Hugging Face's logo

Spaces:
ppaihack
/
ClairVault
Running on CPU Upgrade

App Files Community
ClairVault
File size: 4,150 Bytes 
6c570a1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
import os
import pandas as pd
import numpy
import pickle
import pefile
import sklearn.ensemble as ek
from sklearn.feature_selection import SelectFromModel
import joblib
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import confusion_matrix
from sklearn import svm
import sklearn.metrics as metrics
from sklearn.model_selection import train_test_split
import pdb

from tqdm import tqdm

dataset = pd.read_csv("data.csv", sep="|")


# Feature
X = dataset.drop(
    ["Name", "md5", "legitimate"], axis=1
).values  # Droping this because classification model will not accept object type elements (float and int only)
# Target variable

ugly = [
    "Machine",
    "SizeOfOptionalHeader",
    "Characteristics",
    "MajorLinkerVersion",
    "MinorLinkerVersion",
    "SizeOfCode",
    "SizeOfInitializedData",
    "SizeOfUninitializedData",
    "AddressOfEntryPoint",
    "BaseOfCode",
    "BaseOfData",
    "ImageBase",
    "SectionAlignment",
    "FileAlignment",
    "MajorOperatingSystemVersion",
    "MinorOperatingSystemVersion",
    "MajorImageVersion",
    "MinorImageVersion",
    "MajorSubsystemVersion",
    "MinorSubsystemVersion",
    "SizeOfImage",
    "SizeOfHeaders",
    "CheckSum",
    "Subsystem",
    "DllCharacteristics",
    "SizeOfStackReserve",
    "SizeOfStackCommit",
    "SizeOfHeapReserve",
    "SizeOfHeapCommit",
    "LoaderFlags",
    "NumberOfRvaAndSizes",
    "SectionsNb",
    "SectionsMeanEntropy",
    "SectionsMinEntropy",
    "SectionsMaxEntropy",
    "SectionsMeanRawsize",
    "SectionsMinRawsize",
    #"SectionsMaxRawsize",
    "SectionsMeanVirtualsize",
    "SectionsMinVirtualsize",
    "SectionMaxVirtualsize",
    "ImportsNbDLL",
    "ImportsNb",
    "ImportsNbOrdinal",
    "ExportNb",
    "ResourcesNb",
    "ResourcesMeanEntropy",
    "ResourcesMinEntropy",
    "ResourcesMaxEntropy",
    "ResourcesMeanSize",
    "ResourcesMinSize",
    "ResourcesMaxSize",
    "LoadConfigurationSize",
    "VersionInformationSize",
]

X = dataset[ugly].values

y = dataset["legitimate"].values


extratrees = ek.ExtraTreesClassifier().fit(X[:1000], y[:1000])
model = SelectFromModel(extratrees, prefit=True)
X_new = model.transform(X)
nbfeatures = X_new.shape[1]


# splitting the data (70% - training and 30% - testing)


X_train, X_test, y_train, y_test = train_test_split(
    X_new, y, test_size=0.29, stratify=y
)


features = []
index = numpy.argsort(extratrees.feature_importances_)[::-1][:nbfeatures]


for f in range(nbfeatures):
    print(
        "%d. feature %s (%f)"
        % (
            f + 1,
            dataset.columns[2 + index[f]],
            extratrees.feature_importances_[index[f]],
        )
    )
    features.append(dataset.columns[2 + f])


model = {
    "DecisionTree": DecisionTreeClassifier(max_depth=10),
    "RandomForest": ek.RandomForestClassifier(n_estimators=50),
}


results = {}
for algo in model:
    clf = model[algo]
    clf.fit(X_train, y_train)
    score = clf.score(X_test, y_test)
    print("%s : %s " % (algo, score))
    results[algo] = score


winner = max(results, key=results.get)  # Selecting the classifier with good result
print("Using", winner, "for classification, with", len(features), "features.")


joblib.dump(model[winner], "classifier.pkl")
open("features.pkl", "wb").write(pickle.dumps(features))


from fhe_utils import (
    client_server_interaction, train_zama,
    setup_network,
    copy_directory,
    setup_client,
)

model_dev_fhe = train_zama(X_train, y_train)
#pdb.set_trace()
network, _ = setup_network(model_dev_fhe)
copied, error_message = copy_directory(network.dev_dir.name, destination="fhe_model")


if not copied:
    print(f"Error copying directory: {error_message}")


network.dev_send_model_to_server()
network.dev_send_clientspecs_and_modelspecs_to_client()

fhemodel_client, serialized_evaluation_keys = setup_client(
    network, network.client_dir.name
)
print(f"Evaluation keys size: {len(serialized_evaluation_keys)} B")

network.client_send_evaluation_key_to_server(serialized_evaluation_keys)



decrypted_predictions, execution_time  = client_server_interaction(network, fhemodel_client, X_test[:100])