def test(X, y, learned_params):
N = np.shape(X)[0] #no of instances
X = np.append(np.ones((N,1)), X,1) #appending a column of ones as bias (used in logistic regression weights prediction)
F = np.shape(X)[1] #no of features+1
p_old = 1
class_prob = []
for w in learned_params.keys():
p = Utils.logistic_transformation( learned_params[w], X )
class_prob.append(p_old-p)
p_old = p
class_prob.append(p_old)
max_prob = np.max(class_prob, 0)
predicted_y = []
output_label = range(min_class_label, max_class_label+1)
for i in xrange(np.size(max_prob)):
class_label = np.where(class_prob == max_prob[i])[0]
#print class_label
predicted_y.append(output_label[class_label[0]])
#print "predicted y :", predicted_y
#print "Actual y:", y
accuracy = Utils.calculate_accuracy(np.array(y), np.array(predicted_y))
f_score_mean, f_score_std = Utils.calculate_average_F1score(np.array(y), np.array(predicted_y), min_class_label, max_class_label)
return (accuracy, f_score_mean, f_score_std)
def Estep(x, w, a, b):
p = Utils.logistic_transformation(w, x)
log_p_a =np.log(p) + np.log(a)
log_p_ab = np.log(p*a + (1-p)*b)
log_ycap = log_p_a - log_p_ab
ycap = np.exp(log_ycap)
return ycap
def test(X, y, learned_params):
N = np.shape(X)[0] # no of instances
X = np.append(
np.ones((N, 1)), X, 1
) # appending a column of ones as bias (used in logistic regression weights prediction)
F = np.shape(X)[1] # no of features+1
p_old = 1
class_prob = []
for w in learned_params.keys():
p = Utils.logistic_transformation(learned_params[w], X)
class_prob.append(p_old - p)
p_old = p
class_prob.append(p_old)
max_prob = np.max(class_prob, 0)
cs = np.array(class_prob)
cs = np.reshape(cs, cs.size, order="F").reshape(np.shape(cs)[1], np.shape(cs)[0])
second_max = np.where(cs.argsort() == cs.shape[-1] - 1)[1] + 1
cs_prob = cs[np.where(cs.argsort() == cs.shape[-1] - 1)] #
predicted_y = []
arbitrary_answer = []
output_label = range(min_class_label, max_class_label + 1)
for i in xrange(np.size(max_prob)):
class_label = np.where(class_prob == max_prob[i])[0]
predicted_y.append(output_label[class_label[0]])
arbitrary_answer.append(
str(output_label[class_label[0]])
+ " prob "
+ str(max_prob[i])
+ " second prob "
+ str(cs_prob[i])
+ "second class"
+ str(second_max[i])
)
# print "predicted y :", predicted_y
# print "Actual y:", y
accuracy = Utils.calculate_accuracy(np.array(y), np.array(predicted_y))
f_score_mean, f_score_std = Utils.calculate_average_F1score(
np.array(y), np.array(predicted_y), min_class_label, max_class_label
)
return (accuracy, f_score_mean, f_score_std)
