Repository navigation
Expand file tree
/
Copy pathtransform.py
More file actions
257 lines (229 loc) · 10.4 KB
/
Copy pathtransform.py
File metadata and controls
257 lines (229 loc) · 10.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
import pandas as pd
import numpy as np
from tensorflow.contrib.learn.python.learn.datasets import base
from tensorflow.python.framework import dtypes
# function for Data pre-processing method according to IDS
def prepare_feature(data):
#data = pd.read_csv('data/UNSW_NB15_training-set.csv')
max_feature = dict()
min_feature = dict()
# can change in exception if you don't want this process on any feature
# exception = ['label', 'proto', 'service', 'state', 'attack_cat']
exception = ['label']
for col in data.columns:
max_feature[col] = data[col].max()
min_feature[col] = data[col].min()
for col in data.columns:
if col in exception:
continue
data[col] = data[col] - min_feature[col]
data[col] = data[col] / (max_feature[col] - min_feature[col])
return data
# Need to find out actual number of classes from data set
def extract_labels(data, one_hot=False, num_classes=2):
# label from label column of Data
labels = data['label'].to_numpy()
if one_hot:
# labels = to_categorical(labels)
return dense_to_one_hot(labels, num_classes)
return labels
def extract_labels_attack(data, one_hot=False, num_classes=2):
# label from label column of Data
labels = data['label'].to_numpy()
label_normal=labels[labels==0]
label_attack = labels[labels != 0]
if one_hot:
# labels = to_categorical(labels)
label_normal=dense_to_one_hot(label_normal, num_classes)
label_attack = dense_to_one_hot(label_attack, num_classes)
return label_normal, label_attack
# Extract feature + label values and replace any non numerical feature value with numerical value with split attack_cat
def extract_features_attack(data):
proto_unique = data['proto'].unique()
service_unique = data['service'].unique()
state_unique = data['state'].unique()
attack_unique = ['Normal', 'Backdoor', 'Analysis', 'Fuzzers', 'Shellcode', 'Reconnaissance',
'Exploits', 'DoS', 'Worms', 'Generic']
# print("attack_unique", attack_unique)
data.pop('label')
data.pop('id')
for i in range(len(proto_unique)):
data = data.replace(proto_unique[i], i)
for i in range(len(service_unique)):
data = data.replace(service_unique[i], i)
for i in range(len(state_unique)):
data = data.replace(state_unique[i], i)
for i in range(len(attack_unique)):
data = data.replace(attack_unique[i], i)
data_normal = data[data['attack_cat'] == 0]
data_attack = data[data['attack_cat'] != 0]
return data, data_normal, data_attack
# Extract feature values and replace any non numerical feature value with numerical value here
def extract_features(data):
#import pandas as pd
#data = pd.read_csv('data/UNSW_NB15_testing-set.csv')
proto_unique = data['proto'].unique()
service_unique = data['service'].unique()
state_unique = data['state'].unique()
attack_unique = ['Normal', 'Backdoor', 'Analysis', 'Fuzzers', 'Shellcode', 'Reconnaissance',
'Exploits', 'DoS', 'Worms', 'Generic']
data.pop('label')
data.pop('id')
for i in range(len(proto_unique)):
data = data.replace(proto_unique[i], i)
for i in range(len(service_unique)):
data = data.replace(service_unique[i], i)
for i in range(len(state_unique)):
data = data.replace(state_unique[i], i)
for i in range(len(attack_unique)):
data = data.replace(attack_unique[i], i)
#print(data)
return data
def extract_labels_multi(data, one_hot=False, num_classes=10):
# label from label column of Data
labels = data['attack_cat'].to_numpy()
if one_hot:
# labels = to_categorical(labels)
return dense_to_one_hot(labels, num_classes)
data.pop('attack_cat')
return labels
def extract_labels_attack_multi(data, one_hot=False, num_classes=10):
# label from label column of Data
labels = data['attack_cat'].to_numpy()
label_normal=labels[labels==0]
label_attack = labels[labels != 0]
if one_hot:
# labels = to_categorical(labels)
label_normal=dense_to_one_hot(label_normal, num_classes)
label_attack = dense_to_one_hot(label_attack, num_classes)
data.pop('attack_cat')
return label_normal, label_attack
def data_importer_GAN(one_hot=False):
TRAIN_SET = pd.read_csv('data/UNSW_NB15_testing-set.csv')
TEST_SET = pd.read_csv('data/UNSW_NB15_training-set.csv')
dtype = dtypes.float32
ACTUAL_TRAIN_SET = TRAIN_SET
train_labels = extract_labels(ACTUAL_TRAIN_SET, one_hot=one_hot)
train_samples = extract_features(ACTUAL_TRAIN_SET)
test_labels_normal, test_labels_attack = extract_labels_attack(TEST_SET, one_hot=one_hot)
test_samples_full, test_samples_normal, test_samples_attack = extract_features_attack(TEST_SET)
test_normal = DataSet(test_samples_normal, test_labels_normal, dtype=dtype)
test_attack = DataSet(test_samples_attack,
