def get_classifier(vocabulary):
'''
需要将抽象的句子分类到某一个模板，这里是训练分类器
'''
# 准备数据集
x_train = []
y_train = []
root = "./Qdata/question/"
filenames = [
filename for filename in os.listdir(root) if filename[0] == "【"
]
for filename in filenames:
label = int(filename[filename.index("【") + 1:filename.index("】")])
with open(root + filename, "r", encoding="utf-8") as f:
sen_list = [line.strip() for line in f.readlines()]
x_train += sen_list
y_train += [label] * len(sen_list)
x_train_array = np.zeros((len(x_train), len(vocabulary)))
for row, sentence in enumerate(x_train):
for col, voc in enumerate(vocabulary):
if voc in sentence:
x_train_array[row, col] = 1
classifier = ComplementNB()
classifier.fit(x_train_array, y_train)
return classifier
def realizar_treinamento(registros_de_treino, vetorizador):
treino_comentarios = [
registro_treino[0] for registro_treino in registros_de_treino
]
treino_respostas = [
registro_treino[1] for registro_treino in registros_de_treino
]
treino_comentarios = vetorizador.fit_transform(treino_comentarios)
# modelo = BernoulliNB()
# modelo = MultinomialNB()
modelo = ComplementNB()
modelo.fit(treino_comentarios, treino_respostas)
# VALIDAÇÃO COM CROSS VALIDATION
# cv = KFold(n_splits=200)
# resultado = cross_val_predict(modelo, treino_comentarios, treino_respostas, cv=cv)
# total = len(resultado)
# acc = 0
#
# score = accuracy_score(treino_respostas, resultado)
# print(score * 100)
#
# for i in range(0, total):
# if resultado[i] == treino_respostas[i]:
# acc += 1
#
# print(acc, total, acc/total * 100)
#
# print(metrics.classification_report(treino_respostas, resultado, [0, 1]))
#
# exit()
return modelo
class CNBTwoStepClassifier(ImbalancedTrainerInterface):
def __init__(self, alpha=1):
self.alpha = alpha
self.clf_yn = ComplementNB(alpha=alpha)
self.clf_n = ComplementNB(alpha=alpha)
def fit(self, X_train, y_train):
X_train = X_train.toarray().tolist()
y_train = pd.Series(y_train)
max_class = self._find_dominant_class(X_train, y_train)
x_w, x_o, y_w, y_o = self._partition(X_train, y_train, max_class[0])
x_yn = x_w + x_o
y_yn = y_w + ['not'] * len(y_o)
# print(y_yn)
# print(y_o)
self.clf_yn.fit(x_yn, y_yn)
self.clf_n.fit(x_o, y_o)
def predict(self, X_test):
y_pred_yn = self.clf_yn.predict(X_test)
y_pred_total = []
for p,x in list(zip(y_pred_yn, X_test)):
if p == 'not':
y_pred_total.append(self.clf_n.predict(x)[0])
else:
y_pred_total.append(p)
return y_pred_total
def score(self, X_test, y_test):
y_test = pd.Series(y_test)
y_pred = self.predict(X_test)
acc = np.mean(y_pred == y_test)
return acc
def confusion_matrix():
'''
Creates a full confusion matrix for the top 15 varieties and displays it.
Currently changes to vectorizer and model must be done manually.
