def experiment_anomaly_detection(train, test, comb, num_train, anom_prob, labels): phi = calc_feature_vecs(comb.X) print phi.size # bayes classifier (DIMS, N) = phi.size w_bayes = co.matrix(1.0, (DIMS, 1)) pred = w_bayes.trans()*phi[:,num_train:] (fpr, tpr, thres) = metric.roc_curve(labels[num_train:], pred.trans()) bayes_auc = metric.auc(fpr, tpr) # train one-class svm kern = Kernel.get_kernel(phi[:,0:num_train], phi[:,0:num_train]) ocsvm = OCSVM(kern, C=1.0/(num_train*anom_prob)) ocsvm.train_dual() kern = Kernel.get_kernel(phi, phi) (oc_as, foo) = ocsvm.apply_dual(kern[num_train:,ocsvm.get_support_dual()]) (fpr, tpr, thres) = metric.roc_curve(labels[num_train:], oc_as) base_auc = metric.auc(fpr, tpr) if (base_auc<0.5): base_auc = 1.0-base_auc # train structured anomaly detection #sad = StructuredOCSVM(train, C=1.0/(num_train*anom_prob)) sad = StructuredOCSVM(train, C=1.0/(num_train*0.5)) (lsol, lats, thres) = sad.train_dc(max_iter=50) (pred_vals, pred_lats) = sad.apply(test) (fpr, tpr, thres) = metric.roc_curve(labels[num_train:], pred_vals) auc = metric.auc(fpr, tpr) if (auc<0.5): auc = 1.0-auc return (auc, base_auc, bayes_auc)
def experiment_anomaly_detection(train, test, comb, num_train, anom_prob, labels): # train one-class svm phi = calc_feature_vecs(comb.X) kern = Kernel.get_kernel(phi[:,0:num_train], phi[:,0:num_train]) ocsvm = OCSVM(kern, C=1.0/(num_train*anom_prob)) ocsvm.train_dual() kern = Kernel.get_kernel(phi, phi) (oc_as, foo) = ocsvm.apply_dual(kern[num_train:,ocsvm.get_support_dual()]) (fpr, tpr, thres) = metric.roc_curve(labels[num_train:], oc_as) base_auc = metric.auc(fpr, tpr) # train structured anomaly detection sad = StructuredOCSVM(train, C=1.0/(num_train*anom_prob)) (lsol, lats, thres) = sad.train_dc(max_iter=40) (pred_vals, pred_lats) = sad.apply(test) (fpr, tpr, thres) = metric.roc_curve(labels[num_train:], pred_vals) auc = metric.auc(fpr, tpr) return (auc, base_auc)
def experiment_anomaly_detection(train, test, comb, num_train, anom_prob,
labels):
# train one-class svm
phi = calc_feature_vecs(comb.X)
kern = Kernel.get_kernel(phi[:, 0:num_train], phi[:, 0:num_train])
ocsvm = OCSVM(kern, C=1.0 / (num_train * anom_prob))
ocsvm.train_dual()
kern = Kernel.get_kernel(phi, phi)
(oc_as, foo) = ocsvm.apply_dual(kern[num_train:, ocsvm.get_support_dual()])
(fpr, tpr, thres) = metric.roc_curve(labels[num_train:], oc_as)
base_auc = metric.auc(fpr, tpr)
# train structured anomaly detection
sad = StructuredOCSVM(train, C=1.0 / (num_train * anom_prob))
(lsol, lats, thres) = sad.train_dc(max_iter=40)
(pred_vals, pred_lats) = sad.apply(test)
(fpr, tpr, thres) = metric.roc_curve(labels[num_train:], pred_vals)
auc = metric.auc(fpr, tpr)
return (auc, base_auc)
fig = plt.figure()
for i in xrange(4):
plt.subplot(2,4,i+1)
if i==0:
plt.title("LatentSVDD")
lsvdd.train_dc()
(scores,lats) = lsvdd.apply(predsobj)
if i==1:
plt.title("StructPCA")
spca.train_dc()
(scores,lats) = spca.apply(predsobj)
if i==2:
plt.title("StructOCSVM")
socsvm.train_dc()
(scores,lats) = socsvm.apply(predsobj)
if i==3:
plt.title("SSVM")
ssvm.train()
(scores,lats) = ssvm.apply(predsobj)
# plot scores
Z = np.reshape(scores,(sx,sy))
plt.contourf(X, Y, Z)
plt.scatter(Dtrain[0,:],Dtrain[1,:],10)
# plot latent variable
Z = np.reshape(lats,(sx,sy))
plt.subplot(2,4,i+4+1)
plt.contourf(X, Y, Z)
plt.scatter(Dtrain[0,:],Dtrain[1,:],10)
fig = plt.figure()
for i in xrange(4):
plt.subplot(2, 4, i + 1)
if i == 0:
plt.title("LatentSVDD")
lsvdd.train_dc()
(scores, lats) = lsvdd.apply(predsobj)
if i == 1:
plt.title("StructPCA")
spca.train_dc()
(scores, lats) = spca.apply(predsobj)
if i == 2:
plt.title("StructOCSVM")
socsvm.train_dc()
(scores, lats) = socsvm.apply(predsobj)
if i == 3:
plt.title("SSVM")
ssvm.train()
(scores, lats) = ssvm.apply(predsobj)
# plot scores
Z = np.reshape(scores, (sx, sy))
plt.contourf(X, Y, Z)
plt.scatter(Dtrain[0, :], Dtrain[1, :], 10)
# plot latent variable
Z = np.reshape(lats, (sx, sy))
plt.subplot(2, 4, i + 4 + 1)
plt.contourf(X, Y, Z)
plt.scatter(Dtrain[0, :], Dtrain[1, :], 10)