test_labels_attack,
dtype=dtype)
train = DataSet(train_samples, train_labels, dtype=dtype)
return base.Datasets(train=train, validation=test_attack, test=test_normal)
def data_importer_IDS(Evaluate=False):
TRAIN_SET = pd.read_csv('data/UNSW_NB15_testing-set.csv')
TEST_SET = pd.read_csv('data/UNSW_NB15_training-set.csv')
total_sample=TRAIN_SET.shape[0]
print("total training sample", total_sample)
print("total test sample", TEST_SET.shape[0])
dtype = dtypes.float32
if Evaluate==False:
df = pd.DataFrame(np.random.randn(len(TRAIN_SET), 2))
mask = np.random.rand(len(df)) < 0.8
ACTUAL_TRAIN_SET = TRAIN_SET[mask]
# print(ACTUAL_TRAIN_SET.head())
VALIDATION_SET = TRAIN_SET[~mask]
# validation_labels = extract_labels(VALIDATION_SET, one_hot=True)
# validation_samples = extract_features(VALIDATION_SET)
#
# test_labels = extract_labels(TEST_SET, one_hot=True)
# test_samples = extract_features(TEST_SET)
validation_samples = extract_features(VALIDATION_SET)
validation_labels = extract_labels_multi(validation_samples, one_hot=True)
test_samples = extract_features(TEST_SET)
test_labels = extract_labels_multi(test_samples, one_hot=True)
validation = DataSet(validation_samples,
validation_labels,
dtype=dtype)
test = DataSet(test_samples, test_labels, dtype=dtype)
else:
ACTUAL_TRAIN_SET=TRAIN_SET
train_samples_normal = TRAIN_SET[TRAIN_SET['attack_cat']=='Normal']
train_samples_attack = TRAIN_SET[TRAIN_SET['attack_cat']!='Normal']
# test_labels_normal, test_labels_attack = extract_labels_attack(TEST_SET, one_hot=True)
# test_samples_full, test_samples_normal, test_samples_attack = extract_features_attack(TEST_SET)
test_samples_full, test_samples_normal, test_samples_attack = extract_features_attack(TEST_SET)
test_labels_normal, test_labels_attack = extract_labels_attack_multi(test_samples_full, one_hot=True)
validation = DataSet(test_samples_attack,
test_labels_attack,
dtype=dtype)
test = DataSet(test_samples_normal, test_labels_normal, dtype=dtype)
print("training normal sample", train_samples_normal.shape[0])
print("training attack sample", train_samples_attack.shape[0])
print("test normal sample", test_samples_normal.shape[0])
print("test attack sample", test_samples_attack.shape[0])
# train_labels = extract_labels(ACTUAL_TRAIN_SET, one_hot=True)
# train_samples = extract_features(ACTUAL_TRAIN_SET)
train_samples = extract_features(ACTUAL_TRAIN_SET)
train_labels = extract_labels_multi(train_samples,one_hot=True)
train = DataSet(train_samples, train_labels, dtype=dtype)
return base.Datasets(train=train, validation=validation, test=test)
def dense_to_one_hot(labels_dense, num_classes):
"""Convert class labels from scalars to one-hot vectors."""
num_labels = labels_dense.shape[0]
index_offset = np.arange(num_labels) * num_classes
labels_one_hot = np.zeros((num_labels, num_classes))
labels_one_hot.flat[index_offset + labels_dense.ravel()] = 1
return labels_one_hot
class DataSet(object):
def __init__(self,
samples,
labels,
one_hot=False,
dtype=dtypes.float32):
"""Construct a DataSet.
one_hot arg is used only if fake_data is true. `dtype` can be either
`uint8` to leave the input as `[0, 255]`, or `float32` to rescale into
`[0, 1]`.
"""
dtype = dtypes.as_dtype(dtype).base_dtype
if dtype not in (dtypes.uint8, dtypes.float32):
raise TypeError('Invalid image dtype %r, expected uint8 or float32' %
dtype)
assert samples.shape[0] == labels.shape[0], (
'samples.shape: %s labels.shape: %s' % (samples.shape, labels.shape))
self._num_examples = samples.shape[0]
# print("self._num_examples", self._num_examples)
if dtype == dtypes.float32:
# Convert from [0, 255] -> [0.0, 1.0].
samples = samples.astype(np.float32)
samples = np.multiply(samples, 1.0 / 255.0)
self._samples = samples
self._labels = labels
self._epochs_completed = 0
self._index_in_epoch = 0
@property
def samples(self):
return self._samples
@property
def labels(self):
return self._labels
@property
def num_examples(self):
return self._num_examples
@property
def epochs_completed(self):
return self._epochs_completed
def next_batch(self, batch_size):
"""Return the next `batch_size` examples from this data set."""
start = self._index_in_epoch
self._index_in_epoch += batch_size
if self._index_in_epoch > self._num_examples:
# Finished epoch
self._epochs_completed += 1
# Shuffle the data
# perm = np.arange(self._num_examples)
# np.random.shuffle(perm)
# sample_shape=self._samples.shape
# self._samples = self._samples[perm]
# label_shape = self._labels.shape
# self._labels = self._labels[perm]
# Start next epoch
start = 0
self._index_in_epoch = batch_size
assert batch_size <= self._num_examples
end = self._index_in_epoch
return self._samples[start:end], self._labels[start:end]