def create_classifier(path, model, **kwargs):
train = dataset.from_npy(path, "train.npy")
test = dataset.from_npy(path, "test.npy")
### Logistic Regression ###
train_X, train_y = train[:, :-1], train[:, -1]
test_X, test_y = test[:, :-1], test[:, -1]
clf = model(**kwargs)
start = time.time()
clf.fit(train_X, train_y)
finit = time.time()
delta = finit - start
print "Training took %0.3f seconds" % delta
print clf
clf_pred = clf.predict(test_X)
print "Accuracy: " + str(accuracy_score(test_y, clf_pred))
print "F1 Score: " + str(f1_score(test_y, clf_pred))
cm = confusion_matrix(test_y, clf_pred)
plot_confusion_matrix(cm)
plt.draw()
return clf
def create_classifier(path, model, **kwargs):
train = dataset.from_npy(path, "train.npy")
test = dataset.from_npy(path, "test.npy")
### Logistic Regression ###
train_X, train_y = train[:, :-1], train[:, -1]
test_X, test_y = test[:, :-1], test[:, -1]
clf = model(**kwargs)
start = time.time()
clf.fit(train_X, train_y)
finit = time.time()
delta = finit - start
print "Training took %0.3f seconds" % delta
print clf
clf_pred = clf.predict(test_X)
print "Accuracy: " + str(accuracy_score(test_y, clf_pred))
print "F1 Score: " + str(f1_score(test_y, clf_pred))
cm = confusion_matrix(test_y, clf_pred)
plot_confusion_matrix(cm)
plt.draw()
return clf
def controller(folder):
path = os.path.join(os.path.expanduser('~'), 'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset', folder)
X = dataset.from_npy(path, 'X.npy')
y = dataset.from_npy(path, 'y.npy')
classifiers = [GaussianNB(), BernoulliNB(), LinearSVC(), DecisionTreeClassifier(), RandomForestClassifier(),
ExtraTreesClassifier(), LogisticRegression(), SGDClassifier()]
for clf in classifiers:
accuracy = get_accuracy(X, y, 10, clf) * 100
print type(clf).__name__, ':', accuracy
def controller(folder):
path = os.path.join(
os.path.expanduser('~'),
'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset', folder)
X = dataset.from_npy(path, 'X.npy')
y = dataset.from_npy(path, 'y.npy')
classifiers = [
GaussianNB(),
BernoulliNB(),
LinearSVC(),
DecisionTreeClassifier(),
RandomForestClassifier(),
ExtraTreesClassifier(),
LogisticRegression(),
SGDClassifier()
]
for clf in classifiers:
accuracy = get_accuracy(X, y, 10, clf) * 100
print type(clf).__name__, ':', accuracy
def initialize(self):
# Initialize grid layout manager
self.grid()
self.entryVariable = Tkinter.StringVar()
# Create the widget
self.entry = Tkinter.Entry(self, textvariable=self.entryVariable)
# When a cell grows larger than the widget is contains,
# you can ask the widget to stick to some edges of the cell.
# That's the sticky='EW'.
# (E=east (left), W=West (right), N=North (top), S=South (bottom))
# We specified 'EW', which means the widget will try to stick to
# both left and right edges of its cell.
self.entry.grid(column=0, row=0, sticky='EW')
# <Return> event handler
self.entry.bind("<Return>", self.OnPressEnter)
self.entryVariable.set(u"Enter URL here.")
# Load classifier (Choose a path where the voter is saved)
path = os.path.join(
os.path.expanduser('~'),
'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset\\Trial_03')
self.voter = load_voter(path)
test = dataset.from_npy(path, 'test.npy')
test_X, test_y = test[:, :-1], test[:, -1]
self.voter.confusion_matrix(test_X, test_y)
### Button ###
# Note that in this case, we do not keep a reference to the button
# (because we will not read or alter its value later)
button = Tkinter.Button(self,
text=u"Submit",
command=self.OnButtonClick)
button.grid(column=1, row=0)
### Label ###
self.labelVariable = Tkinter.StringVar()
# White text on a blue background
self.label = Tkinter.Label(self,
textvariable=self.labelVariable,
anchor="w",
fg="white",
bg="blue")
self.label.grid(column=0, row=1, columnspan=2, sticky='EW')
self.labelVariable.set(u"Welcome to URL Classifier!")
# Enable resizing
self.grid_columnconfigure(0, weight=1)
# Adding constraint - Prevent vertical resizing
self.resizable(width=True, height=False)
# Set window to not grow and shrink automatically
self.update()
self.geometry(self.geometry())
self.entry.focus_set()
self.entry.selection_range(0, Tkinter.END)
def initialize(self):
# Initialize grid layout manager
self.grid()
self.entryVariable = Tkinter.StringVar()
# Create the widget
self.entry = Tkinter.Entry(self, textvariable=self.entryVariable)
# When a cell grows larger than the widget is contains,
# you can ask the widget to stick to some edges of the cell.
# That's the sticky='EW'.
# (E=east (left), W=West (right), N=North (top), S=South (bottom))
# We specified 'EW', which means the widget will try to stick to
# both left and right edges of its cell.
self.entry.grid(column=0, row=0, sticky='EW')
# <Return> event handler
self.entry.bind("<Return>", self.OnPressEnter)
self.entryVariable.set(u"Enter URL here.")
