def create(
country_id,
measure_type,
measure_value,
measure_date
):
""" Create a new measure """
measure = Measure(
country_id,
measure_type,
measure_value,
measure_date
)
return measure.save()
def handle_measure():
"""
Trae lista de medida (GET) y agrega una medida (POST)
"""
# POST request
if request.method == 'POST':
body = request.get_json()
if body is None:
raise APIException("You need to specify the request body as a json object", status_code=400)
if 'name' not in body:
raise APIException('You need to specify the name', status_code=400)
if 'unit' not in body:
raise APIException('You need to specify the unit', status_code=400)
if 'symbol' not in body:
raise APIException('You need to specify the symbol', status_code=400)
measure1 = Measure(name=body['name'], unit=body['unit'], symbol=body['symbol'])
db.session.add(measure1)
db.session.commit()
return "ok", 200
# GET request
if request.method == 'GET':
all_measures = Measure.query.all()
all_measures = list(map(lambda x: x.serialize(), all_measures))
return jsonify(all_measures), 200
return "Invalid Method", 404
def LSTM(X_train_seq, X_test_seq, y_train, y_test):
max_words = 10000
max_len = 200
lstm_model = Sequential()
lstm_model.add(Embedding(max_words, 50, input_length=max_len))
lstm_model.add(LSTM_lib(128, dropout=0.25, recurrent_dropout=0.25))
lstm_model.add(Dense(1, activation='sigmoid'))
lstm_model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
print('Train model')
lstm_model.fit(X_train_seq,
y_train,
batch_size=32,
epochs=3,
validation_data=(X_test_seq, y_test))
y_pred = lstm_model.predict_classes(X_test_seq)
lstm_model.save('51_acc_language_model.h5')
acc_score = accuracy_score(y_test, y_pred)
precision, recall, fscore, support = score(y_test,
y_pred,
average='weighted')
target_names = ['Non-Spam', 'Spam']
print(classification_report(y_test, y_pred, target_names=target_names))
return Measure.Measure(acc_score, precision, recall, fscore)
def GET(self, name):
"""Get Measures with descriptions"""
results = Measure.get_all()
data = results_to_measures(results, True)
if not data:
raise Error(NORESULT)
return data
def save_measure_in_db(session, data, point):
objects = []
value = data.pop(0)
my_ssid = Ssid(value[0])
my_bssid = Bssid(value[1])
my_measure = Measure(value[2])
my_channel = Channel(value[3])
my_security = Security(value[4])
my_point = Point(point)
entry = session.query(Ssid).filter(Ssid.ssid_value.like(value[0])).first()
if entry is None:
my_ssid.measure.append(my_measure)
objects.append(my_ssid)
else:
entry.measure.append(my_measure)
entry = session.query(Bssid).filter(Bssid.bssid_value.like(
value[1])).first()
if entry is None:
my_bssid.measure.append(my_measure)
objects.append(my_bssid)
else:
entry.measure.append(my_measure)
entry = session.query(Channel).filter(Channel.channel_number.like(
value[3])).first()
if entry is None:
my_channel.measure.append(my_measure)
objects.append(my_channel)
else:
entry.measure.append(my_measure)
entry = session.query(Security).filter(
Security.security_type.like(value[4])).first()
if entry is None:
my_security.measure.append(my_measure)
objects.append(my_security)
else:
entry.measure.append(my_measure)
entry = session.query(Point).filter(Point.x_location == point.x).filter(
Point.y_location == point.y).first()
if entry is None:
my_point.measure.append(my_measure)
objects.append(my_point)
else:
entry.measure.append(my_measure)
objects.append(my_measure)
session.add_all(objects)
session.commit()
if len(data) != 0:
save_measure_in_db(session, data, point)
else:
return
def postSensor(id):
sensor = Sensor.query.get(id)
data = request.json
if sensor:
data['sensor_id'] = id
data['update_on'] = datetime.now()
m = Measure(**data)
db.session.add(m)
db.session.commit()
else:
abort(404)
return json.dumps({"result": "OK"})
def fill_measures():
"""
Llena con medidas (POST)
"""
# POST request
if request.method == 'POST':
for i in range(3):
measures1 = Measure(name="medida"+str(i+1), unit="unidad"+str(i+1), symbol="símbolo"+str(i+1))
db.session.add(measures1)
db.session.commit()
return "ok", 200
return "Invalid Method", 404
def KNN(X_train, X_test, y_train, y_test):
KNN = KNeighborsClassifier(n_neighbors=1)
KNN.fit(X_train, y_train)
y_knc = KNN.predict(X_test)
print('KNeighbors Accuracy_score: ', accuracy_score(y_test, y_knc))
print('KNeighbors confusion_matrix:/n', confusion_matrix(y_test, y_knc))
print(classification_report(y_test, KNN.predict(X_test)))
joblib.dump(KNN, 'KNN.pkl')
