refactor dataset generation

This commit is contained in:
René Knaebel 2017-07-05 21:19:19 +02:00
parent 772b07847f
commit b2f5c56019
3 changed files with 50 additions and 36 deletions

View File

@ -1,3 +1,3 @@
test: test:
python3 main.py --epochs 1 --batch 64 python3 main.py --epochs 1 --batch 64 --train data/rk_data.csv.gz --test data/rk_data.csv.gz

View File

@ -117,12 +117,27 @@ def create_dataset_from_flows(user_flow_df, char_dict, max_len, window_size=10,
break break
print("create training dataset") print("create training dataset")
return create_dataset_from_lists( domain_tr, flow_tr, hits_tr, names_tr, server_tr, trusted_hits_tr = create_dataset_from_lists(
domains=domains, features=features, vocab=char_dict, domains=domains, features=features, vocab=char_dict,
max_len=max_len, max_len=max_len,
use_cisco_features=use_cisco_features, urlSIPDIct=dict(), use_cisco_features=use_cisco_features, urlSIPDIct=dict(),
window_size=window_size) window_size=window_size)
# make client labels discrete with 4 different values
# TODO: use trusted_hits_tr for client classification too
client_labels = np.apply_along_axis(lambda x: discretize_label(x, 3), 0, np.atleast_2d(hits_tr))
# select only 1.0 and 0.0 from training data
pos_idx = np.where(client_labels == 1.0)[0]
neg_idx = np.where(client_labels == 0.0)[0]
idx = np.concatenate((pos_idx, neg_idx))
# choose selected sample to train on
domain_tr = domain_tr[idx]
flow_tr = flow_tr[idx]
client_labels = client_labels[idx]
server_labels = server_tr[idx]
return domain_tr, flow_tr, client_labels, server_labels
def create_dataset_from_lists(domains, features, vocab, max_len, def create_dataset_from_lists(domains, features, vocab, max_len,
use_cisco_features=False, urlSIPDIct=dict(), use_cisco_features=False, urlSIPDIct=dict(),
@ -185,9 +200,11 @@ def discretize_label(values, threshold):
return 0.0 return 0.0
def get_user_flow_data(): def get_user_flow_data(csv_file):
df = pd.read_csv("data/rk_data.csv.gz") df = pd.read_csv(csv_file)
df.drop("Unnamed: 0", 1, inplace=True) keys = ["duration", "bytes_down", "bytes_up", "domain", "timeStamp", "server_ip", "user_hash", "virusTotalHits",
"serverLabel", "trustedHits"]
df = df[keys]
df.set_index(keys=['user_hash'], drop=False, inplace=True) df.set_index(keys=['user_hash'], drop=False, inplace=True)
return df return df

