mefsantos/tsanalyse

A python package for (physiological) time series analysis

Installation Notes

Pre-requisites:

• Python
• pip
• python-dev

Installing pip:

    curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
    sudo python get-pip.py

Further Information on pip:

    https://pip.pypa.io/en/stable/installing/#do-i-need-to-install-pip

This tool may also require some external libraries in order to work on 64-bit processors. namely:

• lib32z1
• lib32stdc++6

To install the package run the following command:

    python setup.py install

This setup will install the package and build the external compressors located under algo/ according to your OS.

External Compressors

This tool was implemented to support additional compressors, namely paq8l and ppmd. algo/ contains the paq8l sources and ppmd binaries. If the code/binaries no longer work on your system try using your archive manager to install them.

Linux (Debian)

(see https://debian.pkgs.org/7/debian-main-amd64/ppmd_10.1-5_amd64.deb.html)

TL;DR:

    sudo apt-get update
    sudo apt-get install ppmd

MacOS

(see https://guide.macports.org/chunked/using.html#using.port.selfupdate also https://github.com/macports/macports-ports/blob/master/archivers/ppmd/Portfile)

(ppmd is only available on mac-ports)

Mac-ports:

    sudo port selfupdate
    sudo port install ppmd

Documentation

The tool set is extensively documented using pydoc. To launch the graphical interface run:

    pydoc -g

in the command line.

TSFilter

This interface should be used before any other as it filters the input data and enables to obtain the dataset in the correct format.

Examples :

Retrieve the hrf within the limits [50, 250]:

    ./TSFilter.py unittest_dataset filter -lim

Retrieve the timestamps and hrf:

    ./TSFilter.py unittest_dataset filter -kt

Retrieve the hrf from the second column of the input file

    ./TSFilter.py unittest_dataset filter -col 2

TSAnalyseDirect

TSAnalyseDirect allows for operations to be applied to files or directories (no subdirectories).

Examples :

• Compress

  Compress using the gzip algorithm (maximum compression level will be used)

    ./TSAnalyseDirect.py unittest_dataset_filtered compress -c gzip
  

  Compress using the brotli algorithm (maximum compression level will be used) and return an additional column with the compression ratio

    ./TSAnalyseDirect.py unittest_dataset_filtered compress -c brotli -cr
  

  Compress using the bzip2 algorithm with minimum compression(1 in this case):

    ./TSAnalyseDirect.py unittest_dataset_filtered compress -c bzip2 --level 1
  

• Entropy

  Calculate the entropy using Approximate entropy with tolerance 0.2 and matrix dimension 2 (reference values for the analysis of biological data)

    ./TSAnalyseDirect.py unittest_dataset_filtered entropy apen -t 0.2
  

• Stv Compute short-term variability using the Arduini algorithm

    ./TSAnalyseDirect.py unittest_dataset_filtered stv -algo arduini
  

TSAnalyseFileBlocks

TSAnalyseFileBlocks partitions the input files and computes the entropy and compression.

Examples:

• Compress

   Cut files into 5min blocks with no overlap and compress each one with the default compressor
    
    ./TSAnalyseFileBlocks.py unittest_dataset_filtered/ -s 300 compress
    
    
   Cut files into blocks with 300 lines with no overlap and compress each one with the default compressor
    
    ./TSAnalyseFileBlocks.py unittest_dataset_filtered/ -s 300 --use-lines compress
  

• Entropy

   Cut files into blocks with 5 min where one block starts 1 min later then the previous one did.
   Calculate each files entropy using the Sample entropy.
    
    ./TSAnalyseFileBlocks.py unittest_dataset_filtered/ -s 300 -g 60 entropy sampen
  

TSAnalyseMultiScale

The last interface TSAnalyseMultiScale is specific to calculate MultiScale entropy and compression:

Examples:

• Multiscale entropy for all the files starting at scale 1(original files) and ending in scale 20

    ./TSAnalyseMultiscale unittest_dataset_filtered entropy sampen
  

• Multiscale compression with rounded results for scale, since the scales are constructed by averaging a given number of point we are bound to have floats, this options rounds those numbers to an integer.

    ./TSAnalyseMultiscale unittest_dataset_filtered --round-to-int compression
  

• Multiscale compression with rounded results for scale, multiplied by 10, the scale point is multiplied by 10 and rounded.

    ./TSAnalyseMultiscale unittest_dataset_filtered --round-to-int --multiply 10 compression -c paq8l
  

---

Copyright (C) 2012 Mara Matias
Edited by Marcelo Santos - 2016

This file is part of TSAnalyse.

TSAnalyse is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published
by the Free Software Foundation, either version 3 of the License,
or (at your option) any later version.

TSAnalyse is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with TSAnalyse.  If not, see
<http://www.gnu.org/licenses/>.