precision, recall, fscore, support = score(y_test,
y_knc,
average='weighted')
acc_score = accuracy_score(y_test, y_knc)
# print('Decision Tree Accuracy: ', accuracy_score(y_test, y_dtc))
# print('Precision : {}'.format(precision))
# print('Recall : {}'.format(recall))
# print('F-score : {}'.format(fscore))
# print('Support : {}'.format(support))
return Measure.Measure(acc_score, precision, recall, fscore)
def Naive_Bayes(X_train, X_test, y_train, y_test):
clf = MultinomialNB()
clf.fit(X_train, y_train)
y_mnb = clf.predict(X_test)
cm = confusion_matrix(y_mnb, y_test)
print(cm)
print('Naive Bayes Accuracy: ', accuracy_score(y_test, y_mnb))
# #in ra độ chính xác của các label
print(classification_report(y_test, clf.predict(X_test)))
joblib.dump(clf, 'NB.pkl')
precision, recall, fscore, support = score(y_test,
y_mnb,
average='weighted')
acc_score = accuracy_score(y_test, y_mnb)
# print('Decision Tree Accuracy: ', accuracy_score(y_test, y_dtc))
# print('Precision : {}'.format(precision))
# print('Recall : {}'.format(recall))
# print('F-score : {}'.format(fscore))
# print('Support : {}'.format(support))
return Measure.Measure(acc_score, precision, recall, fscore)
def SVM(X_train, X_test, y_train, y_test):
SVM = svm.SVC(kernel='linear') # Linear Kernel
# Train the model using the training sets
SVM.fit(X_train, y_train)
# Predict the response for test dataset
y_pred = SVM.predict(X_test)
cm = confusion_matrix(y_pred, y_test)
print(cm)
print('svm Accuracy: ', accuracy_score(y_test, y_pred))
# #in ra độ chính xác của các label
print(classification_report(y_test, SVM.predict(X_test)))
joblib.dump(SVM, 'SVM.pkl')
precision, recall, fscore, support = score(y_test,
y_pred,
average='weighted')
acc_score = accuracy_score(y_test, y_pred)
# print('Decision Tree Accuracy: ', accuracy_score(y_test, y_dtc))
# print('Precision : {}'.format(precision))
# print('Recall : {}'.format(recall))
# print('F-score : {}'.format(fscore))
# print('Support : {}'.format(support))
return Measure.Measure(acc_score, precision, recall, fscore)
def DecisionTree(X_train, X_test, y_train, y_test):
# huấn luyện mô hình bằng tập train và test
dtc = DecisionTreeClassifier(min_samples_split=7, random_state=None)
dtc.fit(X_train, y_train)
# dự đoán cho tập dữ liệu test
y_dtc = dtc.predict(X_test)
cm = confusion_matrix(y_dtc, y_test)
print(cm)
print('Decision Tree Accuracy: ', accuracy_score(y_test, y_dtc))
# in ra độ chính xác của các label
print(classification_report(y_test, dtc.predict(X_test)))
joblib.dump(dtc, 'DecisionTree.pkl')
precision, recall, fscore, support = score(y_test,
y_dtc,
average='weighted')
acc_score = accuracy_score(y_test, y_dtc)
# print('Decision Tree Accuracy: ', accuracy_score(y_test, y_dtc))
# print('Precision : {}'.format(precision))
# print('Recall : {}'.format(recall))
# print('F-score : {}'.format(fscore))
# print('Support : {}'.format(support))
return Measure.Measure(acc_score, precision, recall, fscore)
def index(request):
if request.method == 'POST':
measure = Measure()
measure.ch1 = request.POST['ch1']
measure.ch2 = request.POST['ch2']
measure.ch3 = request.POST['ch3']
measure.ch4 = request.POST['ch4']
measure.ch5 = request.POST['ch5']
measure.ch6 = request.POST['ch6']
measure.ch7 = request.POST['ch7']
measure.ch8 = request.POST['ch8']
measure.ch9 = request.POST['ch9']
measure.ch10 = request.POST['ch10']
measure.ch11 = request.POST['ch11']
measure.ch12 = request.POST['ch12']
measure.therm = request.POST['therm']
measure.temp = request.POST['temp']
measure.excv = request.POST['excv']
measure.batt = request.POST['batt']
measure.time = request.POST['time']
measure.date = request.POST['date']
measure.save()
list = Measure.objects
listOfMeasure = []
for m in list:
listOfMeasure.append(m)
context = {
'listOfMeasure': listOfMeasure,
}
return render(request, 'crud/index.html', context)
def load(request):
import os
fh = open(os.path.join(FILE_LOAD_DIR, 'L4286F08.txt'), "r")
fh.readline()
p = fh.readline()
while p != '':
p = p.split('\t')
p[-1] = p[-1].split('\n')[0]
measure = Measure()
measure.ch1 = p[0]
measure.ch2 = p[1]
measure.ch3 = p[2]
measure.ch4 = p[3]
measure.ch5 = p[4]
measure.ch6 = p[5]
measure.ch7 = p[6]
measure.ch8 = p[7]
measure.ch9 = p[8]
measure.ch10 = p[9]
measure.ch11 = p[10]
measure.ch12 = p[11]
measure.therm = p[12]
measure.temp = p[13]
measure.excv = p[14]
measure.batt = p[15]
measure.time = p[16]
measure.date = p[17]
measure.save()
p = fh.readline()
return HttpResponseRedirect(reverse('crud:index'))
