120e438fcd6357fc81a979d92b39b2bd7101dd34,pymc/step_methods/metropolis.py,BinaryMetropolis,astep,#BinaryMetropolis#Any#Any#,152

Before Change

        // Calculate probabilities of each value
        logp_true = logp([True]*len(q0))
        logp_false = logp([False]*len(q0))

        p_true = exp(logp_true)
        p_false = exp(logp_false)

        // Stochastically set value according to relative
        // probabilities of True and False
        return (random(len(q0)) < p_true / (p_true + p_false)).astype(int)

After Change

        // Convert adaptive_scale_factor to a jump probability
        p_jump = 1.-.5**self.scaling

        rand_array = random(q0.shape)
        q = copy(q0)
        // Locations where switches occur, according to p_jump
        switch_locs = where(rand_array<p_jump)
        q[switch_locs] = True - q[switch_locs]

        q_new = metrop_select(logp(q) - logp(q0), q, q0)

        return q_new

def subst(a, val, index):
    ap = a.copy()
    ap[index]=val

Instances