'''
wrangler = Data_Handler('data/cleaned_data.csv')
df = wrangler.get_top_num(15)
stops = wrangler.stop_words
X = df['description']
y = df['variety']
X_train, X_test, y_train, y_test = train_test_split(X, y)
vecto = TfidfVectorizer(stop_words=stops)
X_train = vecto.fit_transform(X_train)
X_test = vecto.transform(X_test)
model = ComplementNB()
model.fit(X_train, y_train)
class_sort = [
'Pinot Noir', 'Cabernet Sauvignon', 'Red Blend',
'Bordeaux-style Red Blend', 'Syrah', 'Merlot', 'Zinfandel',
'Sangiovese', 'Malbec', 'Nebbiolo', 'Rosé', 'Chardonnay',
'Sauvignon Blanc', 'Riesling', 'White Blend'
]
plot_confusion_matrix(model,
X_test,
y_test,
normalize='true',
xticks_rotation='vertical',
labels=class_sort,
include_values=False)
plt.show()
def get_optimal_values_ComplementNB(x_train, y_train, x_val, y_val):
alphas = [x / 10 for x in range(0, 11)]
fit_priors = [True, False]
norms = [True, False]
max_score = 0
optimal_fit_prior = True
optimal_alpha = 1.0
optiomal_norm = False
# Evaluamos para escoger el mejor parámetro
for alpha in alphas:
for fit_prior in fit_priors:
for norm in norms:
naive = ComplementNB(alpha=alpha,
fit_prior=fit_prior,
norm=norm)
naive.fit(x_train, y_train)
y_pred = naive.predict(x_val)
if max_score < accuracy_score(y_val, y_pred) * 100:
optimal_alpha = alpha
optimal_fit_prior = fit_prior
optiomal_norm = norm
max_score = accuracy_score(y_val, y_pred) * 100
print(max_score, optimal_alpha, optimal_fit_prior, optiomal_norm)
return max_score, optimal_alpha, optimal_fit_prior, optiomal_norm
def test_cnb():
# Tests ComplementNB when alpha=1.0 for the toy example in Manning,
# Raghavan, and Schuetze's "Introduction to Information Retrieval" book:
# http://nlp.stanford.edu/IR-book/html/htmledition/the-bernoulli-model-1.html
# Training data points are:
# Chinese Beijing Chinese (class: China)
# Chinese Chinese Shanghai (class: China)
# Chinese Macao (class: China)
# Tokyo Japan Chinese (class: Japan)
# Features are Beijing, Chinese, Japan, Macao, Shanghai, and Tokyo.
X = np.array([[1, 1, 0, 0, 0, 0],
[0, 1, 0, 0, 1, 0],
[0, 1, 0, 1, 0, 0],
[0, 1, 1, 0, 0, 1]])
# Classes are China (0), Japan (1).
Y = np.array([0, 0, 0, 1])
# Verify inputs are nonnegative.
clf = ComplementNB(alpha=1.0)
assert_raises(ValueError, clf.fit, -X, Y)
clf.fit(X, Y)
# Check that counts are correct.
feature_count = np.array([[1, 3, 0, 1, 1, 0], [0, 1, 1, 0, 0, 1]])
assert_array_equal(clf.feature_count_, feature_count)
class_count = np.array([3, 1])
assert_array_equal(clf.class_count_, class_count)
feature_all = np.array([1, 4, 1, 1, 1, 1])
assert_array_equal(clf.feature_all_, feature_all)
# Check that weights are correct. See steps 4-6 in Table 4 of
# Rennie et al. (2003).
theta = np.array([
[
(0 + 1) / (3 + 6),
(1 + 1) / (3 + 6),
(1 + 1) / (3 + 6),
(0 + 1) / (3 + 6),
(0 + 1) / (3 + 6),
(1 + 1) / (3 + 6)
],
[
(1 + 1) / (6 + 6),
(3 + 1) / (6 + 6),
(0 + 1) / (6 + 6),
(1 + 1) / (6 + 6),
(1 + 1) / (6 + 6),
(0 + 1) / (6 + 6)
]])
weights = np.zeros(theta.shape)
for i in range(2):
weights[i] = np.log(theta[i])
weights[i] /= weights[i].sum()
assert_array_equal(clf.feature_log_prob_, weights)
def _complementnb(*,
train,
test,
x_predict=None,
metrics,
alpha=1.0,
fit_prior=True,
class_prior=None,
norm=False):
"""For for info visit :
https://scikit-learn.org/stable/modules/generated/sklearn.naive_bayes.ComplementNB.html#sklearn.naive_bayes.ComplementNB
"""
model = ComplementNB(alpha=alpha,
fit_prior=fit_prior,
class_prior=class_prior,
norm=norm)
model.fit(train[0], train[1])
model_name = 'ComplementNB'
y_hat = model.predict(test[0])
if metrics == 'f1_score':
accuracy = f1_score(test[1], y_hat)
if metrics == 'jaccard_score':
accuracy = jaccard_score(test[1], y_hat)
if metrics == 'accuracy_score':
accuracy = accuracy_score(test[1], y_hat)
if x_predict is None:
return (model_name, accuracy, None)
y_predict = model.predict(x_predict)
return (model_name, accuracy, y_predict)
def findBestFitCluster(orphanCorpus, corpusCluster={}):
"""
Given a set of questions without a cluster and a set of other clusters, find the best cluster to put the orphaned questions
Parameters:
orphanCorpus (tagged_question_corpus.TaggedQuestionCorpus): corpus of the questions without a cluster.
