TimeSeries analysis toolbox for Matlab
Dylan R. Muir, Roth, Morgane, Antonin Blot · Zenodo (CERN European Organization for Nuclear Research) · 2020
This repository contains a Matlab toolbox for analysis of time-series data, such as imaging stacks or behavioural recordings. How do I get set up? Clone the repository https://bitbucket.org/DylanMuir/timeseriesanalysis to your machine where you want to work Decide whether you want to use the object-oriented toolbox (recommended; "development" branch) or the Matlab struct version (not recommended; "master" branch). Make sure you switch to the correct branch. Add the toolbox directory to the Matlab path Which version to use Use the object-oriented version of the toolbox unless you really can't ("development" branch). At some point the struct version ("master" branch) will no longer be updated. At present I am trying to keep everything in synch between the two versions. Getting started — Object-oriented version vtTimeTrace = 0:(1/100):10; tfSignal = repmat(linspace(0, 1, numel(vtTimeTrace)), [1 10 10]); ts = TimeSeries(vtTimeTrace, tfSignal); % Make a 2D time series ts.plot; % or plot(ts); plot the time series plot(ts + ts); % Add two time series ts.var_t() % or var_t(ts); Compute the variance of 'ts' over time plot(mean(ts)); % Plot the mean of all series within 'ts' plot(ts.choose(5)); % Plot only series '5' within 'ts' (linear indexing) tsEvent = TimeSeries(vtTimeTrace, [], 'event'); % Make an event time series tfTriggeredResponse = ContingentBinTS(tsEvent, ts, [], -1:.5:1); % Compute the triggered response of 'ts', at trigger events from 'tsEvent'. Struct version vtTimeTrace = 0:(1/100):10; tfSignal = repmat(linspace(0, 1, numel(vtTimeTrace)), [1 10 10]); ts = TimeSeriesStructure(vtTimeTrace, tfSignal); plot_ts(ts); % Plot the time series plot_ts(sumTS(ts, ts)); % Add two time series var_tTS(ts) % Compute the variance of 'ts' over time plot_ts(meanTS(ts)); % Plot the mean of all series within 'ts' plot_ts(ts, 5); % Plot only series '5' within 'ts' (linear indexing) tsEvent = TimeSeriesStructure(vtTimeTrace, [], 'event'); % Make an event time series tfTriggeredResponse = ContingentBinTS(tsEvent, ts, [], -1:.5:1); % Compute the triggered response of 'ts', at trigger events from 'tsEvent'. Getting help Look through the list of functions / methods to see if what you need already exists If there's a bug, try to debug the problem yourself as much as possible (using breakpoints, "dbstop error", stepping through the code, etc) Make an issue report: https://bitbucket.org/DylanMuir/timeseriesanalysis/issues