Skip to content

Add files via upload #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Lasciarti wants to merge 3 commits into
base: master
Choose a base branch
from
Open

Add files via upload #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a7ad58b
Add files via upload
Lasciarti Oct 7, 2020
1ba918e
Add files via upload
Lasciarti Oct 24, 2020
af466d0
Add files via upload
Lasciarti Oct 24, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 26 additions & 12 deletions examples/rnn_main_single.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,8 @@
from pcap.parser import PCAP_PKTS, _pcap2flows
from util.tool import dump_data, load_data

from time import *

RANDOM_STATE = 100

cuda = True if torch.cuda.is_available() else False
Expand All @@ -42,7 +44,7 @@ def set_random_state(random_state=100):
set_random_state(random_state=RANDOM_STATE)


def raw2features(raw_features, header=True, MTU=1500, normalize=True):
def raw2features(raw_features, data_type=True, MTU=1500, normalize=True):
"""Extract features for the detection model

Parameters
Expand All @@ -64,10 +66,13 @@ def normalize_bytes(flow):
feat_0 = v_lst[0]
feat_i_lst = v_lst[1:]

if header:
if data_type == 'header':
tmp_v = [v['header'] for v in feat_i_lst]
tmp_v = [v + [0] * (40 - len(v)) if len(v) < 40 else v[:40] for v in tmp_v]
elif data_type == 'header_payload':
tmp_v = [v['header'] + v['payload'] for v in feat_i_lst]
tmp_v = [v + [0] * (MTU - len(v)) if len(v) < MTU else v[:MTU] for v in tmp_v]
else:
else: # data_type=='payload':
payload_len = MTU - 40
tmp_v = [v['payload'] for v in feat_i_lst]
tmp_v = [v + [0] * (payload_len - len(v)) if len(v) < payload_len else v[:payload_len] for v in tmp_v]
Expand All @@ -80,7 +85,7 @@ def normalize_bytes(flow):
return X


def load_flow_data(overwrite=False, random_state=100, full_flow=True):
def load_flow_data(overwrite=False, random_state=100, full_flow=True, data_type='header'):
"""Get raw features from PCAP and store them into disk

Parameters
Expand All @@ -101,8 +106,8 @@ def load_flow_data(overwrite=False, random_state=100, full_flow=True):
# # # # # # #
# # # # # # 'DS20_PU_SMTV/DS21-srcIP_10.42.0.1',
# # # # # # # #
#'DS40_CTU_IoT/DS41-srcIP_10.0.2.15',
'DS40_CTU_IoT/DS42-srcIP_192.168.1.196',
'DS40_CTU_IoT/DS41-srcIP_10.0.2.15',
# 'DS40_CTU_IoT/DS42-srcIP_192.168.1.196',
# # # # #
# # # # # # 'DS50_MAWI_WIDE/DS51-srcIP_202.171.168.50',
# # # 'DS50_MAWI_WIDE/DS51-srcIP_202.171.168.50',
Expand All @@ -128,7 +133,7 @@ def load_flow_data(overwrite=False, random_state=100, full_flow=True):

dataset_name = datasets[0]
print(f'dataset: {dataset_name}')
in_dir = 'data/data_reprst/pcaps'
in_dir = './examples/data/data_reprst/pcaps'
if dataset_name == 'DS40_CTU_IoT/DS42-srcIP_192.168.1.196':
in_norm_file = f'{in_dir}/{dataset_name}/2019-01-09-22-46-52-src_192.168.1.196_CTU_IoT_CoinMiner_anomaly.pcap'
in_abnorm_file = f'{in_dir}/{dataset_name}/2018-12-21-15-50-14-src_192.168.1.195-CTU_IoT_Mirai_normal.pcap'
Expand Down Expand Up @@ -175,9 +180,9 @@ def load_flow_data(overwrite=False, random_state=100, full_flow=True):
out_abnorm_file = in_abnorm_file + '-raw_abnormal_features.dat'
dump_data(abnorm_pp.features, out_abnorm_file)

X_norm = raw2features(load_data(out_norm_file), header=False)
X_norm = raw2features(load_data(out_norm_file), data_type=data_type)
y_norm = [0] * len(X_norm)
X_abnorm = raw2features(load_data(out_abnorm_file), header=False)
X_abnorm = raw2features(load_data(out_abnorm_file), data_type=data_type)
y_abnorm = [1] * len(X_abnorm)

return split_train_test(X_norm, y_norm, X_abnorm, y_abnorm, random_state)
Expand All @@ -202,6 +207,7 @@ def split_train_test(X_norm, y_norm, X_abnorm, y_abnorm, random_state=100):
# X_norm = sklearn.utils.shuffle(X_norm, random_state)
random.Random(random_state).shuffle(X_norm) #注意此处打乱数据的作用
size = int(len(y_norm) // 2) if len(y_norm) <= len(y_abnorm) else min(400, len(y_abnorm))
size = 8800
X_test = X_norm[-size:] + X_abnorm[:size]
y_test = y_norm[-size:] + y_abnorm[:size]
X_train = X_norm[:-size]
Expand All @@ -212,13 +218,21 @@ def split_train_test(X_norm, y_norm, X_abnorm, y_abnorm, random_state=100):


def main(random_state=100):
X_train, y_train, X_test, y_test = load_flow_data(random_state=random_state)

rnn = RNN(n_epochs=100, in_dim=1460, out_dim=10, n_layers=1, lr=1e-3, bias=False, random_state=random_state)
data_type = 'payload' # header, header_payload
X_train, y_train, X_test, y_test = load_flow_data(random_state=random_state, data_type=data_type)
in_dim = len(X_train[0][0])
rnn = RNN(n_epochs=100, in_dim=in_dim, out_dim=10, n_layers=1, lr=1e-3, bias=False, random_state=random_state)

rnn.train(X_train=X_train, y_train=y_train, X_val=X_test, y_val=y_test, split=True)

#计算test运行时间
begin_time = time()

rnn.test(X_test=X_test, y_test=y_test, split=True)

end_time = time()
run_time = end_time-begin_time
print("test time：",run_time)


if __name__ == '__main__':
Expand Down
Loading