corpusCluster ({tagged_question_corpus.TaggedQuestionCorpus}): Object containing different clusters and their corpuses
Returns:
xxx
"""
# corpusCluster = {
# "questions": [ 'and the moon too guys', 'lets show some or a lot of love for the moon!!' ],
# "question_vectors": [[], []],
# "clusterIds": [ '4', '4' ]
# }
# orphanCorpus = [ {
# "id": 11, "question": 'Another one about the sun?', "question_vector": []
# },
# {
# "id": 33,
# "question": 'What is the distance from the sun though?', "question_vector": [] },
# {
# "id": 37,
# "question": 'what\'s the changing factors of the sun and moon together?', "question_vector": []
# } ]
# Fit the Naive bayes model on existing clusters
clf = ComplementNB()
clf.fit(corpusCluster["question_vectors"], corpusCluster["clusterIds"])
predictions = clf.predict_proba(
[doc["question_vector"] for doc in orphanCorpus])
def ComplementNB_classification(train,
test,
train_labels,
test_labels,
res={}):
"""
:param train: training data, iterable/list
:param test: testing data, iterable/list
:param train_labels: training labels, iterable/list
:param test_labels: testing labels, iterable/list
:return: / --> Saves data in folder "Results"
"""
print("Classifying with Complement Nive Bayes...")
complNB = ComplementNB()
complNB.fit(train, train_labels)
prediction = complNB.predict(test)
utils.report_and_confmat(test_labels, prediction, "ComplementNB")
score = complNB.score(test, test_labels)
res["ComplementNB"] = {
"model": complNB,
"accuracy": score,
"name": "ComplementNB"
}
print("Complement ended...")
return score, complNB
class ComplementNBImpl():
def __init__(self,
alpha=1.0,
fit_prior=True,
class_prior=None,
norm=False):
self._hyperparams = {
'alpha': alpha,
'fit_prior': fit_prior,
'class_prior': class_prior,
'norm': norm
}
def fit(self, X, y=None):
self._sklearn_model = SKLModel(**self._hyperparams)
if (y is not None):
self._sklearn_model.fit(X, y)
else:
self._sklearn_model.fit(X)
return self
def predict(self, X):
return self._sklearn_model.predict(X)
def predict_proba(self, X):
return self._sklearn_model.predict_proba(X)
def main():
# Iris or breast cancer dataset can be used too
x, y = datasets.load_wine(return_X_y=True)
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=2405)
# Multinomial Naive Bayes
MNB = MultinomialNB()
MNB.fit(x_train, y_train)
mnb_accuracy = MNB.score(x_test, y_test)
print(f"MultinomialNB accuracy is {round(mnb_accuracy, 4)}")
# Gaussian Naive Bayes
GNB = GaussianNB()
GNB.fit(x_train, y_train)
gnb_accuracy = GNB.score(x_test, y_test)
print(f"GaussianNB accuracy is {round(gnb_accuracy, 4)}")
# Complement Naive Bayes
CNB = ComplementNB()
CNB.fit(x_train, y_train)
cnb_accuracy = CNB.score(x_test, y_test)
print(f"ComplementNB accuracy is {round(cnb_accuracy, 4)}")
class bayes(object):
def __init__(self, data, target, algorithm="GNB"):
self.algorithm = algorithm
self.data = data
self.target = target
if algorithm == 'GNB':
self.model = GaussianNB()
elif algorithm == 'MNB':
self.model = MultinomialNB()
elif algorithm == 'BNB':
self.model = BernoulliNB()
else:
self.model = ComplementNB()
self.model.fit(data, target)
def save_model(self, path):
_joblib.dump(self.model, path)
def load_model(self, path):
self.model = _joblib.load(path)
def predict(self, x):
res = self.model.predict(x)
return res
def pickling():
'''
Creates and pickles both the vectorizer and model for use in prediction.