59
main.py
View File

@ -1,6 +1,5 @@
import argparse import argparse
import numpy as np
from keras.utils import np_utils from keras.utils import np_utils
import dataset import dataset
@ -8,17 +7,20 @@ import models
parser = argparse.ArgumentParser() parser = argparse.ArgumentParser()
parser.add_argument("--modes", action="store", dest="modes", nargs="+") # parser.add_argument("--modes", action="store", dest="modes", nargs="+")
parser.add_argument("--train", action="store", dest="train_data",
default="data/full_dataset.csv.tar.bz2")
parser.add_argument("--test", action="store", dest="test_data",
default="data/full_future_dataset.csv.tar.bz2")
# parser.add_argument("--data", action="store", dest="data",
# default="data/")
#
# parser.add_argument("--h5data", action="store", dest="h5data", # parser.add_argument("--h5data", action="store", dest="h5data",
# default="") # default="")
# #
# parser.add_argument("--model", action="store", dest="model", parser.add_argument("--model", action="store", dest="model",
# default="model_x") default="model_x")
#
# parser.add_argument("--pred", action="store", dest="pred", # parser.add_argument("--pred", action="store", dest="pred",
# default="") # default="")
# #
@ -66,8 +68,7 @@ parser.add_argument("--domain_embd", action="store", dest="domain_embedding",
# #
# parser.add_argument("--tmp", action="store_true", dest="tmp") # parser.add_argument("--tmp", action="store_true", dest="tmp")
# #
# parser.add_argument("--test", action="store", dest="test_image", # parser.add_argument("--test", action="store_true", dest="test")
# default=6, choices=range(7), type=int)
args = parser.parse_args() args = parser.parse_args()
@ -82,37 +83,24 @@ def main():
# parameter # parameter
cnnDropout = 0.5 cnnDropout = 0.5
cnnHiddenDims = 1024 cnnHiddenDims = 1024
flowFeatures = 3
numCiscoFeatures = 30 numCiscoFeatures = 30
kernel_size = 3 kernel_size = 3
drop_out = 0.5 drop_out = 0.5
filters = 128 filters = 128
char_dict = dataset.get_character_dict() char_dict = dataset.get_character_dict()
user_flow_df = dataset.get_user_flow_data() user_flow_df = dataset.get_user_flow_data(args.train_data)
print("create training dataset") print("create training dataset")
domain_tr, flow_tr, hits_tr, names_tr, server_tr, trusted_hits_tr = dataset.create_dataset_from_flows( domain_tr, flow_tr, client_tr, server_tr = dataset.create_dataset_from_flows(
user_flow_df, char_dict, user_flow_df, char_dict,
max_len=args.domain_length, window_size=args.window) max_len=args.domain_length, window_size=args.window)
# make client labels discrete with 4 different values
# TODO: use trusted_hits_tr for client classification too
client_labels = np.apply_along_axis(lambda x: dataset.discretize_label(x, 3), 0, np.atleast_2d(hits_tr))
# select only 1.0 and 0.0 from training data
pos_idx = np.where(client_labels == 1.0)[0]
neg_idx = np.where(client_labels == 0.0)[0]
idx = np.concatenate((pos_idx, neg_idx))
# choose selected sample to train on
domain_tr = domain_tr[idx]
flow_tr = flow_tr[idx]
client_labels = client_labels[idx]
server_labels = server_tr[idx]
shared_cnn = models.renes_networks.get_embedding(len(char_dict) + 1, args.embedding, args.domain_length, shared_cnn = models.renes_networks.get_embedding(len(char_dict) + 1, args.embedding, args.domain_length,
args.hidden_char_dims, kernel_size, args.domain_embedding, 0.5) args.hidden_char_dims, kernel_size, args.domain_embedding, 0.5)
shared_cnn.summary() shared_cnn.summary()
model = models.renes_networks.get_model(cnnDropout, flowFeatures, args.domain_embedding, model = models.renes_networks.get_model(cnnDropout, flow_tr.shape[-1], args.domain_embedding,
args.window, args.domain_length, filters, kernel_size, args.window, args.domain_length, filters, kernel_size,
cnnHiddenDims, shared_cnn) cnnHiddenDims, shared_cnn)
model.summary() model.summary()
@ -121,14 +109,23 @@ def main():
loss='binary_crossentropy', loss='binary_crossentropy',
metrics=['accuracy']) metrics=['accuracy'])
client_labels = np_utils.to_categorical(client_labels, 2) client_tr = np_utils.to_categorical(client_tr, 2)
server_labels = np_utils.to_categorical(server_labels, 2) server_tr = np_utils.to_categorical(server_tr, 2)
model.fit([domain_tr, flow_tr], model.fit([domain_tr, flow_tr],
[client_labels, server_labels], [client_tr, server_tr],
batch_size=args.batch_size, batch_size=args.batch_size,
epochs=args.epochs, epochs=args.epochs,
shuffle=True) shuffle=True,
# TODO: for validation we use future data -> validation_data=(testData,testLabel)) validation_split=0.2)
def test():
char_dict = dataset.get_character_dict()
user_flow_df = dataset.get_user_flow_data(args.test_data)
domain_val, flow_val, client_val, server_val = dataset.create_dataset_from_flows(
user_flow_df, char_dict,
max_len=args.domain_length, window_size=args.window)
# TODO: get model and exec model.evaluate(...)
if __name__ == "__main__": if __name__ == "__main__":