print "Final results :\n"
pprint(EM_result)
#for e in xrange(len(no_of_experts)):
# Utils.visualize( EM_result[e], min_class_label, max_class_label, no_of_experts[e] )
return EM_result
def test(test_data):
print "Test data:"
pprint (test_data)
#k_fold_cross_validation()
train(data, 0)
Utils.showPlot()
"""else:
EM_perf = crowds_EM.predict_EM(crowds_EM.x, crowds_EM.y)
print "EM perf ", EM_perf
EM_acc += EM_perf
if EM_perf > EM_highest_performance['accuracy']:
EM_highest_performance['accuracy'] = EM_perf
EM_highest_performance['results'] = crowds_EM.results
MV_acc += crowds_EM.predict_MV(crowds_EM.x, crowds_EM.y)
#np.save('X.npy', crowds_EM.x)"""
"""print "No. of failed iterations : ", failed
print "Average EM accuracy after ", iterations," iter : ", EM_acc/(total_iter-failed)