Parameters
----------
None
Returns
----------
None
'''
wrangler = Data_Handler('data/cleaned_data.csv')
stops = wrangler.stop_words
df = wrangler.get_top_num(15)
X = df['description']
y = df['variety']
vecto = TfidfVectorizer(stop_words=stops)
X = vecto.fit_transform(df['description'])
f = open('pickles/text_vec.pkl', 'wb')
pickle.dump(vecto, f)
model = ComplementNB()
model.fit(X, y)
m = open('pickles/model.pkl', 'wb')
pickle.dump(model, m)
def main(args):
model_name = args.model_name
model_dir = os.path.join(args.root, "model") # get model dir
data_dir = os.path.join(args.root, "data") # get data dir
data_path = os.path.join(data_dir, args.inFile)
print('load data from' + data_path)
data = pickle.load(open(data_path, 'rb'))
out_path = os.path.join(data_dir, args.outFileName + '.csv')
assert 'data' in data
if args.train:
ratio = args.ratio
clf = ComplementNB(alpha=args.alpha,
fit_prior=args.fit_prior,
norm=args.norm)
assert 'target' in data
features = data['data']
labels = data['target']
rs = ShuffleSplit(n_splits=1, test_size=ratio)
train_index, val_index = next(rs.split(features, labels))
x_train = features[train_index]
x_test = features[val_index]
y_train = labels[train_index]
y_test = labels[val_index]
clf.fit(x_train, y_train)
y_pred = clf.predict(x_test)
# The accuracy
print('Accuracy: \n', accuracy_score(y_test, y_pred))
df = pd.DataFrame({
'pred': y_pred,
'target': y_test,
})
print(f'validation results save to:{args.outFileName}.csv')
df.to_csv(out_path)
print("Some results of validation:")
print(df.head())
model_path = os.path.join(model_dir, f'{model_name}_{model}.model')
dump(clf, model_path)
else:
# TODO: How to Save the prediction?
model_path = os.path.join(model_dir, args.model_path)
clf = load(args.model)
x = data['data']
pred = clf.predict(x)
df = pd.DataFrame({
'pred': pred,
})
df.to_csv(out_path)
class NaiveBayes():
def __init__(self, division="sents", ngram=1):
self.df_train = pd.read_csv(f"../data/{division}_train.csv",
sep='\t',
names=['sentence', 'author', 'work'])
self.df_val = pd.read_csv(f"../data/{division}_val.csv",
sep='\t',
names=['sentence', 'author', 'work'])
self.df_test = pd.read_csv(f"../data/{division}_test.csv",
sep='\t',
names=['sentence', 'author', 'work'])
self.df_spurious = pd.read_csv(f"../data/{division}_spurious.csv",
sep='\t',
names=['sentence', 'work'])
self.df_epistles = self.df_spurious[self.df_spurious['work'] == 36]
self.df_spurious = self.df_spurious[self.df_spurious['work'] != 36]
self.tfidf = TfidfVectorizer(lowercase=False,
stop_words=list(
map(strip_accents, STOPS_LIST)),
ngram_range=(1, ngram))
self.tfidf_train = self.tfidf.fit_transform(self.df_train['sentence'])
self.tfidf_val = self.tfidf.transform(self.df_val['sentence'])
self.tfidf_test = self.tfidf.transform(self.df_test['sentence'])
self.tfidf_spurious = self.tfidf.transform(
self.df_spurious['sentence'])
self.tfidf_epistles = self.tfidf.transform(
self.df_epistles['sentence'])
self.label = LabelEncoder()
self.author_train = self.label.fit_transform(self.df_train['author'])
self.author_val = self.label.transform(self.df_val['author'])
self.author_test = self.label.transform(self.df_test['author'])
self.nb = ComplementNB()
self.nb.fit(self.tfidf_train, self.author_train)
def eval(self):
author_train_pred = self.nb.predict(self.tfidf_train)
author_val_pred = self.nb.predict(self.tfidf_val)
author_test_pred = self.nb.predict(self.tfidf_test)
print(classification_report(self.author_train, author_train_pred))
print(classification_report(self.author_val, author_val_pred))
def predict(self):
epistles_labels = self.label.inverse_transform(
self.nb.predict(self.tfidf_epistles))
print((epistles_labels == "Plato").mean())
print(epistles_labels)
spurious_labels = self.label.inverse_transform(
