TSPatternFinding-0.0.1-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ea2768d23d2f39af9a23cc7a5e3d17a2b263dfd58f565de9c1395759eeac3b9c |
|
MD5 | f52975a17f384d803f72b3b24ae240cc |
|
BLAKE2b-256 | 9a35f17c69d7ad6a9ad93eb710d914d73ea7f9c732fc0b869d04aa159dd9d0c0 |