Load_NN_model.py

from __future__ import absolute_import, division, print_function, unicode_literals

import time

import tensorflow as tf
from tensorflow import keras
import numpy as np
from sklearn.model_selection import KFold
import matplotlib.pyplot as plt
import csv
import os


FOLDER='testResult'

start_time = time.time()
# запись векторов разрещений и результата
Y_vector=None
Permissions_vector=None

App_Names=[]
with open(FOLDER+'/permission_dict.csv', newline='') as File:
    reader = csv.reader(File,delimiter=';')
    for row in reader:
        if reader.line_num == 1:
            continue
        else:
            if Y_vector is None:
                if row[1].find('virus') != -1:
                    Y_vector=np.array([1])
                else:
                    Y_vector=np.array([0])
            else:
                if row[1].find('virus') != -1:
                    Y_vector=np.append(Y_vector,[1],axis=0)
                else:
                    Y_vector=np.append(Y_vector,[0],axis=0)
            temp = [int(i) for i in row[2:]]
            if Permissions_vector is None:
                Permissions_vector=np.array([temp])
            else:
                Permissions_vector=np.append(Permissions_vector,[temp],axis=0)



OtherDict_vector=None
with open(FOLDER+'/OtherDict.csv', newline='') as File:
    reader = csv.reader(File,delimiter=';')
    for row in reader:
        if reader.line_num == 1:
            continue
        else:
            temp = [int(i) for i in row[2:]]
            if OtherDict_vector is None:
                OtherDict_vector=np.array([temp])
            else:
                OtherDict_vector=np.append(OtherDict_vector,[temp],axis=0)

API_CALLS_vector=None
with open(FOLDER+'/API_CALLS.csv', newline='') as File:
    reader = csv.reader(File,delimiter=';')
    for row in reader:
        if reader.line_num == 1:
            continue
        else:
            temp = [int(i) for i in row[2:]]

            if API_CALLS_vector is None:
                API_CALLS_vector=np.array([temp])
            else:
                API_CALLS_vector=np.append(API_CALLS_vector,[temp],axis=0)
#API_CALLS_vector=normalize(API_CALLS_vector,axis=1)#нормализация тут- сомниельные результаты
# загрузка нормализации
temp=None
with open('API_CALLS_normalization.npy', 'rb') as f:
    temp = np.load(f)

API_CALLS_vector=API_CALLS_vector/temp

groupAPI_dict_vector=None
with open(FOLDER+'/groupAPI_dict.csv', newline='') as File:
    reader = csv.reader(File,delimiter=';')
    for row in reader:
        if reader.line_num == 1:
            continue
        else:
            temp = [int(i) for i in row[2:]]

            if groupAPI_dict_vector is None:
                groupAPI_dict_vector=np.array([temp])
            else:
                groupAPI_dict_vector=np.append(groupAPI_dict_vector,[temp],axis=0)
#groupAPI_dict_vector=normalize(groupAPI_dict_vector,axis=1)
# загрузка нормализации
temp=None
with open('groupAPI_dict_normalization.npy', 'rb') as f:
    temp = np.load(f)

groupAPI_dict_vector=groupAPI_dict_vector/temp

strings_dict_vector=None
with open(FOLDER+'/strings_dict.csv', newline='') as File:
    reader = csv.reader(File,delimiter=';')
    for row in reader:
        if reader.line_num == 1:
            continue
        else:
            temp = [int(i) for i in row[2:]]

            if strings_dict_vector is None:
                strings_dict_vector=np.array([temp])
            else:
                strings_dict_vector=np.append(strings_dict_vector,[temp],axis=0)
# загрузка нормализации
temp=None
with open('string_normalization.npy', 'rb') as f:
    temp = np.load(f)

strings_dict_vector=strings_dict_vector/temp

result_vector=np.append(Permissions_vector,OtherDict_vector,axis=1)
result_vector=np.append(result_vector,API_CALLS_vector,axis=1)
result_vector=np.append(result_vector,groupAPI_dict_vector,axis=1)
result_vector=np.append(result_vector,strings_dict_vector,axis=1)

print("Data loaded:{} ".format(time.time() - start_time))
# Восстановим в точности ту же модель, включая веса и оптимизатор
new_model = keras.models.load_model('my_model.h5')

# Покажем архитектуру модели
new_model.summary()
# Evaluate the restored model

predictions = new_model.predict(result_vector)
#for i in predictions:
#    print(f"Appp{}predictions)
print(predictions)
print("All DONE:{} ".format(time.time() - start_time))