def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
valid_inputs, valid_targets = load_valid()
test_inputs, test_targets = load_test()
# add your code here (it should be less than 10 lines)
self = NaiveBayesClassifier()
self.train(train_inputs, train_targets)
test_prediction = self.predict(test_inputs)
valid_accuracy = self.compute_accuracy(valid_inputs, valid_targets)
test_accuracy = self.compute_accuracy(test_inputs, test_targets)
print "valid accuracy: ", valid_accuracy, "\n test accuracy: ", test_accuracy
print((self.mean))
plot_digits(self.mean)
plt.savefig("2.4 mean.png")
plt.clf()
plot_digits(self.var)
plt.savefig("2.4 variance.png")
plt.clf()
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
nbc = NaiveBayesClassifier()
nbc.train(load_train()[0], load_train()[1])
print 'Training accuracy: %.4f Test accuracy: %.4f' % (
nbc.compute_accuracy(load_train()[0],
load_train()[1]),
nbc.compute_accuracy(load_test()[0],
load_test()[1]))
plot_digits(nbc.mean)
plot_digits(nbc.var)
'''
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
test_inputs, test_targets = load_test()
# add your code here (it should be less than 10 lines)
NB = NaiveBayesClassifier()
NB.train(train_inputs, train_targets)
print(NB.compute_accuracy(train_inputs, train_targets))
print(NB.compute_accuracy(test_inputs, test_targets))
plot_digits(NB.mean)
plot_digits(NB.var)
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
nbc = NaiveBayesClassifier()
nbc.train(load_train()[0],load_train()[1])
print 'Training accuracy: %.4f Test accuracy: %.4f' % (nbc.compute_accuracy(load_train()[0],load_train()[1]), nbc.compute_accuracy(load_test()[0],load_test()[1]))
plot_digits(nbc.mean)
plot_digits(nbc.var)
'''
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
test_inputs, test_targets = load_test()
nbc = NaiveBayesClassifier()
nbc.train(train_inputs, train_targets)
acc = nbc.compute_accuracy(test_inputs, test_targets)
print "the accuracy is {}".format(acc)
plot_digits(nbc.mean)
plot_digits(nbc.var)
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
test_inputs, test_targets = load_test()
my_model = NaiveBayesClassifier()
my_model.train(train_inputs, train_targets)
print "training accuracy: %.1f%%\ntest accuracy %.1f%%" % \
(100.0*my_model.compute_accuracy(train_inputs, train_targets), \
100.0*my_model.compute_accuracy(test_inputs, test_targets))
plot_digits( np.concatenate((my_model.mean, my_model.var), axis=0) )
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
test_inputs, test_targets = load_test()
# add your code here (it should be less than 10 lines)
classifier = NaiveBayesClassifier()
while (not classifier.model_learned):
classifier.train(train_inputs, train_targets)
print 'Training accuracy: ', classifier.compute_accuracy(train_inputs, train_targets)
print 'Test accuracy: ', classifier.compute_accuracy(test_inputs, test_targets)
plot_digits(classifier.mean)
def main():
"""
Learn a Naive Bayes classifier on the digit dataset, evaluate its
performance on training and test sets, then visualize the mean and variance
for each class.
"""
train_inputs, train_targets = load_train()
test_inputs, test_targets = load_test()
# add your code here (it should be less than 10 lines)
nbc = NaiveBayesClassifier()
nbc.train(train_inputs, train_targets)
print nbc.compute_accuracy(test_inputs, test_targets)
print nbc.compute_accuracy(train_inputs, train_targets)
plot_digits(nbc.mean.reshape((2,784)))
plot_digits(nbc.var.reshape((2,784)))