self.nb.predict(self.tfidf_spurious))
print((spurious_labels == "Plato").mean())
def run_compnb(x_train, x_test, y_train, y_test, x):
'''Complement Naive Bayes'''
logger.info("Running ComplementNB")
compnb = ComplementNB()
compnb.fit(x_train, y_train)
compnb_pred = compnb.predict(x_test)
model_dict['compnb'] = get_model_results(compnb, x_test, y_test,
compnb_pred, x)
return compnb_pred
class DocClfTfidfCNB():
def __init__(self,maxStringLength=MAXSTRINGLENGH, \
firstStringLength=FIRSTSTRINGLENGTH):
self.maxStringLength=maxStringLength
self.firstStringLength=firstStringLength
self.message="Complement Naive Bayes using TF-IDF with "+"%5d" % maxFeatures + " features " + \
" ngram-range "+"%2d" % ngramrange[0]+" to "+"%2d" % ngramrange[1] + \
" maxString Length "+ "%6d" % self.maxStringLength
return
def preprocess(self,x):
xprocessed=[]
xbegin=[]
for item in x:
xprocessed.append(item[0:self.maxStringLength])
xbegin.append(item[0:self.firstStringLength])
return xprocessed,xbegin
def fit(self,x,y):
# generate dictionary of words and numb of word occurences
# in each document
xprocessed,xbegin=self.preprocess(x)
self.vectorizer=\
TfidfVectorizer(max_df=maxdf,min_df=mindf,max_features=maxFeatures,
ngram_range=ngramrange)
xv=self.vectorizer.fit_transform(xprocessed)
self.nbclf=ComplementNB(alpha=alphasmooth)
self.nbclf.fit(xv,y)
ytrain=self.nbclf.predict(xv)
return ytrain
#predict for a group of x value
def predict(self,x):
if (len(x[0])<minLength):
y=["No input"]
return y
try:
xprocessed,xbegin=self.preprocess(x)
xv=self.vectorizer.transform(xprocessed)
y=self.nbclf.predict(xv)
except:
raise
return y
# Compute confidence given predicted values & return confusion matrix
def confidence(self,ytest,ytestpred):
conf_mat = confusion_matrix(ytest, ytestpred)
# compute accuracy given predicted value
labels = sorted(set(ytest))
self.confidence=dict(zip(labels, conf_mat.diagonal()/
(.1+conf_mat.sum(axis=0))))
return conf_mat
# get the Confidence score for a single item:
def getConfidence(self,x,y):
try:
return self.confidence[y]
except:
return -1.0;
def train_complement_naivebayes(params,
x_train,
y_train,
n_folds,
random_state,
stratified=True,
shuffle=True):
# Model and hyperparameter selection
if stratified:
kf = StratifiedKFold(n_splits=n_folds,
random_state=random_state,
shuffle=shuffle)
else:
kf = KFold(n_splits=n_folds,
random_state=random_state,
shuffle=shuffle)
cnb_model = ComplementNB(**params)
i = 0
# Model Training
for (train_index, test_index) in kf.split(x_train, y_train):
# cross-validation randomly splits train data into train and validation data
print('\n Fold %d' % (i + 1))
x_train_cv, x_val_cv = x_train.iloc[train_index], x_train.iloc[
test_index]
y_train_cv, y_val_cv = y_train.iloc[train_index], y_train.iloc[
test_index]
# declare your model
cnb_model.fit(x_train_cv, y_train_cv)
# predict train and validation set accuracy and get eval metrics
scores_cv = cnb_model.predict(x_train_cv)
scores_val = cnb_model.predict(x_val_cv)
# training evaluation
train_pc = accuracy_score(y_train_cv, scores_cv)
train_pp = precision_score(y_train_cv, scores_cv)
train_re = recall_score(y_train_cv, scores_cv)
print('\n train-Accuracy: %.6f' % train_pc)
print(' train-Precision: %.6f' % train_pp)
print(' train-Recall: %.6f' % train_re)
eval_pc = accuracy_score(y_val_cv, scores_val)
eval_pp = precision_score(y_val_cv, scores_val)
eval_re = recall_score(y_val_cv, scores_val)
print('\n eval-Accuracy: %.6f' % eval_pc)
print(' eval-Precision: %.6f' % eval_pp)
print(' eval-Recall: %.6f' % eval_re)
i = i + 1
# return model for evaluation and prediction
return cnb_model
class Recommender(object):
'''
A class to house the text vectorizer and stacked Naive Bayes/Random Forest
Classifiers that form the heart of this wine recommender.
