├── README.md
└── dga
    ├── __pycache__
        ├── common.cpython-36.pyc
        └── load.cpython-36.pyc
    ├── common.py
    ├── common.pyc
    ├── load.py
    ├── load.pyc
    ├── lstm.py
    ├── main.py
    ├── main.pyc
    ├── svm.ipynb
    ├── train.txt
    └── word2vec.txt


/README.md:
--------------------------------------------------------------------------------
 1 | # dga_check
 2 | 
 3 | 基于机器学习识别DGA域名
 4 | 
 5 | 包括SVM , Word2Vec + LSTM两种方法实现
 6 | 
 7 | acc:
 8 | 
 9 | SVM~90%    
10 | 
11 | LSTM~99%
12 | 
13 | 
14 | 依赖：
15 | 
16 | python3
17 | 
18 | numpy
19 | 
20 | pandas
21 | 
22 | matplotlib
23 | 
24 | sklearn
25 | 
26 | gensim
27 | 
28 | tensorflow
29 | 
30 | keras
31 | 
32 | 数据集来源：
33 | 
34 | https://github.com/phunterlau/dga_classifier
35 | 


--------------------------------------------------------------------------------
/dga/__pycache__/common.cpython-36.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/0FuzzingQ/dga_check/48dcee824187d4e8bcdef537f31926de8469f7ca/dga/__pycache__/common.cpython-36.pyc


--------------------------------------------------------------------------------
/dga/__pycache__/load.cpython-36.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/0FuzzingQ/dga_check/48dcee824187d4e8bcdef537f31926de8469f7ca/dga/__pycache__/load.cpython-36.pyc


--------------------------------------------------------------------------------
/dga/common.py:
--------------------------------------------------------------------------------
 1 | import numpy as np
 2 | import pandas as pd
 3 | import math
 4 | from hmmlearn import hmm 
 5 | from gensim.models import Word2Vec
 6 | 
 7 | def getshan(domain):
 8 | 	tmp_dict = {}
 9 | 	domain_len = len(domain)
10 | 	for i in range(0,domain_len):
11 | 		if domain[i] in tmp_dict.keys():
12 | 			tmp_dict[domain[i]] = tmp_dict[domain[i]] + 1
13 | 		else:
14 | 			tmp_dict[domain[i]] = 1
15 | 	shannon = 0
16 | 	for i in tmp_dict.keys():
17 | 		p = float(tmp_dict[i]) / domain_len
18 | 		shannon = shannon - p * math.log(p,2)
19 | 	return shannon
20 | 
21 | def getyuanyin(domain):
22 | 	yuan_list = ['a','e','i','o','u']
23 | 	domain = domain.lower()
24 | 	count_word = 0
25 | 	count_yuan = 0
26 | 	yuan_ratio = 0
27 | 	for i in range(0,len(domain)):
28 | 		if ord(domain[i]) >= ord('a') and ord(domain[i]) <= ord('z'):
29 | 			count_word = count_word + 1
30 | 			if domain[i] in yuan_list:
31 | 				count_yuan = count_yuan + 1
32 | 	if count_word == 0:
33 | 		return yuan_ratio
34 | 	else:
35 | 		yuan_ratio = float(count_yuan) / count_word
36 | 		return yuan_ratio
37 | 
38 | def getroot(domain):
39 | 	return domain.split('.')[-1]
40 | 
41 | def getrootclass(root):
42 | 	common_root = ['cn','com','cc','net','org','gov','info']
43 | 	if root in common_root:
44 | 		return 0
45 | 	else:
46 | 		return 1
47 | 
48 | def getlen(domain):
49 | 	return len(domain)
50 | 
51 | def getw2v(domain_list,label_list):
52 | 	stop = ['/','~']
53 | 	w2v_list = []
54 | 	for i in range(0,domain_list.size):
55 | 		tmp = []
56 | 		name = domain_list[i].split('.')[0]
57 | 		for j in range(0,len(name)):
58 | 			tmp.append(name[j])
59 | 		w2v_list.append(tmp)
60 | 
61 | 	model = Word2Vec(w2v_list, min_count = 1)
62 | 	#print (model._vocabulary)
63 | 	model.wv.save_word2vec_format('word2vec.txt',binary=False) 
64 | 	#print model['a']
65 | 	label_vect = []
66 | 	wv_vect = []
67 | 	for i in range(0,domain_list.size):
68 | 		name = domain_list[i].replace('.','')
69 | 		tmp = []
70 | 		vect = []
71 | 		for j in range(0,len(name)):
72 | 			if name[j] in stop:
73 | 				continue
74 | 			tmp.append(model[name[j]])
75 | 			if j >= 9:
76 | 				break
77 | 		if len(tmp) < 10:
78 | 			for k in range(0,10-len(tmp)):
79 | 				tmp.append([0]*100)
80 | 		vect = np.vstack((x for x in tmp))
81 | 		wv_vect.append(vect)
82 | 		label_vect.append(label_list[i])
83 | 		if i ==100000:
84 | 			break
85 | 	wv_vect = np.array(wv_vect)
86 | 	label_vect = np.array(label_vect)
87 | 	return wv_vect,label_vect
88 | 		
89 | 
90 | 		
91 | 
92 | 


--------------------------------------------------------------------------------
/dga/common.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/0FuzzingQ/dga_check/48dcee824187d4e8bcdef537f31926de8469f7ca/dga/common.pyc


--------------------------------------------------------------------------------
/dga/load.py:
--------------------------------------------------------------------------------
 1 | # -*- coding: utf-8 -*-
 2 | 
 3 | import sys
 4 | import os
 5 | import numpy as np
 6 | import pandas as pd
 7 | import matplotlib.pyplot as plt
 8 | import sklearn
 9 | 
10 | def load_data(path,method):
11 | 	if method == 'train':
12 | 		url_list = []
13 | 		label_list = []
14 | 		
15 | 		with open(path,'r') as f:
16 | 			content = f.readlines()
17 | 			for i in range(0,len(content)):
18 | 				tmp = content[i].strip('\n').lower()
19 | 				#print tmp.split(' ')
20 | 				url_list.append(tmp.split('\t')[0].strip()[0:-1])
21 | 				label_list.append(tmp.split('\t')[1].strip())
22 | 		f.close()
23 | 		
24 | 		#print 'can not read file'
25 | 		
26 | 		tmp = np.vstack((label_list,url_list))
27 | 		train_data = pd.DataFrame(tmp.T,columns = ['label','domain'])
28 | 		#print train_data.info()
29 | 		#print train_data.head()
30 | 		return train_data
31 | 
32 | 	elif method == 'test':
33 | 		exit()
34 | 
35 | #load_data('train.txt','train')
36 | 


--------------------------------------------------------------------------------
/dga/load.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/0FuzzingQ/dga_check/48dcee824187d4e8bcdef537f31926de8469f7ca/dga/load.pyc


--------------------------------------------------------------------------------
/dga/lstm.py:
--------------------------------------------------------------------------------
 1 | import sys
 2 | import os
 3 | import numpy as np
 4 | import pandas as pd
 5 | import matplotlib.pyplot as plt
 6 | import sklearn
 7 | import argparse
 8 | from load import load_data
 9 | from common import getshan,getyuanyin,getroot,getrootclass,getlen,getw2v
10 | from sklearn.feature_extraction.text import CountVectorizer
11 | import re
12 | from sklearn import preprocessing
13 | from sklearn.utils import shuffle
14 | from sklearn.svm import SVC
15 | from sklearn import cross_validation
16 | from keras.preprocessing import sequence
17 | from keras.models import Sequential
18 | from keras.layers import Dense,Embedding
19 | from keras.layers import LSTM
20 | 
21 | 
22 | method = 'train'
23 | path = 'train.txt'
24 | 
25 | if 'train' in method:
26 | 	train_data = load_data(path,method)
27 | 	train_data = shuffle(train_data)
28 | 	w2v_word_list,label_list = getw2v(train_data['domain'],train_data['label'])
29 | 	#print (type(w2v_word_list))
30 | 	#exit()
31 | 	x_train = w2v_word_list[0:80000]
32 | 	y_train = label_list[0:80000]
33 | 	x_test = w2v_word_list[80000:]
34 | 	y_test = label_list[80000:]
35 | 	model = Sequential()
36 | 	#model.add(Embedding())
37 | 	model.add(LSTM(128,dropout = 0.2,recurrent_dropout = 0.2))
38 | 	model.add(Dense(1,activation='sigmoid'))
39 | 
40 | 	model.compile(loss = 'binary_crossentropy',optimizer = 'adam',metrics = ['accuracy'])
41 | 	print ('now training....')
42 | 	model.fit(x_train,y_train,nb_epoch = 50,batch_size = 32)
43 | 	print ('now evaling....')
44 | 	score,acc = model.evaluate(x_test,y_test)
45 | 	print (score,acc)
46 | 
47 | 


--------------------------------------------------------------------------------
/dga/main.py:
--------------------------------------------------------------------------------
 1 | 
 2 | import sys
 3 | import os
 4 | import numpy as np
 5 | import pandas as pd
 6 | import matplotlib.pyplot as plt
 7 | import sklearn
 8 | import argparse
 9 | from load import load_data
10 | from common import getshan,getyuanyin,getroot,getrootclass,getlen,getw2v
11 | from sklearn.feature_extraction.text import CountVectorizer
12 | import re
13 | from sklearn import preprocessing
14 | from sklearn.utils import shuffle
15 | from sklearn import cross_vali
16 | from sklearn.svm import SVCdation
17 | 
18 | 
19 | method = 'train'
20 | path = 'train.txt'
21 | 
22 | if 'train' in method:
23 | 	train_data = load_data(path,method)
24 | 	train_data['shan'] = train_data['domain'].map(lambda  x:getshan(x)).astype(float)
25 | 	train_data['yuan_ratio'] = train_data['domain'].map(lambda x:getyuanyin(x)).astype(float)
26 | 	train_data['root'] = train_data['domain'].map(lambda x:getroot(x)).astype(str)
27 | 	train_data['rootclass'] = train_data['root'].map(lambda x:getrootclass(x)).astype(int)
28 | 	train_data['len'] = train_data['domain'].map(lambda x:getlen(x)).astype(int)
29 | 
30 | 	scaler = preprocessing.StandardScaler()
31 | 	len_scale_param = scaler.fit(train_data['len'].values.reshape(-1,1))
32 | 	train_data['len_scaled'] = scaler.fit_transform(train_data['len'].values.reshape(-1,1),len_scale_param)
33 | 	shan_scale_param = scaler.fit(train_data['shan'].values.reshape(-1,1))
34 | 	train_data['shan_sclaed'] = scaler.fit_transform(train_data['shan'].values.reshape(-1,1),shan_scale_param)
35 | 
36 | 
37 | 	train_pre = train_data.filter(regex = 'label|yuan_ratio|len_scaled|shan_scale_param|rootclass')
38 | 	train_pre = shuffle(train_pre)
39 | 	print (train_data.info())
40 | 	print (train_data.head(10))
41 | 	print (train_data.tail(10))
42 | 	train_pre =train_pre.as_matrix()
43 | 	y_train = train_pre[0:200000,0]
44 | 	x_train = train_pre[0:200000,1:]
45 | 	y_test = train_pre[200000:201000,0]
46 | 	x_test = train_pre[200000:201000,1:]
47 | 	print ('now training')
48 | 	lr = SVC(kernel='rbf',C=0.4).fit(x_train,y_train)
49 | 	print ('training finished')
50 | 	model = cross_validation.cross_val_score(lr,x_train,y_train,cv = 5)
51 | 	print (model)
52 | 
53 | 
54 | 
55 | 
56 | 
57 | 
58 | 
59 | 


--------------------------------------------------------------------------------
/dga/main.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/0FuzzingQ/dga_check/48dcee824187d4e8bcdef537f31926de8469f7ca/dga/main.pyc


--------------------------------------------------------------------------------
/dga/svm.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {
   6 |     "slideshow": {
   7 |      "slide_type": "slide"
   8 |     }
   9 |    },
  10 |    "source": [
  11 |     "想要了解该内容的看客们一定对DGA相关内容有一定了解了，这里就不再赘述，只是做简单说明。\n",
  12 |     "\n",
  13 |     "僵尸网络中用来控制主机，负责处理信息，下发任务的中心机器，也称C&C服务器（control & command）。一个成熟的僵尸网络往往具有多个C&C服务 器。\n",
  14 |     "对于安全人员来说，查杀了僵尸网络的C&c服务器也就等于毁灭而这个僵尸网络，那么作为安全对抗，僵尸网络的缔造者也会想尽办法对其进行隐藏和保护其主控端。\n",
  15 |     "\n",