# Load classifier (Choose a path where the voter is saved)
path = os.path.join(os.path.expanduser('~'), 'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset\\Trial_03')
self.voter = load_voter(path)
test = dataset.from_npy(path, 'test.npy')
test_X, test_y = test[:, :-1], test[:, -1]
self.voter.confusion_matrix(test_X, test_y)
### Button ###
# Note that in this case, we do not keep a reference to the button
# (because we will not read or alter its value later)
button = Tkinter.Button(self, text=u"Submit",
command=self.OnButtonClick)
button.grid(column=1, row=0)
### Label ###
self.labelVariable = Tkinter.StringVar()
# White text on a blue background
self.label = Tkinter.Label(self, textvariable=self.labelVariable,
anchor="w", fg="white", bg="blue")
self.label.grid(column=0, row=1, columnspan=2, sticky='EW')
self.labelVariable.set(u"Welcome to URL Classifier!")
# Enable resizing
self.grid_columnconfigure(0,weight=1)
# Adding constraint - Prevent vertical resizing
self.resizable(width=True, height=False)
# Set window to not grow and shrink automatically
self.update()
self.geometry(self.geometry())
self.entry.focus_set()
self.entry.selection_range(0, Tkinter.END)
trial = 'Trial_03'
# trial = 'Lexical'
path = os.path.join(
os.path.expanduser('~'),
'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset', trial)
# voter = create_voter()
#
# train = dataset.from_npy(path, 'train.npy')
# train_X, train_y = train[:, :-1], train[:, -1]
# voter.fit(train_X, train_y)
# save_voter(voter, path)
voter = load_voter(path)
test = dataset.from_npy(path, 'test.npy')
test_X, test_y = test[:, :-1], test[:, -1]
voter.confusion_matrix(test_X, test_y)
# for clf in voter._classifiers:
# print type(clf).__name__
# cm = confusion_matrix(test_y, clf.predict(test_X))
# plot_confusion_matrix(cm)
# plt.draw()
#
test_urls = [
'https://www.youtube.com/watch?v=4WM6hB7l4Lc&list=PLQVvvaa0QuDd0flgGphKCej-9jp-QdzZ3&index=12&feature=iv&src_vid=81ZGOib7DTk&annotation_id=annotation_1856532697',
'http://ld.mediaget.com/index.php?l=ru&fu=http:/www.playground.ru/download/?cheat=grand_theft_auto_4_gta_iv_episodes_from_liberty_city_eflc_sohranenie_100-41709&r=playground.ru&f=grand_theft_auto_4_gta_iv_episodes_from_liberty_city_eflc__&%23x421;&%23x43e;&%23x445;&%23x440;&%23x430;&%23x43d;&%23x435;&%23x43d;&%23x438;&%23x435;_100%25',
'https://raw.github.com/inquisb/shellcodeexec/master/windows/shellcodeexec.x32.exe',
'http://www.ezthemes.com/site_advertisers/extrafindWD.exe'
]
# controller('Lexical')
# controller('Trial_03')
trial = 'Trial_03'
# trial = 'Lexical'
path = os.path.join(os.path.expanduser('~'), 'OneDrive\\RPI\\Summer Project\\URL Classifier\\Dataset', trial)
# voter = create_voter()
#
# train = dataset.from_npy(path, 'train.npy')
# train_X, train_y = train[:, :-1], train[:, -1]
# voter.fit(train_X, train_y)
# save_voter(voter, path)
voter = load_voter(path)
test = dataset.from_npy(path, 'test.npy')
test_X, test_y = test[:, :-1], test[:, -1]
voter.confusion_matrix(test_X, test_y)
# for clf in voter._classifiers:
# print type(clf).__name__
# cm = confusion_matrix(test_y, clf.predict(test_X))
# plot_confusion_matrix(cm)
# plt.draw()
#
test_urls = [
'https://www.youtube.com/watch?v=4WM6hB7l4Lc&list=PLQVvvaa0QuDd0flgGphKCej-9jp-QdzZ3&index=12&feature=iv&src_vid=81ZGOib7DTk&annotation_id=annotation_1856532697',
'http://ld.mediaget.com/index.php?l=ru&fu=http:/www.playground.ru/download/?cheat=grand_theft_auto_4_gta_iv_episodes_from_liberty_city_eflc_sohranenie_100-41709&r=playground.ru&f=grand_theft_auto_4_gta_iv_episodes_from_liberty_city_eflc__&%23x421;&%23x43e;&%23x445;&%23x440;&%23x430;&%23x43d;&%23x435;&%23x43d;&%23x438;&%23x435;_100%25',
'https://raw.github.com/inquisb/shellcodeexec/master/windows/shellcodeexec.x32.exe',
'http://www.ezthemes.com/site_advertisers/extrafindWD.exe']
for test_url in test_urls: