This fork is out of date since Jon gave me permission to write to his repo. I've made all my updates to that repo instead. Nothing to see here, move along.
Contributors: jdherman and matthewjwoodruff
See example_images/
for the outputs from these files (PNG). Note that publication-quality figures will usually require a vector graphics format (SVG/EPS/PDF) along with some manual editing to improve font clarity, etc. The examples shown here are in unedited raster format and are only meant to be illustrative.
From example_data/column_data.txt
. Used with:
line_plot.{m,py}
: Standard line plot with different formatting options
fill_between.py
: Same data asline_plot
, with shaded fill between Series 1 and Series 2
stacked_area.m
: Stacked area plot with color options
animated_gif.{m,py}
: Save loop of images as a.gif
animation. The Python version requires you to useimagemagick
after the fact to stick the frames together:convert -delay 15 -loop 0 frame*png animation.gif
.
histogram_cdf.m
: Standard plots of frequency distributions
parallel_coordinate.m
: Plots row entries as lines instead of columns. Options to scale all values between[0,1]
and toggle solutions on/off above/below a certain threshold. (These functions are included in/util/
).
From example_data/gridded_2D.txt
. Used with:
imagesc_plot.m
: Show matrix values directly as color blocks
contour_surf.{m,py}
: Contour, filled contour, and surface plots of meshgrid data
From example_data/non_gridded_2D.txt
. The data file contains rows of (x,y,z)
points that are not on a regular grid. Used with:
scatter_plot.{m,py}
: Plot the irregularly spaced points as dots, withZ
values as the color
contourf_non_gridded.{m,py}
: Use Matlab's built-ingriddata
function to interpolate the data onto a standard meshgrid, which can then be used for standard contour/mesh/surf plots, etc.