die_rolled error

[jbwhit]

I think that I've found a bug in the logic behind this bit:

die_rolled(9, hypotheses)
Out[323]:
{6: 0.0,
 8: 0.0,
 12: 0.72553987580240009,
 20: 0.15671661317331853,
 22: 0.11774351102428135}

You need to reset the hypotheses, because they are being overwritten each time you roll the die, so the correct answer should be:

hypotheses = {
    6: 1,
    8: 1,
    12:1,
    20:1,
    22:1
}

# Normalize
total_possibilities = sum(hypotheses.values())
for key, value in hypotheses.items():
    hypotheses[key] = hypotheses[key]/(1.0*total_possibilities)

die_rolled(9, hypotheses)

{6: 0.0,
 8: 0.0,
 12: 0.4661016949152542,
 20: 0.2796610169491526,
 22: 0.25423728813559326}

Unless, I suppose, the idea is that each roll is an additional number reported so that by roll 9, you've also seen 5 and 6 as well.