  16 |     "那么对于固定的C&C服务器域名，安全人员一般来说很容易对其进行查杀，因此，基于DGA方法僵尸网络也就产生了，DGA全程随机域名生成算法，是指使用主控端和被控端协商好的一种基于随机算法的域名生成协议，简单来说就是生成一个随机字符串来作为域名并进行注册，将其作为C&C服务器的域名并不定时经常性更换。由于具备强随机性，短时效性，通过DGA生成的域名往往在查杀上更具备难度。\n",
  17 |     "\n",
  18 |     "使用机器学习的方法来对DGA域名进行查杀，业界已有实现；笔者首次接触实在2016年wooyun年会上，phunter神做的一篇报告，在报告中他详细介绍了包括二层，三层flux方法等在内的诸多高级生成DGA方法，并从数学上介绍了一些生成算法以及机器学习的检测方法.本文是参考https://github.com/phunterlau/dga_classifier 的一个简单实现版本，本文全部代码可以在 笔者GitHub项目中查看https://github.com/0FuzzingQ/dga_check\n"
  19 |    ]
  20 |   },
  21 |   {
  22 |    "cell_type": "code",
  23 |    "execution_count": 1,
  24 |    "metadata": {},
  25 |    "outputs": [
  26 |     {
  27 |      "name": "stderr",
  28 |      "output_type": "stream",
  29 |      "text": [
  30 |       "C:\\Users\\aldin\\AppData\\Roaming\\Python\\Python36\\site-packages\\gensim\\utils.py:1212: UserWarning: detected Windows; aliasing chunkize to chunkize_serial\n",
  31 |       "  warnings.warn(\"detected Windows; aliasing chunkize to chunkize_serial\")\n",
  32 |       "c:\\program files\\python36\\lib\\site-packages\\sklearn\\cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
  33 |       "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
  34 |      ]
  35 |     }
  36 |    ],
  37 |    "source": [
  38 |     "import sys\n",
  39 |     "import os\n",
  40 |     "import numpy as np\n",
  41 |     "import pandas as pd\n",
  42 |     "import matplotlib.pyplot as plt\n",
  43 |     "import sklearn\n",
  44 |     "import argparse\n",
  45 |     "from load import load_data\n",
  46 |     "from common import getshan,getyuanyin,getroot,getrootclass,getlen,getw2v\n",
  47 |     "from sklearn.feature_extraction.text import CountVectorizer\n",
  48 |     "import re\n",
  49 |     "from sklearn import preprocessing\n",
  50 |     "from sklearn.utils import shuffle\n",
  51 |     "from sklearn import cross_validation\n",
  52 |     "from sklearn.svm import SVC\n",
  53 |     "import seaborn as sns\n",
  54 |     "plt.style.use('seaborn')\n",
  55 |     "sns.set(font_scale=2)\n",
  56 |     "pd.set_option('display.max_columns', 500)"
  57 |    ]
  58 |   },
  59 |   {
  60 |    "cell_type": "code",
  61 |    "execution_count": 2,
  62 |    "metadata": {},
  63 |    "outputs": [],
  64 |    "source": [
  65 |     "method = 'train'\n",
  66 |     "path = 'train.txt'"
  67 |    ]
  68 |   },
  69 |   {
  70 |    "cell_type": "markdown",
  71 |    "metadata": {},
  72 |    "source": [
  73 |     "本次使用训练集位于train.txt中，共包含10w正常url数据以及15w 使用DGA算法生成的域名数据\n",
  74 |     "首先载入数据集看一下数据集情况"
  75 |    ]
  76 |   },
  77 |   {
  78 |    "cell_type": "code",
  79 |    "execution_count": 3,
  80 |    "metadata": {},
  81 |    "outputs": [],
  82 |    "source": [
  83 |     "train_data = load_data(path,method)"
  84 |    ]
  85 |   },
  86 |   {
  87 |    "cell_type": "code",
  88 |    "execution_count": 4,
  89 |    "metadata": {},
  90 |    "outputs": [
  91 |     {
  92 |      "name": "stdout",
  93 |      "output_type": "stream",
  94 |      "text": [
  95 |       "<class 'pandas.core.frame.DataFrame'>\n",
  96 |       "RangeIndex: 249999 entries, 0 to 249998\n",
  97 |       "Data columns (total 2 columns):\n",
  98 |       "label     249999 non-null object\n",
  99 |       "domain    249999 non-null object\n",
 100 |       "dtypes: object(2)\n",
 101 |       "memory usage: 3.8+ MB\n"
 102 |      ]
 103 |     }
 104 |    ],
 105 |    "source": [
 106 |     "train_data.info()"
 107 |    ]
 108 |   },
 109 |   {
 110 |    "cell_type": "code",
 111 |    "execution_count": 5,
 112 |    "metadata": {},
 113 |    "outputs": [
 114 |     {
 115 |      "data": {
 116 |       "text/html": [
 117 |        "<div>\n",
 118 |        "<style scoped>\n",
 119 |        "    .dataframe tbody tr th:only-of-type {\n",
 120 |        "        vertical-align: middle;\n",
 121 |        "    }\n",
 122 |        "\n",
 123 |        "    .dataframe tbody tr th {\n",
 124 |        "        vertical-align: top;\n",
 125 |        "    }\n",
 126 |        "\n",
 127 |        "    .dataframe thead th {\n",
 128 |        "        text-align: right;\n",
 129 |        "    }\n",
 130 |        "</style>\n",
 131 |        "<table border=\"1\" class=\"dataframe\">\n",
 132 |        "  <thead>\n",
 133 |        "    <tr style=\"text-align: right;\">\n",
 134 |        "      <th></th>\n",
 135 |        "      <th>label</th>\n",
 136 |        "      <th>domain</th>\n",
 137 |        "    </tr>\n",
 138 |        "  </thead>\n",
 139 |        "  <tbody>\n",
 140 |        "    <tr>\n",
 141 |        "      <th>0</th>\n",
 142 |        "      <td>0</td>\n",
 143 |        "      <td>facebook.com</td>\n",
 144 |        "    </tr>\n",
 145 |        "    <tr>\n",
 146 |        "      <th>1</th>\n",
 147 |        "      <td>0</td>\n",
 148 |        "      <td>youtube.com</td>\n",
 149 |        "    </tr>\n",
 150 |        "    <tr>\n",
 151 |        "      <th>2</th>\n",
 152 |        "      <td>0</td>\n",
 153 |        "      <td>yahoo.com</td>\n",
 154 |        "    </tr>\n",
 155 |        "    <tr>\n",
 156 |        "      <th>3</th>\n",
 157 |        "      <td>0</td>\n",
 158 |        "      <td>baidu.com</td>\n",
 159 |        "    </tr>\n",
 160 |        "    <tr>\n",
 161 |        "      <th>4</th>\n",
 162 |        "      <td>0</td>\n",
 163 |        "      <td>wikipedia.org</td>\n",
 164 |        "    </tr>\n",
 165 |        "  </tbody>\n",
 166 |        "</table>\n",
 167 |        "</div>"
 168 |       ],
 169 |       "text/plain": [
 170 |        "  label         domain\n",
 171 |        "0     0   facebook.com\n",
 172 |        "1     0    youtube.com\n",
 173 |        "2     0      yahoo.com\n",
 174 |        "3     0      baidu.com\n",
 175 |        "4     0  wikipedia.org"
 176 |       ]
 177 |      },
 178 |      "execution_count": 5,
 179 |      "metadata": {},
 180 |      "output_type": "execute_result"
 181 |     }
 182 |    ],
 183 |    "source": [
 184 |     "train_data.head(5)"
 185 |    ]
 186 |   },
 187 |   {
 188 |    "cell_type": "code",
 189 |    "execution_count": 6,
 190 |    "metadata": {},
 191 |    "outputs": [
 192 |     {
 193 |      "data": {
 194 |       "text/html": [
 195 |        "<div>\n",
 196 |        "<style scoped>\n",
 197 |        "    .dataframe tbody tr th:only-of-type {\n",
 198 |        "        vertical-align: middle;\n",
 199 |        "    }\n",
 200 |        "\n",
 201 |        "    .dataframe tbody tr th {\n",
 202 |        "        vertical-align: top;\n",
 203 |        "    }\n",
 204 |        "\n",
 205 |        "    .dataframe thead th {\n",
 206 |        "        text-align: right;\n",
 207 |        "    }\n",
 208 |        "</style>\n",
 209 |        "<table border=\"1\" class=\"dataframe\">\n",
 210 |        "  <thead>\n",
 211 |        "    <tr style=\"text-align: right;\">\n",
 212 |        "      <th></th>\n",
 213 |        "      <th>label</th>\n",
 214 |        "      <th>domain</th>\n",
 215 |        "    </tr>\n",
 216 |        "  </thead>\n",
 217 |        "  <tbody>\n",
 218 |        "    <tr>\n",
 219 |        "      <th>249994</th>\n",
 220 |        "      <td>1</td>\n",
 221 |        "      <td>zzznnx.ir</td>\n",
 222 |        "    </tr>\n",
 223 |        "    <tr>\n",
 224 |        "      <th>249995</th>\n",
 225 |        "      <td>1</td>\n",
 226 |        "      <td>zzztqrq.ht</td>\n",
 227 |        "    </tr>\n",
 228 |        "    <tr>\n",
 229 |        "      <th>249996</th>\n",
 230 |        "      <td>1</td>\n",
 231 |        "      <td>zzzvr.su</td>\n",
 232 |        "    </tr>\n",
 233 |        "    <tr>\n",
 234 |        "      <th>249997</th>\n",
 235 |        "      <td>1</td>\n",
 236 |        "      <td>zzzwtef.com.pa</td>\n",
 237 |        "    </tr>\n",
 238 |        "    <tr>\n",
 239 |        "      <th>249998</th>\n",
 240 |        "      <td>1</td>\n",
 241 |        "      <td>zzzzuvfbt.co.uk</td>\n",
 242 |        "    </tr>\n",
 243 |        "  </tbody>\n",
 244 |        "</table>\n",
 245 |        "</div>"
 246 |       ],
 247 |       "text/plain": [
 248 |        "       label           domain\n",
 249 |        "249994     1        zzznnx.ir\n",
 250 |        "249995     1       zzztqrq.ht\n",
 251 |        "249996     1         zzzvr.su\n",
 252 |        "249997     1   zzzwtef.com.pa\n",
 253 |        "249998     1  zzzzuvfbt.co.uk"
 254 |       ]
 255 |      },
 256 |      "execution_count": 6,
 257 |      "metadata": {},
 258 |      "output_type": "execute_result"
 259 |     }
 260 |    ],
 261 |    "source": [
 262 |     "train_data.tail(5)"
 263 |    ]
 264 |   },
 265 |   {
 266 |    "cell_type": "markdown",
 267 |    "metadata": {},
 268 |    "source": [
 269 |     "可以看到正常域名都是大家比较熟悉的那些，Google，youtube等等,恶意域名，感觉就很奇怪，而且...根域名更奇怪，根域名，也许可以看成是一个特征向量也说不定"
 270 |    ]
 271 |   },
 272 |   {
 273 |    "cell_type": "markdown",
 274 |    "metadata": {},
 275 |    "source": [
 276 |     "那么就先把根域名尝试作为一个特征吧，探测一下！周所周知，正大光明的网站一般使用正大光明的域名，比如说很常见的如com;cn;org;cc；net等等，因此我们把这个作为一个点来分类；由于是一个简单的项目，我们不是用one-hot编码,频率编码等手段，仅仅依照经验将其分为常见，不常见两个类别\n",
 277 |     "\n",
 278 |     "注：本文的功能函数代码均位于github中common.py文件中，本notebook仅仅演示主要的机器学习的使用流程和手法\n",
 279 |     "\n",
 280 |     "以下第一行代码用于获取每一个域名的根域名并加入DataFrame中\n",
 281 |     "第二行代码用于分类并将类别信息加入DataFrame中"
 282 |    ]
 283 |   },
 284 |   {
 285 |    "cell_type": "code",
 286 |    "execution_count": 7,
 287 |    "metadata": {},
 288 |    "outputs": [],
 289 |    "source": [
 290 |     "train_data['root'] = train_data['domain'].map(lambda x:getroot(x)).astype(str)\n",
 291 |     "train_data['rootclass'] = train_data['root'].map(lambda x:getrootclass(x)).astype(int)"
 292 |    ]
 293 |   },
 294 |   {
 295 |    "cell_type": "markdown",
 296 |    "metadata": {},
 297 |    "source": [
 298 |     "再看一下数据情况，会发现咱们原本的DataFrame多了两列信息，就是我们刚才加上去的~"
 299 |    ]
 300 |   },
 301 |   {
 302 |    "cell_type": "code",
 303 |    "execution_count": 8,
 304 |    "metadata": {},
 305 |    "outputs": [
 306 |     {
 307 |      "data": {
 308 |       "text/html": [
 309 |        "<div>\n",
 310 |        "<style scoped>\n",
 311 |        "    .dataframe tbody tr th:only-of-type {\n",
 312 |        "        vertical-align: middle;\n",
 313 |        "    }\n",
 314 |        "\n",
 315 |        "    .dataframe tbody tr th {\n",
 316 |        "        vertical-align: top;\n",
 317 |        "    }\n",
 318 |        "\n",
 319 |        "    .dataframe thead th {\n",
 320 |        "        text-align: right;\n",
 321 |        "    }\n",
 322 |        "</style>\n",
 323 |        "<table border=\"1\" class=\"dataframe\">\n",
 324 |        "  <thead>\n",
