def plot_hist(data, dim, level):
# TODO(ipince): which bins to use?
#utils.hist_wrap((data, range(int(min(data)), int(max(data)))),
utils.hist_wrap((data, 200),
title='%s diff by %s' % (dim, level),
xlabel='%s_2013 - %s_2012' % (dim, dim),
ylabel='frequency',
filename='%s_diff_hist_by_%s.png' % (dim, level))
# save a zoomed-in copy
plt.ylim(0, 20)
plt.draw()
plt.savefig('%s_diff_hist_by_%s_zoom.png' % (dim, level))
plot_benford_2nd(states, 'state')
plot_benford_2nd(munis, 'muni')
plot_benford_2nd(parishes, 'parish')
plot_benford_2nd(centers, 'center')
plot_benford_2nd(tables, 'table')
plot_all_benford()
print "min for cap2013 is %d" % min([tables[1][code]['cap'] for code in tables[1]])
print "max for cap2013 is %d" % max([tables[1][code]['cap'] for code in tables[1]])
print "min for gov2013 is %d" % min([tables[1][code]['gov'] for code in tables[1]])
print "max for gov2013 is %d" % max([tables[1][code]['gov'] for code in tables[1]])
print "min for voters is %d" % min([tables[1][code]['voters'] for code in tables[1]])
print "max for voters is %d" % max([tables[1][code]['voters'] for code in tables[1]])
utils.hist_wrap(([tables[1][code]['voters'] for code in tables[1]], 200), filename='table-voters.png')
utils.hist_wrap(([tables[1][code]['cap'] for code in tables[1]], 200), filename='table-cap.png')
utils.hist_wrap(([tables[1][code]['gov'] for code in tables[1]], 200), filename='table-gov.png')
utils.hist_wrap(([centers[1][code]['voters'] for code in centers[1]], 200), filename='center-voters.png')
import matplotlib.pyplot as plt
plt.figure()
plt.hist([centers[1][code]['voters'] for code in centers[1]], 200, cumulative=True)
plt.show()
plt.savefig('center-cum_hist.png')
plt.figure()
plt.hist([tables[1][code]['voters'] for code in tables[1]], 200, cumulative=True)
plt.show()
plt.savefig('table-cum_hist.png')
utils.hist_wrap(([centers[1][code]['cap'] for code in centers[1]], 200), filename='center-cap.png')
utils.hist_wrap(([centers[1][code]['gov'] for code in centers[1]], 200), filename='center-gov.png')