def train():
op = Operation()
op.user = request.form['login']
op.pressedtime = float(request.form['pressedtime'])
op.flytime = float(request.form['onflytime'])
op.keytimes = request.form['keytimes']
op.save()
return render_template("train.html", myOp=op)
def train():
global nObservations
records = []
for v in collection.find({}):
op = Operation()
op.pressedtime = v["pressedtime"]
op.flytime = v["flytime"]
op.keytimes = v["keytimes"]
op.user = v["user"]
records.append(op.toTuple())
features = ["flytime", "pressedtime"]
for x in range(1, 19):
features.append("kt" + ` x `)
target = "user"
labels = features + [target]
print(labels)
df = pd.DataFrame.from_records(records, columns=labels)
# print ("Before filtering ", df.shape)
#
# df = df[df.ellapsed < 10]
#
# print ("After filtering ", df.shape)
#
# if df.shape[0] <= nObservations:
# return
#
# nObservations = df.shape[0]
#LABEL ENCODER DE USUARIOS
le = preprocessing.LabelEncoder()
df[target] = le.fit_transform(df[target])
#MOSTRAR BASE DE DATOS
print(df)
params = {'n_estimators': 500, 'max_depth': 4}
regr = ensemble.GradientBoostingClassifier(**params)
X = df[features]
y = df[target]
regr = regr.fit(X, y)
prediction = regr.predict(X)
print("Error", mean_squared_error(prediction, y))
joblib.dump(regr, 'regr.pkl', protocol=2)
joblib.dump(le, 'labelenc.pkl', protocol=2)
print("DUMP")
return df.shape[0]
def logindificil():
op = Operation()
op.pressedtime = float(request.form['pressedtime'])
op.flytime = float(request.form['onflytime'])
op.keytimes = request.form['keytimes']
regr = joblib.load('regr.pkl')
op.predict(regr)
le = joblib.load('labelenc.pkl')
op.prediction = le.inverse_transform([op.prediction])[0]
return render_template("prediccion.html", myOp=op)
def loginsencillo():
op = Operation()
op.user = request.form['login']
op.pressedtime = float(request.form['pressedtime'])
op.flytime = float(request.form['onflytime'])
op.keytimes = request.form['keytimes']
regr = joblib.load('regr.pkl')
op.predict(regr)
le = joblib.load('labelenc.pkl')
if (le.inverse_transform([op.prediction])[0] == op.user):
return render_template("accepted.html", myOp=op)
else:
return render_template("notaccepted.html", myOp=op)