 325 |        "    <tr style=\"text-align: right;\">\n",
 326 |        "      <th></th>\n",
 327 |        "      <th>label</th>\n",
 328 |        "      <th>domain</th>\n",
 329 |        "      <th>root</th>\n",
 330 |        "      <th>rootclass</th>\n",
 331 |        "    </tr>\n",
 332 |        "  </thead>\n",
 333 |        "  <tbody>\n",
 334 |        "    <tr>\n",
 335 |        "      <th>0</th>\n",
 336 |        "      <td>0</td>\n",
 337 |        "      <td>facebook.com</td>\n",
 338 |        "      <td>com</td>\n",
 339 |        "      <td>0</td>\n",
 340 |        "    </tr>\n",
 341 |        "    <tr>\n",
 342 |        "      <th>1</th>\n",
 343 |        "      <td>0</td>\n",
 344 |        "      <td>youtube.com</td>\n",
 345 |        "      <td>com</td>\n",
 346 |        "      <td>0</td>\n",
 347 |        "    </tr>\n",
 348 |        "    <tr>\n",
 349 |        "      <th>2</th>\n",
 350 |        "      <td>0</td>\n",
 351 |        "      <td>yahoo.com</td>\n",
 352 |        "      <td>com</td>\n",
 353 |        "      <td>0</td>\n",
 354 |        "    </tr>\n",
 355 |        "    <tr>\n",
 356 |        "      <th>3</th>\n",
 357 |        "      <td>0</td>\n",
 358 |        "      <td>baidu.com</td>\n",
 359 |        "      <td>com</td>\n",
 360 |        "      <td>0</td>\n",
 361 |        "    </tr>\n",
 362 |        "    <tr>\n",
 363 |        "      <th>4</th>\n",
 364 |        "      <td>0</td>\n",
 365 |        "      <td>wikipedia.org</td>\n",
 366 |        "      <td>org</td>\n",
 367 |        "      <td>0</td>\n",
 368 |        "    </tr>\n",
 369 |        "  </tbody>\n",
 370 |        "</table>\n",
 371 |        "</div>"
 372 |       ],
 373 |       "text/plain": [
 374 |        "  label         domain root  rootclass\n",
 375 |        "0     0   facebook.com  com          0\n",
 376 |        "1     0    youtube.com  com          0\n",
 377 |        "2     0      yahoo.com  com          0\n",
 378 |        "3     0      baidu.com  com          0\n",
 379 |        "4     0  wikipedia.org  org          0"
 380 |       ]
 381 |      },
 382 |      "execution_count": 8,
 383 |      "metadata": {},
 384 |      "output_type": "execute_result"
 385 |     }
 386 |    ],
 387 |    "source": [
 388 |     "train_data.head(5)"
 389 |    ]
 390 |   },
 391 |   {
 392 |    "cell_type": "code",
 393 |    "execution_count": 9,
 394 |    "metadata": {},
 395 |    "outputs": [
 396 |     {
 397 |      "data": {
 398 |       "text/html": [
 399 |        "<div>\n",
 400 |        "<style scoped>\n",
 401 |        "    .dataframe tbody tr th:only-of-type {\n",
 402 |        "        vertical-align: middle;\n",
 403 |        "    }\n",
 404 |        "\n",
 405 |        "    .dataframe tbody tr th {\n",
 406 |        "        vertical-align: top;\n",
 407 |        "    }\n",
 408 |        "\n",
 409 |        "    .dataframe thead th {\n",
 410 |        "        text-align: right;\n",
 411 |        "    }\n",
 412 |        "</style>\n",
 413 |        "<table border=\"1\" class=\"dataframe\">\n",
 414 |        "  <thead>\n",
 415 |        "    <tr style=\"text-align: right;\">\n",
 416 |        "      <th></th>\n",
 417 |        "      <th>label</th>\n",
 418 |        "      <th>domain</th>\n",
 419 |        "      <th>root</th>\n",
 420 |        "      <th>rootclass</th>\n",
 421 |        "    </tr>\n",
 422 |        "  </thead>\n",
 423 |        "  <tbody>\n",
 424 |        "    <tr>\n",
 425 |        "      <th>249994</th>\n",
 426 |        "      <td>1</td>\n",
 427 |        "      <td>zzznnx.ir</td>\n",
 428 |        "      <td>ir</td>\n",
 429 |        "      <td>1</td>\n",
 430 |        "    </tr>\n",
 431 |        "    <tr>\n",
 432 |        "      <th>249995</th>\n",
 433 |        "      <td>1</td>\n",
 434 |        "      <td>zzztqrq.ht</td>\n",
 435 |        "      <td>ht</td>\n",
 436 |        "      <td>1</td>\n",
 437 |        "    </tr>\n",
 438 |        "    <tr>\n",
 439 |        "      <th>249996</th>\n",
 440 |        "      <td>1</td>\n",
 441 |        "      <td>zzzvr.su</td>\n",
 442 |        "      <td>su</td>\n",
 443 |        "      <td>1</td>\n",
 444 |        "    </tr>\n",
 445 |        "    <tr>\n",
 446 |        "      <th>249997</th>\n",
 447 |        "      <td>1</td>\n",
 448 |        "      <td>zzzwtef.com.pa</td>\n",
 449 |        "      <td>pa</td>\n",
 450 |        "      <td>1</td>\n",
 451 |        "    </tr>\n",
 452 |        "    <tr>\n",
 453 |        "      <th>249998</th>\n",
 454 |        "      <td>1</td>\n",
 455 |        "      <td>zzzzuvfbt.co.uk</td>\n",
 456 |        "      <td>uk</td>\n",
 457 |        "      <td>1</td>\n",
 458 |        "    </tr>\n",
 459 |        "  </tbody>\n",
 460 |        "</table>\n",
 461 |        "</div>"
 462 |       ],
 463 |       "text/plain": [
 464 |        "       label           domain root  rootclass\n",
 465 |        "249994     1        zzznnx.ir   ir          1\n",
 466 |        "249995     1       zzztqrq.ht   ht          1\n",
 467 |        "249996     1         zzzvr.su   su          1\n",
 468 |        "249997     1   zzzwtef.com.pa   pa          1\n",
 469 |        "249998     1  zzzzuvfbt.co.uk   uk          1"
 470 |       ]
 471 |      },
 472 |      "execution_count": 9,
 473 |      "metadata": {},
 474 |      "output_type": "execute_result"
 475 |     }
 476 |    ],
 477 |    "source": [
 478 |     "train_data.tail(5)"
 479 |    ]
 480 |   },
 481 |   {
 482 |    "cell_type": "markdown",
 483 |    "metadata": {},
 484 |    "source": [
 485 |     "有没有发现看上去好像正常域名的rootclass都等于0，不正常的都等于1；\n",
 486 |     "感觉好像是的！那么我们来画个图看一看再整个数据集上的分布吧\n",
 487 |     "\n",
 488 |     "python科学计算可视化神器 matplotlib和seaborn"
 489 |    ]
 490 |   },
 491 |   {
 492 |    "cell_type": "code",
 493 |    "execution_count": 10,
 494 |    "metadata": {},
 495 |    "outputs": [],
 496 |    "source": [
 497 |     "suby = train_data.label.values\n",
 498 |     "subx = train_data.rootclass.values"
 499 |    ]
 500 |   },
 501 |   {
 502 |    "cell_type": "code",
 503 |    "execution_count": 11,
 504 |    "metadata": {},
 505 |    "outputs": [
 506 |     {
 507 |      "data": {
 508 |       "text/plain": [
 509 |        "<matplotlib.axes._subplots.AxesSubplot at 0x2dc542b0cf8>"
 510 |       ]
 511 |      },
 512 |      "execution_count": 11,
 513 |      "metadata": {},
 514 |      "output_type": "execute_result"
 515 |     },
 516 |     {
 517 |      "data": {
 518 |       "image/png": "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\n",
 519 |       "text/plain": [
 520 |        "<Figure size 432x288 with 1 Axes>"
 521 |       ]
 522 |      },
 523 |      "metadata": {
 524 |       "needs_background": "light"
 525 |      },
 526 |      "output_type": "display_data"
 527 |     }
 528 |    ],
 529 |    "source": [
 530 |     "sns.countplot(x='rootclass', hue='label',data=train_data)"
 531 |    ]
 532 |   },
 533 |   {
 534 |    "cell_type": "markdown",
 535 |    "metadata": {},
 536 |    "source": [
 537 |     "ok 很明显可以看出来数据分布上的差异了，在正常样本上也就是label=0时，常见根域名出现的概率远远大于非正常域名，几乎可以说是忽略不计，而对于不正常域名label=1时，非常见根域名出现的概率大概是正常域名的4-5倍，\n",
 538 |     "\n",
 539 |     "所以我们完全可以将根域名作为我们模型的一个特征\n",
 540 |     "PS：笔者采用的这种方法是较为简陋的一种，更科学地特征工程可以更好的挖掘这一特征"
 541 |    ]
 542 |   },
 543 |   {
 544 |    "cell_type": "markdown",
 545 |    "metadata": {},
 546 |    "source": [
 547 |     "接下来挖掘一些其他可能的特征，比如说最简单的，长度？\n",
 548 |     "\n",
 549 |     "试试看，还是使用一样的思路，将每一条域名的长度计算出来来，作为一列加入到我们的train_data中，并采用可视化的方法来看一看域名的长度是否在不同label上有着不同的分布\n",
 550 |     "\n",
 551 |     "以下代码用于计算每个域名的长度并返回结果，作为train_data中的一列"
 552 |    ]
 553 |   },
 554 |   {
 555 |    "cell_type": "code",
 556 |    "execution_count": 12,
 557 |    "metadata": {},
 558 |    "outputs": [],
 559 |    "source": [
 560 |     "train_data['len'] = train_data['domain'].map(lambda x:getlen(x)).astype(int)"
 561 |    ]
 562 |   },
 563 |   {
 564 |    "cell_type": "code",
 565 |    "execution_count": 13,
 566 |    "metadata": {},
 567 |    "outputs": [
 568 |     {
 569 |      "data": {
 570 |       "text/html": [
 571 |        "<div>\n",
 572 |        "<style scoped>\n",
 573 |        "    .dataframe tbody tr th:only-of-type {\n",
 574 |        "        vertical-align: middle;\n",
 575 |        "    }\n",
 576 |        "\n",
 577 |        "    .dataframe tbody tr th {\n",
 578 |        "        vertical-align: top;\n",
 579 |        "    }\n",
 580 |        "\n",
 581 |        "    .dataframe thead th {\n",
 582 |        "        text-align: right;\n",
 583 |        "    }\n",
 584 |        "</style>\n",
 585 |        "<table border=\"1\" class=\"dataframe\">\n",
 586 |        "  <thead>\n",
 587 |        "    <tr style=\"text-align: right;\">\n",
 588 |        "      <th></th>\n",
 589 |        "      <th>label</th>\n",
 590 |        "      <th>domain</th>\n",
 591 |        "      <th>root</th>\n",
 592 |        "      <th>rootclass</th>\n",
 593 |        "      <th>len</th>\n",
 594 |        "    </tr>\n",
 595 |        "  </thead>\n",
 596 |        "  <tbody>\n",
 597 |        "    <tr>\n",
 598 |        "      <th>0</th>\n",
 599 |        "      <td>0</td>\n",
 600 |        "      <td>facebook.com</td>\n",
 601 |        "      <td>com</td>\n",
 602 |        "      <td>0</td>\n",
 603 |        "      <td>12</td>\n",
 604 |        "    </tr>\n",
 605 |        "    <tr>\n",
 606 |        "      <th>1</th>\n",
 607 |        "      <td>0</td>\n",
 608 |        "      <td>youtube.com</td>\n",
 609 |        "      <td>com</td>\n",
 610 |        "      <td>0</td>\n",
 611 |        "      <td>11</td>\n",
 612 |        "    </tr>\n",
 613 |        "    <tr>\n",
 614 |        "      <th>2</th>\n",
 615 |        "      <td>0</td>\n",
 616 |        "      <td>yahoo.com</td>\n",
 617 |        "      <td>com</td>\n",
 618 |        "      <td>0</td>\n",
 619 |        "      <td>9</td>\n",
 620 |        "    </tr>\n",
 621 |        "    <tr>\n",
 622 |        "      <th>3</th>\n",
 623 |        "      <td>0</td>\n",
 624 |        "      <td>baidu.com</td>\n",
 625 |        "      <td>com</td>\n",
 626 |        "      <td>0</td>\n",
 627 |        "      <td>9</td>\n",
 628 |        "    </tr>\n",
 629 |        "    <tr>\n",
 630 |        "      <th>4</th>\n",
 631 |        "      <td>0</td>\n",
 632 |        "      <td>wikipedia.org</td>\n",
 633 |        "      <td>org</td>\n",
 634 |        "      <td>0</td>\n",
 635 |        "      <td>13</td>\n",
 636 |        "    </tr>\n",
 637 |        "    <tr>\n",
 638 |        "      <th>5</th>\n",
 639 |        "      <td>0</td>\n",
 640 |        "      <td>qq.com</td>\n",
