def main():
nb = NaiveBayes()
nb.evaluate()
vocab_size = None
max_length = None
embed_matrix = None
else:
train_corpus, test_corpus, vocab_size, max_length = Word2Vec(
X_train, X_test, y_train, y_test)
#print("Before Over Sampling: ",Counter(y_train))
#sm = SMOTE(n_jobs=cores)
#train_corpus, y_train = sm.fit_resample(train_corpus, y_train)
#print("After Over Sampling: ", Counter(y_train))
for i, v in enumerate(models):
if v == 'NaiveBayes':
model = NaiveBayes(train_corpus, test_corpus, y_train)
result = model.evaluate()
elif v == 'LogisticRegression':
model = LogisticRegress(train_corpus, test_corpus, y_train)
result = model.evaluate()
elif v == 'SVM':
model = SVM(train_corpus, test_corpus, y_train)
result = model.evaluate()
elif v == 'FeedForward':
y_train_one = np.array(y_train)
y_test_one = np.array(y_test)
y_train_one -= y_train_one.min()
y_test_one -= y_test_one.min()
model = FeedForward(train_corpus, test_corpus,
to_categorical(y_train_one),
to_categorical(y_test_one))
'Cluster 2 - Malignant', 'Cluster 3 - Malignant'
])
plt.show()
# ------------------------------------------ Question 2 ------------------------------------------
tags = BreastCancerData['y']
nb = NaiveBayes()
# train
nb_learn_start = timeit.default_timer()
nb.train(data_for_train, tags_for_train)
nb_learn_end = timeit.default_timer()
# evaluate
nb_train_error = nb.evaluate(data_for_train, tags_for_train)
nb_clf_start = timeit.default_timer()
nb_test_error = nb.evaluate(data_for_test, tags_for_test)
nb_clf_end = timeit.default_timer()
# Print the results
print('\n ---------------------------------------------------- \n')
print('Results for - Naive Bayes \n')
print('Train Set Error = ' + str(nb_train_error) + '.\n')
print('Test Set Error = ' + str(nb_test_error) + '.\n')
print('Train Time = ' + str(nb_learn_end - nb_learn_start) + '.\n')
print('Classification Time = ' + str(nb_clf_end - nb_clf_start) + '.\n')
print('\n ---------------------------------------------------- \n')
# ------------------------------------------ Question 3 ------------------------------------------
print('Logistic Regression: \n')
