How to plot specific rows and columns of pandas dataframe (based on name of row and name of column) in bar plot with error bars?

Question

I have data in a csv file structured like so:

    Subject    group    Result1    Result2...    ResultN
    101        a        .5         .1            .2
    103        b        .1         .2            .5
    104        b        .2         .3            .4
    mean_a     a        .5         .1            .2
    mean_b     b        .1         .6            .4
    ste_a      a        .05        .02           .03
    ste_b      b        .01        .05           .04

I just want to end up with a bar plot, grouped by the Result, of the mean rows' values for each group, with the stes as the error bars. Unfortunately, I'm having trouble doing so. I can convert the dataframe to two separate dataframes, one for means and one for stes, like this:

               a        b
    Result1    .5       .1
    Result2    .1       .6

However, I cannot figure out how to plot the second dataframe of stes as error bars, and my method seems overly complicated, so I was wondering if anyone knows of a simpler way to do this and, if not, how to use the one dataframe to plot error bars for the other dataframe.

Answer 1

It is quite easy, just pass your error data to yerr argument the same as you will do in matplotlib.

DF=pd.DataFrame({'a':[.5,.1],'b':[.1,.6]})
DF.index=['Result1','Result2']
DF.plot(kind='bar',yerr=DF.b)