 641 |        "      <td>com</td>\n",
 642 |        "      <td>0</td>\n",
 643 |        "      <td>6</td>\n",
 644 |        "    </tr>\n",
 645 |        "    <tr>\n",
 646 |        "      <th>6</th>\n",
 647 |        "      <td>0</td>\n",
 648 |        "      <td>linkedin.com</td>\n",
 649 |        "      <td>com</td>\n",
 650 |        "      <td>0</td>\n",
 651 |        "      <td>12</td>\n",
 652 |        "    </tr>\n",
 653 |        "    <tr>\n",
 654 |        "      <th>7</th>\n",
 655 |        "      <td>0</td>\n",
 656 |        "      <td>taobao.com</td>\n",
 657 |        "      <td>com</td>\n",
 658 |        "      <td>0</td>\n",
 659 |        "      <td>10</td>\n",
 660 |        "    </tr>\n",
 661 |        "    <tr>\n",
 662 |        "      <th>8</th>\n",
 663 |        "      <td>0</td>\n",
 664 |        "      <td>twitter.com</td>\n",
 665 |        "      <td>com</td>\n",
 666 |        "      <td>0</td>\n",
 667 |        "      <td>11</td>\n",
 668 |        "    </tr>\n",
 669 |        "    <tr>\n",
 670 |        "      <th>9</th>\n",
 671 |        "      <td>0</td>\n",
 672 |        "      <td>live.com</td>\n",
 673 |        "      <td>com</td>\n",
 674 |        "      <td>0</td>\n",
 675 |        "      <td>8</td>\n",
 676 |        "    </tr>\n",
 677 |        "  </tbody>\n",
 678 |        "</table>\n",
 679 |        "</div>"
 680 |       ],
 681 |       "text/plain": [
 682 |        "  label         domain root  rootclass  len\n",
 683 |        "0     0   facebook.com  com          0   12\n",
 684 |        "1     0    youtube.com  com          0   11\n",
 685 |        "2     0      yahoo.com  com          0    9\n",
 686 |        "3     0      baidu.com  com          0    9\n",
 687 |        "4     0  wikipedia.org  org          0   13\n",
 688 |        "5     0         qq.com  com          0    6\n",
 689 |        "6     0   linkedin.com  com          0   12\n",
 690 |        "7     0     taobao.com  com          0   10\n",
 691 |        "8     0    twitter.com  com          0   11\n",
 692 |        "9     0       live.com  com          0    8"
 693 |       ]
 694 |      },
 695 |      "execution_count": 13,
 696 |      "metadata": {},
 697 |      "output_type": "execute_result"
 698 |     }
 699 |    ],
 700 |    "source": [
 701 |     "train_data.head(10)"
 702 |    ]
 703 |   },
 704 |   {
 705 |    "cell_type": "code",
 706 |    "execution_count": 14,
 707 |    "metadata": {},
 708 |    "outputs": [
 709 |     {
 710 |      "data": {
 711 |       "text/html": [
 712 |        "<div>\n",
 713 |        "<style scoped>\n",
 714 |        "    .dataframe tbody tr th:only-of-type {\n",
 715 |        "        vertical-align: middle;\n",
 716 |        "    }\n",
 717 |        "\n",
 718 |        "    .dataframe tbody tr th {\n",
 719 |        "        vertical-align: top;\n",
 720 |        "    }\n",
 721 |        "\n",
 722 |        "    .dataframe thead th {\n",
 723 |        "        text-align: right;\n",
 724 |        "    }\n",
 725 |        "</style>\n",
 726 |        "<table border=\"1\" class=\"dataframe\">\n",
 727 |        "  <thead>\n",
 728 |        "    <tr style=\"text-align: right;\">\n",
 729 |        "      <th></th>\n",
 730 |        "      <th>label</th>\n",
 731 |        "      <th>domain</th>\n",
 732 |        "      <th>root</th>\n",
 733 |        "      <th>rootclass</th>\n",
 734 |        "      <th>len</th>\n",
 735 |        "    </tr>\n",
 736 |        "  </thead>\n",
 737 |        "  <tbody>\n",
 738 |        "    <tr>\n",
 739 |        "      <th>249989</th>\n",
 740 |        "      <td>1</td>\n",
 741 |        "      <td>zzzewwhfc.com.hn</td>\n",
 742 |        "      <td>hn</td>\n",
 743 |        "      <td>1</td>\n",
 744 |        "      <td>16</td>\n",
 745 |        "    </tr>\n",
 746 |        "    <tr>\n",
 747 |        "      <th>249990</th>\n",
 748 |        "      <td>1</td>\n",
 749 |        "      <td>zzzgx.co.vi</td>\n",
 750 |        "      <td>vi</td>\n",
 751 |        "      <td>1</td>\n",
 752 |        "      <td>11</td>\n",
 753 |        "    </tr>\n",
 754 |        "    <tr>\n",
 755 |        "      <th>249991</th>\n",
 756 |        "      <td>1</td>\n",
 757 |        "      <td>zzzjxt.as</td>\n",
 758 |        "      <td>as</td>\n",
 759 |        "      <td>1</td>\n",
 760 |        "      <td>9</td>\n",
 761 |        "    </tr>\n",
 762 |        "    <tr>\n",
 763 |        "      <th>249992</th>\n",
 764 |        "      <td>1</td>\n",
 765 |        "      <td>zzzkf.tc</td>\n",
 766 |        "      <td>tc</td>\n",
 767 |        "      <td>1</td>\n",
 768 |        "      <td>8</td>\n",
 769 |        "    </tr>\n",
 770 |        "    <tr>\n",
 771 |        "      <th>249993</th>\n",
 772 |        "      <td>1</td>\n",
 773 |        "      <td>zzzlxtam.co.kr</td>\n",
 774 |        "      <td>kr</td>\n",
 775 |        "      <td>1</td>\n",
 776 |        "      <td>14</td>\n",
 777 |        "    </tr>\n",
 778 |        "    <tr>\n",
 779 |        "      <th>249994</th>\n",
 780 |        "      <td>1</td>\n",
 781 |        "      <td>zzznnx.ir</td>\n",
 782 |        "      <td>ir</td>\n",
 783 |        "      <td>1</td>\n",
 784 |        "      <td>9</td>\n",
 785 |        "    </tr>\n",
 786 |        "    <tr>\n",
 787 |        "      <th>249995</th>\n",
 788 |        "      <td>1</td>\n",
 789 |        "      <td>zzztqrq.ht</td>\n",
 790 |        "      <td>ht</td>\n",
 791 |        "      <td>1</td>\n",
 792 |        "      <td>10</td>\n",
 793 |        "    </tr>\n",
 794 |        "    <tr>\n",
 795 |        "      <th>249996</th>\n",
 796 |        "      <td>1</td>\n",
 797 |        "      <td>zzzvr.su</td>\n",
 798 |        "      <td>su</td>\n",
 799 |        "      <td>1</td>\n",
 800 |        "      <td>8</td>\n",
 801 |        "    </tr>\n",
 802 |        "    <tr>\n",
 803 |        "      <th>249997</th>\n",
 804 |        "      <td>1</td>\n",
 805 |        "      <td>zzzwtef.com.pa</td>\n",
 806 |        "      <td>pa</td>\n",
 807 |        "      <td>1</td>\n",
 808 |        "      <td>14</td>\n",
 809 |        "    </tr>\n",
 810 |        "    <tr>\n",
 811 |        "      <th>249998</th>\n",
 812 |        "      <td>1</td>\n",
 813 |        "      <td>zzzzuvfbt.co.uk</td>\n",
 814 |        "      <td>uk</td>\n",
 815 |        "      <td>1</td>\n",
 816 |        "      <td>15</td>\n",
 817 |        "    </tr>\n",
 818 |        "  </tbody>\n",
 819 |        "</table>\n",
 820 |        "</div>"
 821 |       ],
 822 |       "text/plain": [
 823 |        "       label            domain root  rootclass  len\n",
 824 |        "249989     1  zzzewwhfc.com.hn   hn          1   16\n",
 825 |        "249990     1       zzzgx.co.vi   vi          1   11\n",
 826 |        "249991     1         zzzjxt.as   as          1    9\n",
 827 |        "249992     1          zzzkf.tc   tc          1    8\n",
 828 |        "249993     1    zzzlxtam.co.kr   kr          1   14\n",
 829 |        "249994     1         zzznnx.ir   ir          1    9\n",
 830 |        "249995     1        zzztqrq.ht   ht          1   10\n",
 831 |        "249996     1          zzzvr.su   su          1    8\n",
 832 |        "249997     1    zzzwtef.com.pa   pa          1   14\n",
 833 |        "249998     1   zzzzuvfbt.co.uk   uk          1   15"
 834 |       ]
 835 |      },
 836 |      "execution_count": 14,
 837 |      "metadata": {},
 838 |      "output_type": "execute_result"
 839 |     }
 840 |    ],
 841 |    "source": [
 842 |     "train_data.tail(10)"
 843 |    ]
 844 |   },
 845 |   {
 846 |    "cell_type": "code",
 847 |    "execution_count": 15,
 848 |    "metadata": {},
 849 |    "outputs": [
 850 |     {
 851 |      "data": {
 852 |       "text/plain": [
 853 |        "<matplotlib.axes._subplots.AxesSubplot at 0x2dc5434fa20>"
 854 |       ]
 855 |      },
 856 |      "execution_count": 15,
 857 |      "metadata": {},
 858 |      "output_type": "execute_result"
 859 |     },
 860 |     {
 861 |      "data": {
 862 |       "image/png": "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\n",
 863 |       "text/plain": [
 864 |        "<Figure size 432x288 with 1 Axes>"
 865 |       ]
 866 |      },
 867 |      "metadata": {
 868 |       "needs_background": "light"
 869 |      },
 870 |      "output_type": "display_data"
 871 |     }
 872 |    ],
 873 |    "source": [
 874 |     "sns.countplot(x='len', hue='label',data=train_data)"
 875 |    ]
 876 |   },
 877 |   {
 878 |    "cell_type": "markdown",
 879 |    "metadata": {},
 880 |    "source": [
 881 |     "我们可以发现非正常域名的长度分布更加集中，而正常域名分布更像是一个正态分布，ok,把他作为一个特征\n",
 882 |     "\n",
 883 |     "下一步要想一想还有什么特征\n",
 884 |     "信息学中有一个很重要的概念，叫熵（虽然它最早出自化学和物理学2333），在信息学中，它用于衡量一个系统的混乱程度，简而言之，对于一个随机字符串，他就代表着字符串中字母出现以及分布的混乱程度，打个比方:\n",
 885 |     "str1 = 'aaabbb'\n",
 886 |     "str2 = 'asdfgh'\n",
 887 |     "两者长度相同，可是看上去str1比str2更整齐，更有规律不少，而str2更像是用脸滚键盘滚出来的，熵主要用于判断器混乱程度，不同于我们用肉眼判断，她有一个著名的公式，叫香农熵公式:\n",
 888 |     "\n",
 889 |     "G = -ΣLog(P(Xi)) * P(Xi)\n",
 890 |     "\n",
 891 |     "其中P(Xi)是字符出现的概率 比如 str1中 P('a') = 3 / (3+3) = 0.5  str2中 P('a') = 1 / (6) = 0.166667 "
 892 |    ]
 893 |   },
 894 |   {
 895 |    "cell_type": "code",
 896 |    "execution_count": 16,
 897 |    "metadata": {},
 898 |    "outputs": [],
 899 |    "source": [
 900 |     "train_data['shan'] = train_data['domain'].map(lambda  x:getshan(x)).astype(float)"
 901 |    ]
 902 |   },
 903 |   {
 904 |    "cell_type": "code",
 905 |    "execution_count": 17,
 906 |    "metadata": {},
 907 |    "outputs": [
 908 |     {
 909 |      "data": {
 910 |       "text/html": [
 911 |        "<div>\n",
 912 |        "<style scoped>\n",
 913 |        "    .dataframe tbody tr th:only-of-type {\n",
 914 |        "        vertical-align: middle;\n",
 915 |        "    }\n",
 916 |        "\n",
 917 |        "    .dataframe tbody tr th {\n",
 918 |        "        vertical-align: top;\n",
 919 |        "    }\n",
 920 |        "\n",
 921 |        "    .dataframe thead th {\n",
 922 |        "        text-align: right;\n",
 923 |        "    }\n",
 924 |        "</style>\n",
 925 |        "<table border=\"1\" class=\"dataframe\">\n",
 926 |        "  <thead>\n",
 927 |        "    <tr style=\"text-align: right;\">\n",
 928 |        "      <th></th>\n",
 929 |        "      <th>label</th>\n",
 930 |        "      <th>domain</th>\n",
 931 |        "      <th>root</th>\n",
 932 |        "      <th>rootclass</th>\n",
 933 |        "      <th>len</th>\n",
 934 |        "      <th>shan</th>\n",
 935 |        "    </tr>\n",
 936 |        "  </thead>\n",
