exponential fit

exponential distribution
y = 1/β e^ -y/β

Maximum likelihood estimation
d1, d2, …, dn
Pr{d1}*Pr{d2}…Pr{dw}

X = time until next tweet
f{x = t} = 1/β e * -t/β

from scipy.optimize import curve_fit
def fitFunc(t, b):
	return b*numpy.exp(-b*t)
count,division = np.histogram(timeUntilNext, bins=100, normed=Ture)
fitParams, fitCov = curve_fit(fitFunc, division[0:len(division)-1], count)