Saxify

An implementation of Symbolic Aggregate approXimation (SAX) in python.

Table of Contents

• Prerequisites
• Quickstart
• Project's Structure
• How to Run
  • Jupyter Notebook
  • Python
• Appendix
  • Quiz Instruction
  • Result
• License

Prerequisites

• Docker Community Edition 17.12.0-ce or higher
• Docker Compose 1.18.0 or higher
• Yarn 1.3.2 or higher (a npm replacement)

Quickstart

1. Start JupyterLab by running docker-compose up or yarn start.
2. Copy and paste the URL with a token key shown in the shell output into your browser to access JupyterLab.

Project's Structure

• All data are located in workspace/data directory.
  1. Time series data: ts_a.txt and ts_b.txt
  2. Normalized time series data: ts_a_normalized.txt and ts_b_normalized.txt
  3. Result: result.txt
• Source code
  1. Jupyter Notebook: saxify.ipynb
  2. Python: saxify.py

How to Run

Jupyter Notebook

1. Open saxify.ipynb by double-clicking this file in the left panel.
2. Go to Run > Run All Cells

Python

1. Open terminal from JupyterLab: File > New > Terminal.
2. Change a directory to /home/jovyan/work/workspace by running cd /home/jovyan/work/workspace.
3. Run python3 saxify.py -a ./data/ts_a.txt -b ./data/ts_b.txt -x 7 -y 5 -z 6

Appendix

Quiz Instruction

Download here

Result

5730329521	Parinthorn Saithong (Aof)
X = 7
Y = 5
Z = 6
n = 128
w = 32
SAX_A = EDDCEDCCCBCDBBDECDDDCCDDDDECBDEC
SAX_B = CDCCCDFFCCBDCBDCDBEBECFBCECCCCDC
Distance between SAX_A & SAX_B = 3.82

License

MIT © Parinthorn Saithong