 937 |        "  <tbody>\n",
 938 |        "    <tr>\n",
 939 |        "      <th>0</th>\n",
 940 |        "      <td>0</td>\n",
 941 |        "      <td>facebook.com</td>\n",
 942 |        "      <td>com</td>\n",
 943 |        "      <td>0</td>\n",
 944 |        "      <td>12</td>\n",
 945 |        "      <td>3.022055</td>\n",
 946 |        "    </tr>\n",
 947 |        "    <tr>\n",
 948 |        "      <th>1</th>\n",
 949 |        "      <td>0</td>\n",
 950 |        "      <td>youtube.com</td>\n",
 951 |        "      <td>com</td>\n",
 952 |        "      <td>0</td>\n",
 953 |        "      <td>11</td>\n",
 954 |        "      <td>3.095795</td>\n",
 955 |        "    </tr>\n",
 956 |        "    <tr>\n",
 957 |        "      <th>2</th>\n",
 958 |        "      <td>0</td>\n",
 959 |        "      <td>yahoo.com</td>\n",
 960 |        "      <td>com</td>\n",
 961 |        "      <td>0</td>\n",
 962 |        "      <td>9</td>\n",
 963 |        "      <td>2.641604</td>\n",
 964 |        "    </tr>\n",
 965 |        "    <tr>\n",
 966 |        "      <th>3</th>\n",
 967 |        "      <td>0</td>\n",
 968 |        "      <td>baidu.com</td>\n",
 969 |        "      <td>com</td>\n",
 970 |        "      <td>0</td>\n",
 971 |        "      <td>9</td>\n",
 972 |        "      <td>3.169925</td>\n",
 973 |        "    </tr>\n",
 974 |        "    <tr>\n",
 975 |        "      <th>4</th>\n",
 976 |        "      <td>0</td>\n",
 977 |        "      <td>wikipedia.org</td>\n",
 978 |        "      <td>org</td>\n",
 979 |        "      <td>0</td>\n",
 980 |        "      <td>13</td>\n",
 981 |        "      <td>3.334679</td>\n",
 982 |        "    </tr>\n",
 983 |        "  </tbody>\n",
 984 |        "</table>\n",
 985 |        "</div>"
 986 |       ],
 987 |       "text/plain": [
 988 |        "  label         domain root  rootclass  len      shan\n",
 989 |        "0     0   facebook.com  com          0   12  3.022055\n",
 990 |        "1     0    youtube.com  com          0   11  3.095795\n",
 991 |        "2     0      yahoo.com  com          0    9  2.641604\n",
 992 |        "3     0      baidu.com  com          0    9  3.169925\n",
 993 |        "4     0  wikipedia.org  org          0   13  3.334679"
 994 |       ]
 995 |      },
 996 |      "execution_count": 17,
 997 |      "metadata": {},
 998 |      "output_type": "execute_result"
 999 |     }
1000 |    ],
1001 |    "source": [
1002 |     "train_data.head(5)"
1003 |    ]
1004 |   },
1005 |   {
1006 |    "cell_type": "code",
1007 |    "execution_count": 18,
1008 |    "metadata": {},
1009 |    "outputs": [
1010 |     {
1011 |      "data": {
1012 |       "text/html": [
1013 |        "<div>\n",
1014 |        "<style scoped>\n",
1015 |        "    .dataframe tbody tr th:only-of-type {\n",
1016 |        "        vertical-align: middle;\n",
1017 |        "    }\n",
1018 |        "\n",
1019 |        "    .dataframe tbody tr th {\n",
1020 |        "        vertical-align: top;\n",
1021 |        "    }\n",
1022 |        "\n",
1023 |        "    .dataframe thead th {\n",
1024 |        "        text-align: right;\n",
1025 |        "    }\n",
1026 |        "</style>\n",
1027 |        "<table border=\"1\" class=\"dataframe\">\n",
1028 |        "  <thead>\n",
1029 |        "    <tr style=\"text-align: right;\">\n",
1030 |        "      <th></th>\n",
1031 |        "      <th>label</th>\n",
1032 |        "      <th>domain</th>\n",
1033 |        "      <th>root</th>\n",
1034 |        "      <th>rootclass</th>\n",
1035 |        "      <th>len</th>\n",
1036 |        "      <th>shan</th>\n",
1037 |        "    </tr>\n",
1038 |        "  </thead>\n",
1039 |        "  <tbody>\n",
1040 |        "    <tr>\n",
1041 |        "      <th>249994</th>\n",
1042 |        "      <td>1</td>\n",
1043 |        "      <td>zzznnx.ir</td>\n",
1044 |        "      <td>ir</td>\n",
1045 |        "      <td>1</td>\n",
1046 |        "      <td>9</td>\n",
1047 |        "      <td>2.419382</td>\n",
1048 |        "    </tr>\n",
1049 |        "    <tr>\n",
1050 |        "      <th>249995</th>\n",
1051 |        "      <td>1</td>\n",
1052 |        "      <td>zzztqrq.ht</td>\n",
1053 |        "      <td>ht</td>\n",
1054 |        "      <td>1</td>\n",
1055 |        "      <td>10</td>\n",
1056 |        "      <td>2.446439</td>\n",
1057 |        "    </tr>\n",
1058 |        "    <tr>\n",
1059 |        "      <th>249996</th>\n",
1060 |        "      <td>1</td>\n",
1061 |        "      <td>zzzvr.su</td>\n",
1062 |        "      <td>su</td>\n",
1063 |        "      <td>1</td>\n",
1064 |        "      <td>8</td>\n",
1065 |        "      <td>2.405639</td>\n",
1066 |        "    </tr>\n",
1067 |        "    <tr>\n",
1068 |        "      <th>249997</th>\n",
1069 |        "      <td>1</td>\n",
1070 |        "      <td>zzzwtef.com.pa</td>\n",
1071 |        "      <td>pa</td>\n",
1072 |        "      <td>1</td>\n",
1073 |        "      <td>14</td>\n",
1074 |        "      <td>3.324863</td>\n",
1075 |        "    </tr>\n",
1076 |        "    <tr>\n",
1077 |        "      <th>249998</th>\n",
1078 |        "      <td>1</td>\n",
1079 |        "      <td>zzzzuvfbt.co.uk</td>\n",
1080 |        "      <td>uk</td>\n",
1081 |        "      <td>1</td>\n",
1082 |        "      <td>15</td>\n",
1083 |        "      <td>3.106891</td>\n",
1084 |        "    </tr>\n",
1085 |        "  </tbody>\n",
1086 |        "</table>\n",
1087 |        "</div>"
1088 |       ],
1089 |       "text/plain": [
1090 |        "       label           domain root  rootclass  len      shan\n",
1091 |        "249994     1        zzznnx.ir   ir          1    9  2.419382\n",
1092 |        "249995     1       zzztqrq.ht   ht          1   10  2.446439\n",
1093 |        "249996     1         zzzvr.su   su          1    8  2.405639\n",
1094 |        "249997     1   zzzwtef.com.pa   pa          1   14  3.324863\n",
1095 |        "249998     1  zzzzuvfbt.co.uk   uk          1   15  3.106891"
1096 |       ]
1097 |      },
1098 |      "execution_count": 18,
1099 |      "metadata": {},
1100 |      "output_type": "execute_result"
1101 |     }
1102 |    ],
1103 |    "source": [
1104 |     "train_data.tail(5)"
1105 |    ]
1106 |   },
1107 |   {
1108 |    "cell_type": "markdown",
1109 |    "metadata": {},
1110 |    "source": [
1111 |     "sns.pointplot(x='label', y='shan',data=train_data)"
1112 |    ]
1113 |   },
1114 |   {
1115 |    "cell_type": "code",
1116 |    "execution_count": 19,
1117 |    "metadata": {},
1118 |    "outputs": [
1119 |     {
1120 |      "data": {
1121 |       "text/plain": [
1122 |        "<matplotlib.axes._subplots.AxesSubplot at 0x2dc544b65c0>"
1123 |       ]
1124 |      },
1125 |      "execution_count": 19,
1126 |      "metadata": {},
1127 |      "output_type": "execute_result"
1128 |     },
1129 |     {
1130 |      "data": {
1131 |       "image/png": "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\n",
1132 |       "text/plain": [
1133 |        "<Figure size 432x288 with 1 Axes>"
1134 |       ]
1135 |      },
1136 |      "metadata": {
1137 |       "needs_background": "light"
1138 |      },
1139 |      "output_type": "display_data"
1140 |     }
1141 |    ],
1142 |    "source": [
1143 |     "sns.pointplot(x='label', y='shan',data=train_data)"
1144 |    ]
1145 |   },
1146 |   {
1147 |    "cell_type": "markdown",
1148 |    "metadata": {},
1149 |    "source": [
1150 |     "接下来还有什么参数呢，plunter的原文从语言学提供了提供了一些新的思路，比如说 好读\n",
1151 |     "\n",
1152 |     "没错，根据常规来说一个正常的网站一定会希望他的网站域名是读起来朗朗上口并且好记的，根据语言学和音韵学来说，貌似在字符中插入一定的元音字母会让一个字符串卡萨诺航区或者读起来更像是一个单词一样，元音是啥 a e i o u ，比如说 baidu 五个字母中有3个都是元音 再比如说 google 六个字母中有3个元音，比如说zzzzuvfbt  9个单词中只有一个元音...\n",
1153 |     "\n",
1154 |     "看上去这会是一个好的特征 我们来看看！\n",
1155 |     "以下一行代码用于计算元音字母所占的比例"
1156 |    ]
1157 |   },
1158 |   {
1159 |    "cell_type": "code",
1160 |    "execution_count": 20,
1161 |    "metadata": {},
1162 |    "outputs": [],
1163 |    "source": [
1164 |     "train_data['yuan_ratio'] = train_data['domain'].map(lambda x:getyuanyin(x)).astype(float)"
1165 |    ]
1166 |   },
1167 |   {
1168 |    "cell_type": "code",
1169 |    "execution_count": 21,
1170 |    "metadata": {},
1171 |    "outputs": [
1172 |     {
1173 |      "data": {
1174 |       "text/html": [
1175 |        "<div>\n",
1176 |        "<style scoped>\n",
1177 |        "    .dataframe tbody tr th:only-of-type {\n",
1178 |        "        vertical-align: middle;\n",
1179 |        "    }\n",
1180 |        "\n",
1181 |        "    .dataframe tbody tr th {\n",
1182 |        "        vertical-align: top;\n",
1183 |        "    }\n",
1184 |        "\n",
1185 |        "    .dataframe thead th {\n",
1186 |        "        text-align: right;\n",
1187 |        "    }\n",
1188 |        "</style>\n",
1189 |        "<table border=\"1\" class=\"dataframe\">\n",
1190 |        "  <thead>\n",
1191 |        "    <tr style=\"text-align: right;\">\n",
1192 |        "      <th></th>\n",
1193 |        "      <th>label</th>\n",
1194 |        "      <th>domain</th>\n",
1195 |        "      <th>root</th>\n",
1196 |        "      <th>rootclass</th>\n",
1197 |        "      <th>len</th>\n",
1198 |        "      <th>shan</th>\n",
1199 |        "      <th>yuan_ratio</th>\n",
1200 |        "    </tr>\n",
1201 |        "  </thead>\n",
1202 |        "  <tbody>\n",
1203 |        "    <tr>\n",
1204 |        "      <th>0</th>\n",
1205 |        "      <td>0</td>\n",
1206 |        "      <td>facebook.com</td>\n",
1207 |        "      <td>com</td>\n",
1208 |        "      <td>0</td>\n",
1209 |        "      <td>12</td>\n",
1210 |        "      <td>3.022055</td>\n",
1211 |        "      <td>0.454545</td>\n",
1212 |        "    </tr>\n",
1213 |        "    <tr>\n",
1214 |        "      <th>1</th>\n",
1215 |        "      <td>0</td>\n",
1216 |        "      <td>youtube.com</td>\n",
1217 |        "      <td>com</td>\n",
1218 |        "      <td>0</td>\n",
1219 |        "      <td>11</td>\n",
1220 |        "      <td>3.095795</td>\n",
1221 |        "      <td>0.500000</td>\n",
1222 |        "    </tr>\n",
1223 |        "    <tr>\n",
1224 |        "      <th>2</th>\n",
1225 |        "      <td>0</td>\n",
1226 |        "      <td>yahoo.com</td>\n",
1227 |        "      <td>com</td>\n",
1228 |        "      <td>0</td>\n",
1229 |        "      <td>9</td>\n",
1230 |        "      <td>2.641604</td>\n",
1231 |        "      <td>0.500000</td>\n",
1232 |        "    </tr>\n",
1233 |        "    <tr>\n",
1234 |        "      <th>3</th>\n",
1235 |        "      <td>0</td>\n",
1236 |        "      <td>baidu.com</td>\n",
1237 |        "      <td>com</td>\n",
1238 |        "      <td>0</td>\n",
