Re: [Matplotlib-users] semilog error plots

[John H. <jdh...@ac...>]

    from matplotlib.matlab import *
    m = array([0.1, 0.2, 0.3, 0.6, 0.8, 1,  3])
    r = array([18, 180, 1800, 1.8, 0.18, 250, 2.8])
    plot(m, r)
    set(gca(), "yscale", "log")
    show()

This is a known issue I've been meaning to work on.  Here's what is
happening.  On the plot command the tick locator is autoranging the
axes and ymin becomes 0.  Since the log transformation applies to the
tick locations as well as the data, it is log transforming the tick
location of zero.  To fix this, I need to check the axes range when
the call to set yscale is made and fix it.  In the meantime, you can
explicity set the ylim after the call to plot and before the call to
yscale.

    from matplotlib.matlab import *
    m = array([0.1, 0.2, 0.3, 0.6, 0.8, 1,  3])
    r = array([18, 180, 1800, 1.8, 0.18, 250, 2.8])
    plot(m, r)
    set(gca(), 'ylim', (0.1, 2000))
    set(gca(), "yscale", "log")
    show()


JDH