'''
def __init__(self):
self.nb = ComplementNB()
self.rf = RandomForestClassifier()
self.vecto = TfidfVectorizer()
def _fit(self, data):
'''
Takes in the data for the recommender to be trained and fit to.
Parameters
----------
data - The filepath to the data being fit.
Returns
----------
None
'''
wrangler = Data_Handler(data)
df = wrangler.get_top_num(15)
X = df['description']
y = df['variety']
X = self.vecto.fit_transform(X)
self.nb.fit(X, y)
X = self.nb.predict_proba(X)
self.rf.fit(X, y)
def predict(self, text):
'''
Takes in a single input of tasting notes and runs it through our
vectorizer and ensemble method to return the top five predicted
varieties.
Parameters
----------
text - str - The input tastings notes.
Returns
----------
top_five - lst - The top five predicted varieties for recommendation.
'''
vect = self.vecto.transform([text])
probs = self.nb.predict_proba(vect)
probs = self.rf.predict_proba(probs)[0]
idx = np.argsort(probs)
top_five_idx = idx[-1:-6:-1]
top_five = self.rf.classes_[top_five_idx]
return top_five
def initFitNaiveBayes(xtrain, ytrain):
nb = ComplementNB(
alpha=1.0,
class_prior=None,
fit_prior=True,
norm=False
)
nb.fit(xtrain, ytrain)
print("Naive Bayes Training: Done")
return nb
def CNB(train_x, train_y, test_x, test_y): #ComplementNB알고리즘 결과출력
cnb = ComplementNB()
cnb.fit(train_x, train_y)
pre_arr = cnb.predict(test_x)
pre_arr = pre_arr.reshape(10, 12)
print('ComplementNB의 테스트 세트 예측 :\n{}'.format(pre_arr))
print('ComplementNB의 테스트 세트 정확도 : {0:0.2f}%'.format(
cnb.score(test_x, test_y) * 100))
print('------------------------------------------------------')
def complement_bayes(train_data, test_data):
train_y = train_data['state']
train_X = train_data.iloc[:, FEATURES_INDICES]
test_y = test_data['state']
test_X = test_data.iloc[:, FEATURES_INDICES]
CNB = ComplementNB()
CNB.fit(train_X, train_y)
pred_y = CNB.predict(test_X)
evaluate(CNB, test_X, test_y, pred_y)
def CNB_train(features, labels, ds):
"""
Use the Complement Naive Bayes classifier to train
and saves the classifier as pickle file
:param features: List of features from training set
:param labels: List of labels from training set
:param ds: Number of the dataset
"""
CNB_Classifier = ComplementNB(alpha=10.0)
CNB_Classifier.fit(features, labels)
save_classifier(CNB_Classifier, "ds" + ds + "CNB_Classifier.pkl")
def complementNB(tr_vec, tr_ans, val_vec, val_ans, te_vec):
from sklearn.naive_bayes import ComplementNB
clf = ComplementNB()
clf.fit(tr_vec, tr_ans)
print(clf.score(val_vec, val_ans))
print('make predictions ...')
#clf_predictions = clf.predict_proba(te_vec)
preds = clf.predict(te_vec)
pred_test_y = (preds > 0.35).astype(int)
return pred_test_y
def train_naive_bayes(list_of_vector_label_pairs: list, binary_classification=False, alpha=1.0, norm=False):
"""Builds and trains a Naive Bayes classifier on a list of (vector, label) tuples.
