search

msmbuilder / msmexplorer

Data visualizations for biomolecular dynamics
http://msmbuilder.org/msmexplorer/
MIT License
17 stars 17 forks source link

Implied timescales plot #90

Closed jeiros closed 7 years ago

jeiros commented 7 years ago

Proposed Feature / Bug-Fix

Adding an implied timescales plot, as I suggested in #89

Example Figure

Plot a list of msm objects like this:

plot_implied_timescales(msm_objs, color_palette=colors,
                        xlabel='Lag time (frames)',
                        ylabel='Implied Timescales ($ns$)')

its-ex

PR Check-list

mpharrigan commented 7 years ago

I would lobby for a log-log plot but looks good!

jeiros commented 7 years ago

Do you mean having a log x axis? I could do that, but it's up to the user? Since it will depend on the lag times that he looks at.

mpharrigan commented 7 years ago

Do you mean having a log x axis? I could do that, but it's up to the user? Since it will depend on the lag times that he looks at.

Yeah, your example and test both have loggy x values, and I think this is the appropriate way to approach the problem. We're generally interested in order of magnitude lag-time. 1ns vs 2ns is inconsequential but you want to compare to e.g. 200 ns

cxhernandez commented 7 years ago

Thanks for this @jeiros! I think maybe allowing users to set the axes scaling as an option might be nice, but maybe have the default be log-log.