1239 |        "      <td>9</td>\n",
1240 |        "      <td>3.169925</td>\n",
1241 |        "      <td>0.500000</td>\n",
1242 |        "    </tr>\n",
1243 |        "    <tr>\n",
1244 |        "      <th>4</th>\n",
1245 |        "      <td>0</td>\n",
1246 |        "      <td>wikipedia.org</td>\n",
1247 |        "      <td>org</td>\n",
1248 |        "      <td>0</td>\n",
1249 |        "      <td>13</td>\n",
1250 |        "      <td>3.334679</td>\n",
1251 |        "      <td>0.500000</td>\n",
1252 |        "    </tr>\n",
1253 |        "  </tbody>\n",
1254 |        "</table>\n",
1255 |        "</div>"
1256 |       ],
1257 |       "text/plain": [
1258 |        "  label         domain root  rootclass  len      shan  yuan_ratio\n",
1259 |        "0     0   facebook.com  com          0   12  3.022055    0.454545\n",
1260 |        "1     0    youtube.com  com          0   11  3.095795    0.500000\n",
1261 |        "2     0      yahoo.com  com          0    9  2.641604    0.500000\n",
1262 |        "3     0      baidu.com  com          0    9  3.169925    0.500000\n",
1263 |        "4     0  wikipedia.org  org          0   13  3.334679    0.500000"
1264 |       ]
1265 |      },
1266 |      "execution_count": 21,
1267 |      "metadata": {},
1268 |      "output_type": "execute_result"
1269 |     }
1270 |    ],
1271 |    "source": [
1272 |     "train_data.head(5)"
1273 |    ]
1274 |   },
1275 |   {
1276 |    "cell_type": "code",
1277 |    "execution_count": 22,
1278 |    "metadata": {},
1279 |    "outputs": [
1280 |     {
1281 |      "data": {
1282 |       "text/html": [
1283 |        "<div>\n",
1284 |        "<style scoped>\n",
1285 |        "    .dataframe tbody tr th:only-of-type {\n",
1286 |        "        vertical-align: middle;\n",
1287 |        "    }\n",
1288 |        "\n",
1289 |        "    .dataframe tbody tr th {\n",
1290 |        "        vertical-align: top;\n",
1291 |        "    }\n",
1292 |        "\n",
1293 |        "    .dataframe thead th {\n",
1294 |        "        text-align: right;\n",
1295 |        "    }\n",
1296 |        "</style>\n",
1297 |        "<table border=\"1\" class=\"dataframe\">\n",
1298 |        "  <thead>\n",
1299 |        "    <tr style=\"text-align: right;\">\n",
1300 |        "      <th></th>\n",
1301 |        "      <th>label</th>\n",
1302 |        "      <th>domain</th>\n",
1303 |        "      <th>root</th>\n",
1304 |        "      <th>rootclass</th>\n",
1305 |        "      <th>len</th>\n",
1306 |        "      <th>shan</th>\n",
1307 |        "      <th>yuan_ratio</th>\n",
1308 |        "    </tr>\n",
1309 |        "  </thead>\n",
1310 |        "  <tbody>\n",
1311 |        "    <tr>\n",
1312 |        "      <th>249994</th>\n",
1313 |        "      <td>1</td>\n",
1314 |        "      <td>zzznnx.ir</td>\n",
1315 |        "      <td>ir</td>\n",
1316 |        "      <td>1</td>\n",
1317 |        "      <td>9</td>\n",
1318 |        "      <td>2.419382</td>\n",
1319 |        "      <td>0.125000</td>\n",
1320 |        "    </tr>\n",
1321 |        "    <tr>\n",
1322 |        "      <th>249995</th>\n",
1323 |        "      <td>1</td>\n",
1324 |        "      <td>zzztqrq.ht</td>\n",
1325 |        "      <td>ht</td>\n",
1326 |        "      <td>1</td>\n",
1327 |        "      <td>10</td>\n",
1328 |        "      <td>2.446439</td>\n",
1329 |        "      <td>0.000000</td>\n",
1330 |        "    </tr>\n",
1331 |        "    <tr>\n",
1332 |        "      <th>249996</th>\n",
1333 |        "      <td>1</td>\n",
1334 |        "      <td>zzzvr.su</td>\n",
1335 |        "      <td>su</td>\n",
1336 |        "      <td>1</td>\n",
1337 |        "      <td>8</td>\n",
1338 |        "      <td>2.405639</td>\n",
1339 |        "      <td>0.142857</td>\n",
1340 |        "    </tr>\n",
1341 |        "    <tr>\n",
1342 |        "      <th>249997</th>\n",
1343 |        "      <td>1</td>\n",
1344 |        "      <td>zzzwtef.com.pa</td>\n",
1345 |        "      <td>pa</td>\n",
1346 |        "      <td>1</td>\n",
1347 |        "      <td>14</td>\n",
1348 |        "      <td>3.324863</td>\n",
1349 |        "      <td>0.250000</td>\n",
1350 |        "    </tr>\n",
1351 |        "    <tr>\n",
1352 |        "      <th>249998</th>\n",
1353 |        "      <td>1</td>\n",
1354 |        "      <td>zzzzuvfbt.co.uk</td>\n",
1355 |        "      <td>uk</td>\n",
1356 |        "      <td>1</td>\n",
1357 |        "      <td>15</td>\n",
1358 |        "      <td>3.106891</td>\n",
1359 |        "      <td>0.230769</td>\n",
1360 |        "    </tr>\n",
1361 |        "  </tbody>\n",
1362 |        "</table>\n",
1363 |        "</div>"
1364 |       ],
1365 |       "text/plain": [
1366 |        "       label           domain root  rootclass  len      shan  yuan_ratio\n",
1367 |        "249994     1        zzznnx.ir   ir          1    9  2.419382    0.125000\n",
1368 |        "249995     1       zzztqrq.ht   ht          1   10  2.446439    0.000000\n",
1369 |        "249996     1         zzzvr.su   su          1    8  2.405639    0.142857\n",
1370 |        "249997     1   zzzwtef.com.pa   pa          1   14  3.324863    0.250000\n",
1371 |        "249998     1  zzzzuvfbt.co.uk   uk          1   15  3.106891    0.230769"
1372 |       ]
1373 |      },
1374 |      "execution_count": 22,
1375 |      "metadata": {},
1376 |      "output_type": "execute_result"
1377 |     }
1378 |    ],
1379 |    "source": [
1380 |     "train_data.tail(5)"
1381 |    ]
1382 |   },
1383 |   {
1384 |    "cell_type": "markdown",
1385 |    "metadata": {},
1386 |    "source": [
1387 |     "结果显而易见 正常域名中元音所占的比率远远大于非正常域名！\n",
1388 |     "\n",
1389 |     "OK，暂时想不到其他的什么特征了，就用这几个建立一个demo版本吧\n",
1390 |     "\n",
1391 |     "特征有了，下一步干啥呢\n",
1392 |     "\n",
1393 |     "归一化咯，所谓归一化，就是不数据的范围放缩至一个相同的范围以避免数据跨度或量级的巨大差别而导致训练上的困难或者结果上的偏差\n",
1394 |     "sklearn给出了丰富强大的接口来帮你"
1395 |    ]
1396 |   },
1397 |   {
1398 |    "cell_type": "code",
1399 |    "execution_count": 23,
1400 |    "metadata": {},
1401 |    "outputs": [
1402 |     {
1403 |      "name": "stderr",
1404 |      "output_type": "stream",
1405 |      "text": [
1406 |       "c:\\program files\\python36\\lib\\site-packages\\sklearn\\utils\\validation.py:475: DataConversionWarning: Data with input dtype int32 was converted to float64 by StandardScaler.\n",
1407 |       "  warnings.warn(msg, DataConversionWarning)\n"
1408 |      ]
1409 |     }
1410 |    ],
1411 |    "source": [
1412 |     "scaler = preprocessing.StandardScaler()\n",
1413 |     "len_scale_param = scaler.fit(train_data['len'].values.reshape(-1,1))\n",
1414 |     "train_data['len_scaled'] = scaler.fit_transform(train_data['len'].values.reshape(-1,1),len_scale_param)\n",
1415 |     "shan_scale_param = scaler.fit(train_data['shan'].values.reshape(-1,1))\n",
1416 |     "train_data['shan_sclaed'] = scaler.fit_transform(train_data['shan'].values.reshape(-1,1),shan_scale_param)"
1417 |    ]
1418 |   },
1419 |   {
1420 |    "cell_type": "markdown",
1421 |    "metadata": {},
1422 |    "source": [
1423 |     "ok 现在看一下我们在经过特征工程和数据预处理以后的数据长啥样吧"
1424 |    ]
1425 |   },
1426 |   {
1427 |    "cell_type": "code",
1428 |    "execution_count": 24,
1429 |    "metadata": {},
1430 |    "outputs": [
1431 |     {
1432 |      "data": {
1433 |       "text/html": [
1434 |        "<div>\n",
1435 |        "<style scoped>\n",
1436 |        "    .dataframe tbody tr th:only-of-type {\n",
1437 |        "        vertical-align: middle;\n",
1438 |        "    }\n",
1439 |        "\n",
1440 |        "    .dataframe tbody tr th {\n",
1441 |        "        vertical-align: top;\n",
1442 |        "    }\n",
1443 |        "\n",
1444 |        "    .dataframe thead th {\n",
1445 |        "        text-align: right;\n",
1446 |        "    }\n",
1447 |        "</style>\n",
1448 |        "<table border=\"1\" class=\"dataframe\">\n",
1449 |        "  <thead>\n",
1450 |        "    <tr style=\"text-align: right;\">\n",
1451 |        "      <th></th>\n",
1452 |        "      <th>label</th>\n",
1453 |        "      <th>domain</th>\n",
1454 |        "      <th>root</th>\n",
1455 |        "      <th>rootclass</th>\n",
1456 |        "      <th>len</th>\n",
1457 |        "      <th>shan</th>\n",
1458 |        "      <th>yuan_ratio</th>\n",
1459 |        "      <th>len_scaled</th>\n",
1460 |        "      <th>shan_sclaed</th>\n",
1461 |        "    </tr>\n",
1462 |        "  </thead>\n",
1463 |        "  <tbody>\n",
1464 |        "    <tr>\n",
1465 |        "      <th>0</th>\n",
1466 |        "      <td>0</td>\n",
1467 |        "      <td>facebook.com</td>\n",
1468 |        "      <td>com</td>\n",
1469 |        "      <td>0</td>\n",
1470 |        "      <td>12</td>\n",
1471 |        "      <td>3.022055</td>\n",
1472 |        "      <td>0.454545</td>\n",
1473 |        "      <td>0.043608</td>\n",
1474 |        "      <td>-0.341005</td>\n",
1475 |        "    </tr>\n",
1476 |        "    <tr>\n",
1477 |        "      <th>1</th>\n",
1478 |        "      <td>0</td>\n",
1479 |        "      <td>youtube.com</td>\n",
1480 |        "      <td>com</td>\n",
1481 |        "      <td>0</td>\n",
1482 |        "      <td>11</td>\n",
1483 |        "      <td>3.095795</td>\n",
1484 |        "      <td>0.500000</td>\n",
1485 |        "      <td>-0.237871</td>\n",
1486 |        "      <td>-0.113708</td>\n",
1487 |        "    </tr>\n",
1488 |        "    <tr>\n",
1489 |        "      <th>2</th>\n",
1490 |        "      <td>0</td>\n",
1491 |        "      <td>yahoo.com</td>\n",
1492 |        "      <td>com</td>\n",
1493 |        "      <td>0</td>\n",
1494 |        "      <td>9</td>\n",
1495 |        "      <td>2.641604</td>\n",
1496 |        "      <td>0.500000</td>\n",
1497 |        "      <td>-0.800828</td>\n",
1498 |        "      <td>-1.513712</td>\n",
1499 |        "    </tr>\n",
1500 |        "    <tr>\n",
1501 |        "      <th>3</th>\n",
1502 |        "      <td>0</td>\n",
1503 |        "      <td>baidu.com</td>\n",
1504 |        "      <td>com</td>\n",
1505 |        "      <td>0</td>\n",
1506 |        "      <td>9</td>\n",
1507 |        "      <td>3.169925</td>\n",
1508 |        "      <td>0.500000</td>\n",
1509 |        "      <td>-0.800828</td>\n",
1510 |        "      <td>0.114791</td>\n",
1511 |        "    </tr>\n",
1512 |        "    <tr>\n",
1513 |        "      <th>4</th>\n",
1514 |        "      <td>0</td>\n",
1515 |        "      <td>wikipedia.org</td>\n",
1516 |        "      <td>org</td>\n",
1517 |        "      <td>0</td>\n",
1518 |        "      <td>13</td>\n",
1519 |        "      <td>3.334679</td>\n",
1520 |        "      <td>0.500000</td>\n",