Returns the Naive Bayes classifier."""
# Repackage the data for training the classifier
list_of_vector_tuples, list_of_labels = split_data_pairs(list_of_vector_label_pairs, binary_classification)
# Make and fit the classifier
classifier = ComplementNB(alpha=1.0, norm=True)
print("Please wait, training the Naive Bayes classifier now. . .")
classifier.fit(list_of_vector_tuples, list_of_labels)
print("Training complete.")
return classifier
def get_accuracy_of_selection(X, y):
# create k-fold cross validation object
kf = StratifiedKFold(n_splits=25, shuffle=True, random_state=None)
# array of accuracy predictions for this selection of features
accuracies = []
# perform a k-fold cross validation to determine accuracy of selected features
for train_index, test_index in kf.split(X, y):
# split into testing and training data based on the splits
X_train, X_test = X[train_index], X[test_index]
y_train, y_test = y[train_index], y[test_index]
# count each occurrence of the classes to determine frequency
class_count = [0, 0]
for i in y_train:
class_count[int(i)] += 1
# calculate total number of observations and determine prior probability
total = class_count[0] + class_count[1]
prior_probability = [class_count[0] / total, class_count[1] / total]
# define smoothing: "portion of the largest variance of all features that is added to variances for calculation stability."
smoothing = 1e-09
# perform a complement naive bayes
gnb = ComplementNB(class_prior=prior_probability)
gnb.fit(X_train, y_train)
y_pred = gnb.predict(X_test) # predicted class
# y_probs = gnb.predict_proba(X_test) # confidence in each prediction
# for i in range(y_pred.shape[0]):
# if y_pred[i] != y_test[i]:
# print(y_probs[i]) # shows that sometimes we are really really confident in the wrong answer
# determine how accurate we were
size = y_test.size
true_count = (y_test == y_pred).sum()
accuracy_percentage = (true_count) / size
# add to array of accuracy predictions for this selection of features
accuracies.append(accuracy_percentage)
# compute the mean and standard deviation of this selection of features
mean = np.mean(accuracies)
sd = np.std(accuracies)
# print("MEAN: " + str(round(mean*100,2)) + "%")
# print("STANDARD DEVIATION: " + str(round(sd*100,2)) + "%")
return mean, sd
class NBClassifier(super.abstract_classifier):
def __init__(self, train_features, train_labels):
self.train_features = train_features
self.train_labels = train_labels
self.nb_Member = ComplementNB()
def train(
self): # after this function the ComplementNB is ready to classify
self.nb_Member.fit(self.train_features, self.train_labels)
def classify(self, newVector):
return self.nb_Member.predict(newVector)
def naive_bayes(x, y):
# import complementNB,MultinomialNB
cpl = ComplementNB()
mnb = MultinomialNB()
# train our dataset
cpl.fit(x, y)
mnb.fit(x, y)
# perform prediction and find accuracy
y_test_cpl = cpl.predict(x)
y_test_mnb = mnb.predict(x)
return y_test_cpl, y_test_mnb
def naive_bayes(self, name="Train_Test"):
X_train, X_test, y_train, y_test = train_test_split(self.X,
self.Y,
test_size=0.4,
random_state=0)
clf = ComplementNB()
clf.fit(X_train, y_train)
predict = clf.predict(X_test)
f, p, r = self.nbeval(y_test, predict)
line = "{}: F score:{:.3f}\tP score:{:.3f}\tR score:{:.3f}.".format(
name, f, p, r)
self.logger.info(line)
def NB_accuracy_complement(X_train, X_test, y_train, y_test, fold):
gnb = ComplementNB()
gnb.fit(X_train, y_train)
y_pred = gnb.predict(X_test)
accuracy_score(y_test, y_pred)
print(classification_report(y_test, y_pred))
print(confusion_matrix(y_test, y_pred))
print ("mean_squared_error: ", mean_squared_error(y_test, y_pred))
results = cross_val_score(gnb, X_train, y_train, cv = fold)
print("After 5-fold: ", results.mean()*100)