File size: 4,067 Bytes
f5ed7d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
342e306
f5ed7d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
342e306
 
 
f5ed7d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
342e306
f5ed7d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
342e306
f5ed7d2
342e306
f5ed7d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
342e306
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import joblib

import os
import shutil

# Define the directory for FHE client/server files
fhe_directory = '/tmp/fhe_client_server_files/'

# Create the directory if it does not exist
if not os.path.exists(fhe_directory):
    os.makedirs(fhe_directory)
else:
    # If it exists, delete its contents
    shutil.rmtree(fhe_directory)
    os.makedirs(fhe_directory)

data=pd.read_csv('data/heart.xls')

data.info()   #checking the info

data_corr=data.corr()

plt.figure(figsize=(20,20))
sns.heatmap(data=data_corr,annot=True)
#Heatmap for data

feature_value=np.array(data_corr['output'])
for i in range(len(feature_value)):
    if feature_value[i]<0:
        feature_value[i]=-feature_value[i]

print(feature_value)

features_corr=pd.DataFrame(feature_value,index=data_corr['output'].index,columns=['correalation'])

feature_sorted=features_corr.sort_values(by=['correalation'],ascending=False)

feature_selected=feature_sorted.index

feature_selected     #selected features which are very much correalated

clean_data=data[feature_selected]

from sklearn.tree import DecisionTreeClassifier                  #using sklearn decisiontreeclassifier
from sklearn.model_selection import train_test_split

#making input and output dataset
X=clean_data.iloc[:,1:]
Y=clean_data['output']

x_train,x_test,y_train,y_test=train_test_split(X,Y,test_size=0.25,random_state=0)

print(x_train.shape,y_train.shape,x_test.shape,y_test.shape)     #data is splited in traing and testing dataset

# feature scaling
from sklearn.preprocessing import StandardScaler
sc=StandardScaler()
x_train=sc.fit_transform(x_train)
x_test=sc.transform(x_test)

#training our model
dt=DecisionTreeClassifier(criterion='entropy',max_depth=6)
dt.fit(x_train,y_train)
#dt.compile(x_trqin)

#predicting the value on testing data
y_pred=dt.predict(x_test)

#ploting the data
from sklearn.metrics import confusion_matrix
conf_mat=confusion_matrix(y_test,y_pred)
print(conf_mat)
accuracy=dt.score(x_test,y_test)
print("\nThe accuracy of decisiontreelassifier on Heart disease prediction dataset is "+str(round(accuracy*100,2))+"%")

joblib.dump(dt, 'heart_disease_dt_model.pkl')

from concrete.ml.sklearn.tree import DecisionTreeClassifier

fhe_compatible = DecisionTreeClassifier.from_sklearn_model(dt, x_train, n_bits = 10)
fhe_compatible.compile(x_train)






#### server
from concrete.ml.deployment import FHEModelDev, FHEModelClient, FHEModelServer

# Setup the development environment
dev = FHEModelDev(path_dir=fhe_directory, model=fhe_compatible)
dev.save()

# Setup the server
server = FHEModelServer(path_dir=fhe_directory)
server.load()







####### client

from concrete.ml.deployment import FHEModelDev, FHEModelClient, FHEModelServer

# Setup the client
client = FHEModelClient(path_dir=fhe_directory, key_dir="/tmp/keys_client")
serialized_evaluation_keys = client.get_serialized_evaluation_keys()


# Load the dataset and select the relevant features
data = pd.read_csv('data/heart.xls')

# Perform the correlation analysis
data_corr = data.corr()

# Select features based on correlation with 'output'
feature_value = np.array(data_corr['output'])
for i in range(len(feature_value)):
    if feature_value[i] < 0:
        feature_value[i] = -feature_value[i]

features_corr = pd.DataFrame(feature_value, index=data_corr['output'].index, columns=['correlation'])
feature_sorted = features_corr.sort_values(by=['correlation'], ascending=False)
feature_selected = feature_sorted.index

# Clean the data by selecting the most correlated features
clean_data = data[feature_selected]

# Extract the first row of feature data for prediction (excluding 'output' column)
sample_data = clean_data.iloc[0, 1:].values.reshape(1, -1)  # Reshape to 2D array for model input

encrypted_data = client.quantize_encrypt_serialize(sample_data)



##### end client

encrypted_result = server.run(encrypted_data, serialized_evaluation_keys)

result = client.deserialize_decrypt_dequantize(encrypted_result)
print(result)