diff options
Diffstat (limited to 'experiments/ml.pyx')
| -rw-r--r-- | experiments/ml.pyx | 46 |
1 files changed, 21 insertions, 25 deletions
diff --git a/experiments/ml.pyx b/experiments/ml.pyx index 853c934..67b561c 100644 --- a/experiments/ml.pyx +++ b/experiments/ml.pyx @@ -15,11 +15,12 @@ cdef DTYPE_t weight_success(int dist, int dt, DTYPE_t alpha, DTYPE_t delta, cdef DTYPE_t structural, temporal, result structural = delta ** dist # structural = plogis(w1,delta) * plogis(w2,delta) * plogis(w3,delta) - temporal = exp(-alpha*dt) * (exp(alpha)-1) + # temporal = exp(-alpha*dt) * (exp(alpha)-1) + temporal = 1 - exp(-alpha*dt) if exp(-alpha*dt)==0.: print 'UNDERFLOW ERROR' # temporal = 1. / (1. + (dt - 1.)/alpha)**0.01 - 1. / (1. + dt/alpha)**0.01 result = log(structural * temporal) - print 'st', structural, temporal + # print 'st', structural, temporal return result cdef DTYPE_t weight_failure(int dist, int dt, DTYPE_t alpha, DTYPE_t delta, @@ -31,17 +32,17 @@ cdef DTYPE_t weight_failure(int dist, int dt, DTYPE_t alpha, DTYPE_t delta, temporal = exp(-alpha * dt) # temporal = 1. - 1. / (1. + dt/alpha)**0.01 result = log(1. - structural + structural * temporal) - print 'stnv', structural, temporal + # print 'stnv', structural, temporal return result def ml(dict root_victims, dict victims, dict non_victims, DTYPE_t age, DTYPE_t alpha, DTYPE_t delta): cdef: int n_roots, n_victims, n_nodes, roots, i, dist, dt, t, l - DTYPE_t beta, ll, beta2 + DTYPE_t beta, ll list parents, failures, successes n_roots, n_victims = len(root_victims), len(victims) - n_nodes = 4#148152 + n_nodes = 11270 cdef: np.ndarray[DTYPE_t] probs = np.zeros(n_victims, dtype=DTYPE) np.ndarray[DTYPE_t] probs_fail = np.zeros(n_victims, dtype=DTYPE) @@ -61,12 +62,12 @@ def ml(dict root_victims, dict victims, dict non_victims, DTYPE_t age, probs[i] = max(s - failures[l] for l, s in enumerate(successes)) # loop through non-victims - for i, parents in enumerate(non_victims.itervalues()): - # for each non victim node, compute the probability that all its - # parents fail to infect it - failures = [weight_failure(dist, dt, alpha, delta, w1, w2, w3) - for (dist, dt, w1, w2, w3) in parents] - probs_nv[i] = sum(failures) + # for i, parents in enumerate(non_victims.itervalues()): + # # for each non victim node, compute the probability that all its + # # parents fail to infect it + # failures = [weight_failure(dist, dt, alpha, delta, w1, w2, w3) + # for (dist, dt, w1, w2, w3) in parents] + # probs_nv[i] = sum(failures) # calculate log likelihood # probs.sort(); probs = probs[::-1] # sort probs in descending order @@ -75,32 +76,27 @@ def ml(dict root_victims, dict victims, dict non_victims, DTYPE_t age, ll = probs_fail.sum() # add probability that all edges to victims fail ll += probs_nv.sum() # add probability that all edges to non_victims fail - print 'probs', probs - max_i = -1 + # print 'probs', probs max_beta_add = float('-inf') # iterate over all victim nodes to find the optimal threshold - for i in xrange(0, n_victims+1, 1): - print - roots = n_roots + n_victims - i - beta = 1. / (1. + exp(-probs[i])) - # beta = float(roots)/float(n_nodes) + for beta in np.arange(0.001, .1, .001): thresh = log(beta/(1.-beta)) - print 'thresh:', thresh + # print 'beta:', beta, 'thresh:', thresh, 'infected:', len(probs[probs>=thresh]) + roots = n_roots + len(probs[probs<thresh]) # add probability for realized edges and subtract probability these edges fail beta_add = (probs[probs>=thresh]).sum() - print 'len(probs[probs>=thresh]):', len(probs[probs>=thresh]) # add probability for the seeds and non-seeds beta_add += roots * log(beta) + (n_nodes-roots) * log(1 - beta) if beta_add > max_beta_add: - max_i = i + max_beta = beta + max_roots = roots max_beta_add = beta_add - print 'i:', max_i, 'add:', max_beta_add, 'roots:', roots - else: - print i + # print 'beta:', max_beta, 'add:', max_beta_add, 'roots:', max_roots ll += max_beta_add - roots = n_roots + n_victims - max_i + roots = max_roots + beta = max_beta # print n_nodes, n_roots, n_victims, max_i, roots return (beta, roots, ll) |