1521 |        "      <td>0.325087</td>\n",
1522 |        "      <td>0.622631</td>\n",
1523 |        "    </tr>\n",
1524 |        "  </tbody>\n",
1525 |        "</table>\n",
1526 |        "</div>"
1527 |       ],
1528 |       "text/plain": [
1529 |        "  label         domain root  rootclass  len      shan  yuan_ratio  len_scaled  \\\n",
1530 |        "0     0   facebook.com  com          0   12  3.022055    0.454545    0.043608   \n",
1531 |        "1     0    youtube.com  com          0   11  3.095795    0.500000   -0.237871   \n",
1532 |        "2     0      yahoo.com  com          0    9  2.641604    0.500000   -0.800828   \n",
1533 |        "3     0      baidu.com  com          0    9  3.169925    0.500000   -0.800828   \n",
1534 |        "4     0  wikipedia.org  org          0   13  3.334679    0.500000    0.325087   \n",
1535 |        "\n",
1536 |        "   shan_sclaed  \n",
1537 |        "0    -0.341005  \n",
1538 |        "1    -0.113708  \n",
1539 |        "2    -1.513712  \n",
1540 |        "3     0.114791  \n",
1541 |        "4     0.622631  "
1542 |       ]
1543 |      },
1544 |      "execution_count": 24,
1545 |      "metadata": {},
1546 |      "output_type": "execute_result"
1547 |     }
1548 |    ],
1549 |    "source": [
1550 |     "train_data.head(5)"
1551 |    ]
1552 |   },
1553 |   {
1554 |    "cell_type": "markdown",
1555 |    "metadata": {},
1556 |    "source": [
1557 |     "看上去是一些靠谱的特征了，那么最后一步，把我们需要的特征都拿出来，带入模型训练\n",
1558 |     "\n",
1559 |     "首先，我们需要的列：rootclass,yuan_ratio,len_scaled,shan_scaled是我们所需要的，把他们拿出来，转化成numpy矩阵的格式"
1560 |    ]
1561 |   },
1562 |   {
1563 |    "cell_type": "code",
1564 |    "execution_count": 25,
1565 |    "metadata": {},
1566 |    "outputs": [],
1567 |    "source": [
1568 |     "train_pre = train_data.filter(regex = 'label|yuan_ratio|len_scaled|shan_scale_param|rootclass')\n",
1569 |     "train_pre = shuffle(train_pre)"
1570 |    ]
1571 |   },
1572 |   {
1573 |    "cell_type": "markdown",
1574 |    "metadata": {},
1575 |    "source": [
1576 |     "shuffle()是一个很好用的函数，他将帮你把你的数据集打乱顺序，避免相同label的数据都堆积在一起\n",
1577 |     "\n",
1578 |     "好啦！接下来，我们把数据转化成矩阵，并且切分成训练集和测试集（正常来说还应该有一个数据集叫做验证集，本例只作简单演示，就去掉啦）"
1579 |    ]
1580 |   },
1581 |   {
1582 |    "cell_type": "code",
1583 |    "execution_count": 26,
1584 |    "metadata": {},
1585 |    "outputs": [
1586 |     {
1587 |      "name": "stderr",
1588 |      "output_type": "stream",
1589 |      "text": [
1590 |       "C:\\Users\\aldin\\AppData\\Roaming\\Python\\Python36\\site-packages\\ipykernel_launcher.py:1: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
1591 |       "  \"\"\"Entry point for launching an IPython kernel.\n"
1592 |      ]
1593 |     }
1594 |    ],
1595 |    "source": [
1596 |     "train_pre =train_pre.as_matrix()\n",
1597 |     "y_train = train_pre[0:200000,0]\n",
1598 |     "x_train = train_pre[0:200000,1:]\n",
1599 |     "y_test = train_pre[200000:201000,0]\n",
1600 |     "x_test = train_pre[200000:201000,1:]"
1601 |    ]
1602 |   },
1603 |   {
1604 |    "cell_type": "markdown",
1605 |    "metadata": {},
1606 |    "source": [
1607 |     "OK  数据处理的过程到这里就已经全部结束了，接下来要做的，就是选择一个合适的模型进行训练了\n",
1608 |     "\n",
1609 |     "本次我们选择经典的SVM函数，并选择RBF作为核方法"
1610 |    ]
1611 |   },
1612 |   {
1613 |    "cell_type": "code",
1614 |    "execution_count": 28,
1615 |    "metadata": {},
1616 |    "outputs": [
1617 |     {
1618 |      "name": "stdout",
1619 |      "output_type": "stream",
1620 |      "text": [
1621 |       "training finished\n"
1622 |      ]
1623 |     }
1624 |    ],
1625 |    "source": [
1626 |     "model = SVC(kernel='rbf',C=0.4).fit(x_train,y_train)\n",
1627 |     "print ('training finished')"
1628 |    ]
1629 |   },
1630 |   {
1631 |    "cell_type": "code",
1632 |    "execution_count": 30,
1633 |    "metadata": {},
1634 |    "outputs": [
1635 |     {
1636 |      "name": "stdout",
1637 |      "output_type": "stream",
1638 |      "text": [
1639 |       "[0.88059701 0.87       0.865      0.9        0.89447236]\n"
1640 |      ]
1641 |     }
1642 |    ],
1643 |    "source": [
1644 |     "pre = cross_validation.cross_val_score(model,x_test,y_test,cv = 5)\n",
1645 |     "print (pre)"
1646 |    ]
1647 |   },
1648 |   {
1649 |    "cell_type": "markdown",
1650 |    "metadata": {},
1651 |    "source": [
1652 |     "使用测试机集做了一哈验证，平均准确率在88%上下，emmmm 效果差强人意，毕竟只是一个小样例，要求不能太高哈23333\n",
1653 |     "\n",
1654 |     "本篇over 如果想了解关机器学习算法  可以参考 李航博士的 《统计学习方法》 和周志华教授的 《机器学习》\n",
1655 |     "\n",
1656 |     "下一篇将会尝试使用深度学习的方法完成这一任务并大大提升准确率"
1657 |    ]
1658 |   },
1659 |   {
1660 |    "cell_type": "code",
1661 |    "execution_count": null,
1662 |    "metadata": {},
1663 |    "outputs": [],
1664 |    "source": []
1665 |   }
1666 |  ],
1667 |  "metadata": {
1668 |   "kernelspec": {
1669 |    "display_name": "Python 3",
1670 |    "language": "python",
1671 |    "name": "python3"
1672 |   },
1673 |   "language_info": {
1674 |    "codemirror_mode": {
1675 |     "name": "ipython",
1676 |     "version": 3
1677 |    },
1678 |    "file_extension": ".py",
1679 |    "mimetype": "text/x-python",
1680 |    "name": "python",
1681 |    "nbconvert_exporter": "python",
1682 |    "pygments_lexer": "ipython3",
1683 |    "version": "3.6.6"
1684 |   }
1685 |  },
1686 |  "nbformat": 4,
1687 |  "nbformat_minor": 2
1688 | }
1689 | 


--------------------------------------------------------------------------------
/dga/word2vec.txt:
--------------------------------------------------------------------------------
 1 | 37 100
 2 | e 0.22894579 0.018125884 -0.047007907 0.026391089 -0.08410629 0.042102605 0.19332448 0.04419649 0.22332841 0.12207879 0.20176673 0.2137046 0.144299 0.1077951 0.17956123 0.11858179 -0.33023158 -0.23130865 0.079242095 -0.1778481 0.18652116 0.20735946 -0.09724756 -0.29475522 -0.040154807 -0.0527085 0.08248274 0.028697744 0.20734191 0.1434786 -0.18361384 0.2517993 0.22861674 -0.018833853 -0.12170067 -0.24051236 0.14046186 0.18904021 -0.052375365 -0.019918216 -0.00026703495 0.065446615 0.01539892 0.17162453 0.23800771 0.17657512 0.12586145 -0.34606647 -0.5106507 0.0014673104 -0.11829126 -0.15732902 0.00059355557 -0.117669284 0.16796382 -0.22589684 0.18324699 -0.1432052 -0.10824973 -0.056051962 -0.03304817 -0.19836514 0.043790042 -0.07902638 -0.38936022 -0.053455934 -0.06137043 0.010520164 0.06788945 -0.09357034 0.08171352 0.037035745 0.23434442 -0.066095695 -0.15416895 -0.09993183 -0.06499494 0.29420602 0.3641291 0.077296965 -0.58880377 0.28161004 0.26197237 -0.09671587 0.07037314 -0.027387964 0.022457862 0.33194247 0.23548254 -0.15657422 -0.36731052 -0.14385466 -0.30986518 -0.01667506 -0.013026664 -0.04292883 -0.14403225 0.41363692 0.18245754 -0.30445033
 3 | a 0.21127716 0.14811696 -0.03003102 0.05515453 0.06496612 -0.03163473 0.0717927 0.17556895 0.1534142 0.047990672 -0.11152343 0.3526315 0.05873441 0.07725593 -0.031657092 0.27710766 -0.18533958 -0.27010047 0.038175747 -0.08855002 0.15211913 0.06758541 -0.19029911 -0.031179864 -0.102941684 -0.21563244 -0.16258068 -0.021262076 0.29576904 0.25551277 -0.19412133 0.32796228 0.12475944 -0.06434012 -0.20379956 -0.1262172 -0.07483858 0.028510088 0.0040211985 -0.08000011 -0.12535138 -0.003330886 0.1633749 0.0709195 0.23351166 0.26572466 0.38097504 -0.33184552 -0.5133895 -0.04694748 -0.090785466 -0.07870579 -0.00729043 0.1079951 0.031106982 -0.42863268 -0.033870514 -0.009291893 0.07710043 0.0560411 0.067954846 -0.16432801 0.06393452 -0.2838682 -0.14935917 -0.11627777 -0.19536492 0.13162853 0.05711524 -0.33464226 -0.19463703 0.115609266 0.3494651 -0.04769006 -0.22387631 -0.13531409 -0.08333421 0.08529311 0.1341444 0.106381 -0.55082947 0.24773557 0.15607637 -0.043889586 0.07568372 0.0297078 0.07108098 0.39598778 0.26912645 -0.11612211 -0.0066814744 -0.09194973 -0.1752576 -0.1068538 -0.15394345 0.033124976 -0.30770463 0.28185263 0.1125245 -0.21409176
 4 | o 0.17241064 -0.11035399 -0.19319221 0.08719776 0.012665477 0.063720874 0.14657238 0.108920276 0.17844374 0.18124618 0.09626339 0.24243796 0.12515062 0.17865439 0.03858434 0.19961677 -0.18180715 -0.12171747 0.022241244 0.058635764 0.21420005 0.19200677 -0.15092774 -0.02963157 -0.1997332 -0.1686099 -0.15107721 -0.01915387 0.22809373 0.06808437 -0.2673725 0.14992628 0.1084274 -0.14297077 -0.06197483 -0.19030072 0.020712152 0.1066805 -0.032178707 0.025710855 -0.08665498 0.031103078 0.048560753 0.14175373 0.27104673 0.14172524 0.19032608 -0.25986394 -0.28857085 0.09405997 -0.014304649 -0.10768316 0.019913027 -0.004183936 0.061272252 -0.29369318 0.04766315 0.02754595 -0.11497659 -0.09197872 0.036736388 -0.22072764 -0.030539906 -0.27466974 -0.088670015 -0.11645919 -0.03442033 0.048364207 0.052036628 -0.2041584 -0.0890612 0.082437105 0.22399682 0.023273056 -0.17331229 -0.08684524 -0.12465624 0.18048517 0.15067668 0.21664228 -0.45735112 0.2692619 0.2516533 -0.079208806 0.11983368 0.049332608 0.13919015 0.3201688 0.1102203 -0.1598032 -0.15656872 0.06059257 -0.27595767 -0.05109261 -0.12955804 0.0150232455 -0.20562074 0.31906363 0.21484847 -0.1636679
 5 | i 0.08579124 -0.15353522 0.14532387 0.17162168 0.030491048 0.21689677 0.09850071 -0.03441859 0.12754868 0.07854485 -0.063718274 0.26944476 0.13883087 0.16292652 0.12092153 0.049513735 -0.35713485 -0.07380078 0.013431112 -0.07059379 0.20382492 0.0149470605 -0.24065535 0.06838101 -0.376426 0.38805887 0.26093885 -0.40312588 0.12572451 0.22894636 -0.34405655 0.02117287 0.011127326 -0.2525314 -0.20267929 -0.37499923 0.085586764 0.072025575 -0.028773377 -0.09175217 -0.14791872 -0.054004148 0.20766218 0.15477945 -0.08297017 -0.00020648059 0.1284345 -0.0743652 -0.339893 -0.06824735 0.072790354 -0.147597 -0.09827369 -0.114853315 0.005115119 -0.14271489 0.24529772 -0.015054964 -0.08862493 -0.22117288 -0.078903824 -0.15407997 -0.037262432 -0.13225567 -0.18591478 -0.0069512017 -0.0482997 0.04666416 0.06428531 -0.1747329 -0.18825261 0.1530093 0.25019535 0.13215926 -0.015347618 -0.018832192 -0.19106537 0.012065393 0.0810764 -0.12915671 -0.45243868 0.07409338 0.27199277 -0.15171236 0.16402017 0.060268417 0.19184433 0.26624554 0.20890296 -0.028436264 -0.19546644 0.029902337 -0.17480354 0.07310163 -0.08564797 0.021542743 -0.0030476353 0.25609583 0.053997647 -0.0931804
