missing log-log scale in solution for Weibull transform (chap05soln.ipynb)

[rjoberon]

The solution for the exercise in Chapter 5 to create a transformation which shows a straight line for Weibull distributed data is missing the log-log transformation:

sample = [random.weibullvariate(2, 1) for _ in range(1000)]
cdf = thinkstats2.Cdf(sample)
thinkplot.Cdf(cdf, transform="weibull")
thinkplot.Config(xlabel="Weibull variate", ylabel="CCDF")

The last line should be

thinkplot.Config(xlabel="Weibull variate", ylabel="CCDF", xscale="log", yscale="log")

instead. Then we get a straight line also for other values of k then just k=1.

(This seems to be a bit more complex, since thinkstats.Cdf correctly sets xscale and yscale to "log" but the scale return value of thinkstats.Cdf is not used here. Another option would be to replace the last two lines with

scale = thinkplot.Cdf(cdf, transform="weibull")
thinkplot.Config(xlabel="Weibull variate", ylabel="CCDF", xscale=scale["xscale"], yscale=scale["yscale"])

but that's less transparent.)