 6 | s -0.038906075 -0.19186044 0.058106318 0.034708466 0.115723334 -0.05148212 0.19656909 -0.036778096 0.2144819 -0.031298615 0.10551325 0.07181609 0.1356639 0.06627228 0.095314175 0.2233329 -0.06858552 0.03285247 -0.030802274 0.12340899 0.14025427 0.15989256 -0.089216724 0.038588263 -0.14061545 0.18755326 0.095213935 0.0013497722 0.082043424 -0.023433786 -0.2122101 -0.01430143 -0.08398005 -0.013458209 -0.009350313 -0.11256453 0.12198539 -0.008325632 -0.017762495 0.23063074 -0.03340255 -0.005555744 -0.075714335 0.19716592 0.21898748 0.06973767 0.2705313 -0.14624457 -0.16610841 0.09301039 0.13343658 -0.09647653 0.15495495 -0.23481649 0.12357538 -0.2015125 0.18966414 -0.06488622 -0.04613218 -0.031202234 0.007159134 0.005381707 -0.14700487 -0.19526857 -0.17941949 -0.09576967 0.076152846 0.08654619 -0.15737411 0.03609946 0.021728737 0.06443218 0.056605686 0.05138919 -0.12325241 -0.19474089 0.056729276 0.12273643 0.12193133 0.055633757 -0.20060377 0.08457941 0.21545549 0.10326889 0.010303045 -0.018707287 0.119305335 0.36140934 -0.03586475 -0.40713573 -0.2681094 -0.07707229 -0.15150537 -0.09441615 0.010284476 0.03640932 -0.29858735 0.30975756 0.2780172 0.05371121
 7 | r -0.20327458 -0.23636568 0.09384572 0.031158773 0.2289692 0.0303982 0.16613124 -0.10020323 0.31345224 -0.14411947 0.10943919 0.0010986998 0.11934069 0.0040077274 0.055554245 0.15027158 0.024243388 0.0668375 -0.27703223 0.22066317 0.23936762 -0.14624953 -0.11353286 0.29784188 -0.30960435 0.41315463 0.076330535 -0.21461122 -0.2324101 0.15791667 -0.33390176 -0.071163915 -0.18273494 -0.3676166 0.055333722 -0.16521733 0.06673048 0.073422216 0.13395134 0.26863196 0.16332804 0.03315188 -0.109662294 0.18849386 -0.11014484 0.25272954 0.2304268 0.104182936 -0.067244954 0.022285644 0.3289033 -0.22867307 0.08606067 -0.35269985 0.0996878 -0.07968207 0.2624428 0.14367388 -0.11171791 -0.24831973 0.09861369 0.11365819 -0.30176517 -0.17232881 0.026358908 -0.03730847 0.20284083 0.008373554 -0.22996326 0.0010437907 -0.23413369 0.07686755 -0.036691856 0.14172187 0.23351674 0.01841283 -0.07065376 -0.10166102 -0.057834752 0.00158888 -0.06887049 -0.099017456 0.20875743 0.16159712 0.040514786 0.16780278 0.18985543 0.29391092 0.0068786386 -0.28702593 -0.19144356 0.20293295 -0.0945797 -0.14578435 -0.019351669 0.21579725 -0.20739892 0.26606876 0.24787079 0.23119392
 8 | t 0.051571563 -0.044262588 0.19267252 -0.071697436 0.02582553 -0.09823833 0.29404947 -0.12359169 0.08968029 -0.037464995 0.15589008 0.22228947 0.13961896 0.16822383 0.083594844 0.119774796 -0.122996986 0.021110581 -0.2280308 0.085657895 0.19839115 0.13722587 -0.025366398 -0.08281521 -0.04566757 0.18459238 0.046905044 0.07509397 -0.118559904 0.1488687 -0.27900502 0.08711708 -0.035022885 -0.09289099 -0.012177606 -0.14905247 0.1327146 -0.07145163 -0.11523198 0.02790688 0.04719339 0.1168076 -0.11890315 0.06262999 0.0504405 0.30570874 0.020443667 -0.0321049 -0.045151446 0.22021271 -0.040174533 -0.13789885 0.15037419 -0.09608593 0.24317545 -0.108704746 0.38741362 -0.2926996 0.18834542 -0.08676612 -0.060467556 0.04454685 -0.10190365 -0.0539734 -0.17071685 0.038140904 0.1986577 0.19001889 -0.08690834 0.030704757 0.19136818 0.003452501 0.02443334 -0.23380738 -0.18563004 -0.2318813 0.20030041 0.15944844 0.155246 0.16214003 -0.3249658 0.16820632 0.016981067 0.17361481 0.016795741 -0.051644772 0.016251558 0.28947088 -0.25431404 -0.16662377 -0.3515427 0.041902866 -0.11543044 -0.0009131393 -0.0049342373 -0.14659387 -0.18363301 0.2578201 0.13474533 0.043137334
 9 | n -0.15715796 -0.03269516 0.19859166 0.1088939 0.15357296 -0.3610224 0.21020764 0.10423414 0.12183934 -0.19091098 0.29871312 0.26966166 0.0021514257 0.02403689 0.13967694 0.3098722 0.017385919 0.05639457 -0.08953687 0.16507633 0.23461512 -0.09380742 0.06877612 0.1392994 -0.0985702 0.2680957 -0.06640008 0.10189534 -0.041834775 0.07114711 -0.14174774 -0.035780665 -0.02987549 -0.16890872 0.18154193 -0.19957766 -0.06486757 0.12269939 0.020898655 0.16835368 0.15734853 -0.031830646 -0.17632274 0.02092848 -0.024912698 0.3019352 0.17609532 0.0895882 -0.2176814 -0.013572483 0.31280953 -0.0025129502 0.048557546 -0.31267792 0.15243411 -0.17912088 0.10530732 -0.1536216 -0.042834997 -0.07519403 0.21967317 0.12131074 -0.044528674 -0.09535464 -0.28027046 0.10899969 0.090873145 0.27388793 -0.043249074 0.073419765 0.05525186 0.04717896 -0.1870437 -0.15909097 -0.10631558 -0.15548371 -0.012766979 0.2547215 -0.010949443 -0.0011207255 -0.061634373 0.031083426 -0.028094629 0.0808704 0.06582945 -0.10414855 -0.16068223 0.17076033 0.01232191 -0.3292273 -0.17298225 0.1287373 -0.07234306 -0.13679287 -0.05272441 0.14064497 -0.2483139 0.33431894 0.26566845 0.18120767
10 | l -0.048551634 -0.19914386 0.08222875 0.051749595 0.025738852 -0.10973121 0.23748897 0.0019035722 0.03639035 -0.047182072 0.32025215 0.18143576 0.11012035 0.18656953 0.08596237 0.107006304 0.011547315 0.11148613 -0.06996193 0.16087556 0.20308693 -0.06203771 0.014369187 0.029327983 -0.05201956 0.24572338 0.048274938 0.019238722 -0.11653762 0.08461819 -0.086142786 0.00086884294 -0.096784584 -0.042108785 0.14708292 -0.17394486 0.06557134 0.10651491 -0.041283846 0.18441342 0.18713129 0.11261896 -0.09630408 -0.0010789753 -0.0053827087 0.23745346 0.14494638 0.16481566 -0.035824053 -0.032790285 0.1491452 -0.1259114 0.06675228 -0.36129922 0.12302404 -0.10279252 0.14231992 0.09799287 -0.26159072 0.05925392 0.10463807 0.23049992 -0.096925534 -0.1846797 -0.15399337 0.03013857 0.20515685 0.07961617 -0.12642978 0.10915804 0.01608021 0.10146244 -0.024828238 -0.028264046 -0.1647261 -0.077488504 0.11618961 0.08667462 0.019568583 0.05047856 -0.08995657 0.0225545 -0.029513646 0.027166415 0.0334521 -0.07422752 0.05197258 0.19981116 -0.019706763 -0.2139917 -0.27181867 0.10430023 -0.01795804 -0.058455013 -0.13021705 0.08123328 -0.052179098 0.27725515 0.053214464 0.22276893
11 | c 0.08000457 -0.08330582 0.2453407 -0.2931559 0.028822 -0.30987367 0.062729925 -0.083584264 0.10434201 0.020334028 -0.01850615 0.15778394 0.1939753 0.13978748 0.15650219 0.09595791 -0.062376272 -0.035523467 -0.07557824 0.10982105 0.22416735 0.076135226 0.02041246 0.104193695 -0.0072231013 0.12848441 0.12529355 0.01109697 0.13087094 -0.00732543 -0.28233492 -0.072510764 -0.0037715344 -0.071858674 -0.09532995 -0.13438103 0.15744628 -0.050115816 -0.037386745 0.109830126 -0.007858402 0.14937218 -0.02855836 0.12448153 0.29072925 0.06363767 0.1306279 -0.122644186 -0.02280697 0.09380472 0.009748608 -0.00087719946 0.046217233 -0.071330614 0.105685316 -0.23214665 0.23194784 -0.03208675 0.18516053 -0.09387045 -0.053946797 -0.062960155 -0.12759787 -0.14678265 -0.15690675 -0.13926685 0.12854733 0.02502689 0.01951735 -0.059447367 0.017551314 0.10849616 0.06442824 -0.1638584 -0.18980858 -0.043902557 0.09503061 0.042790785 0.11453516 0.13607611 -0.21599218 0.080438614 0.20710008 0.23613109 -0.19927496 -0.09485832 0.13768028 0.20073548 0.047885954 -0.26737127 -0.15453348 -0.15976828 -0.1515766 -0.013798511 -0.13175327 0.022583863 -0.17367665 0.3059841 0.08350704 -0.03470979
12 | m 0.044085637 -0.15230167 0.13428076 -0.01328807 -0.012635109 -0.15117355 0.20662898 -0.050564382 0.04522055 0.008407537 0.21055092 0.14035371 0.1614966 0.18903442 0.19000924 -0.0060392763 -0.014484248 0.06842681 0.015658347 0.035436638 0.265111 -0.07612328 0.03803459 0.11329411 -0.11816754 0.36420166 0.2793012 0.008441783 -0.15503058 0.06341103 -0.11931916 -0.10939805 -0.00893742 -0.108218364 -0.037419055 -0.1944905 0.12317719 0.14205419 -0.03279027 0.21149494 0.11442936 0.19663261 -0.096492596 0.08218529 -0.027971566 0.07933011 0.014139461 0.16922581 0.021938158 -0.007847776 0.13709748 -0.12844145 0.024598267 -0.3352289 0.10471103 -0.10531424 0.21474554 0.09260015 -0.1821319 0.02180027 -0.040729176 0.19193923 -0.054027162 -0.04968127 -0.24506718 -0.0049960897 0.2707473 0.036457825 -0.023052951 0.10886628 0.05585261 0.09242132 -0.102869816 -0.014259061 -0.23200555 -0.0034001905 0.1429071 0.08826313 0.07309256 -0.045844533 -0.16386724 -0.0034340834 0.01733052 0.05073898 -0.017130718 -0.15897289 0.069454804 0.08487128 0.01977906 -0.106472515 -0.2629216 0.05996235 -0.1428506 -0.012611174 -0.25503466 0.04196877 0.054529205 0.28759983 0.05229882 0.1815697
13 | d 0.017383365 -0.13716057 0.2949029 -0.06199102 -0.08093327 -0.18764783 0.18196931 -0.07593696 0.0760895 -0.1295881 0.24118142 0.2014097 0.19134977 0.1678178 0.09810296 0.18012947 -0.10538177 0.046037067 0.0015852929 0.04068107 0.2313376 -0.048293427 2.8288754e-05 -0.05518404 0.009440827 0.23652285 0.14618093 0.047574192 -0.009265087 -0.021858027 -0.13267243 0.013915686 -0.13707317 -0.034453746 0.0815704 -0.19007906 0.12735792 0.054074932 -0.02511241 0.15683855 0.07042538 0.08108197 -0.027077124 0.08774708 0.025747458 0.09877187 0.13808525 -0.015637197 -0.0667053 0.06207111 0.09061058 -0.0707796 0.06739539 -0.10915792 0.13443996 -0.1593033 0.24231735 -0.10487516 -0.062210716 0.09420475 -0.061529275 0.20280473 0.07039333 -0.18639512 -0.32782438 -0.04117493 0.13286397 0.07464029 0.041200798 0.08440293 0.20271082 -0.010833351 -0.06288818 -0.14645678 -0.2694866 -0.14088172 0.1660197 0.13049613 0.045604825 0.0490851 -0.16395184 0.045036666 0.09743542 -0.02892811 0.0551652 -0.039587248 0.026036907 0.063418694 -0.08332066 -0.2575671 -0.21684201 -0.036762215 -0.0073884944 0.019529132 -0.14741892 0.06451546 0.10173043 0.27500647 0.105756894 0.15968797
14 | u -0.0073406594 -0.028009437 -0.17585872 -0.107637174 0.01027438 -0.023028746 0.1269347 0.111341774 -0.070842884 0.17183873 0.032443807 0.2068273 0.10692973 0.23336956 -0.08387705 0.18644425 -0.13910075 -0.09687045 -0.04263326 0.0030651574 0.22196819 -0.055053424 -0.15223096 0.06929536 -0.06537478 0.0436268 0.08306868 0.07537791 0.010320458 0.21634068 -0.1347581 0.10239488 0.1128446 -0.1567407 -0.007041023 -0.07048346 0.03133988 0.007886676 0.027175495 -0.06903642 -0.08509622 0.0583257 0.22452044 -0.0018463854 0.08749768 0.27111855 0.04505391 -0.18933856 -0.009482135 0.1466052 0.051436044 -0.15379842 0.20748207 0.009357561 0.011487865 -0.25649017 0.07785343 0.022872359 0.0094504235 0.05409313 -0.020444114 -0.06928668 0.098536044 -0.06860429 -0.13906056 -0.046036597 0.13984309 -0.087815635 -0.020382917 -0.20997903 -0.06458964 0.018012868 0.30056703 -0.11324126 -0.29543832 -0.03672114 -0.0069154273 0.11037812 0.08863553 0.20385626 -0.41320607 0.28151765 0.17076047 -0.010347054 0.028610142 -0.11256742 0.12744789 0.15663426 0.14239341 -0.0073220045 -0.18215615 0.06915716 -0.14931116 -0.1470049 -0.35397407 -0.0033556526 -0.21471384 0.28188688 0.057611402 -0.24946018
15 | p 0.07161328 -0.18414274 0.124919794 -0.2135404 -0.014137368 -0.24078125 0.23159668 -0.15286137 0.01840469 0.12833591 0.11921622 0.1747227 0.26824084 0.21670537 0.28430748 0.03754729 -0.11138533 -0.038502723 0.057748184 -0.031249322 0.28353658 0.14727202 0.08045136 0.085289866 0.05558851 0.2784051 0.30122817 0.14110534 0.040095095 -0.0065152207 -0.09707669 -0.097474515 0.13208471 -0.06749392 0.0009859109 -0.21851803 0.12911223 0.086672634 -0.081714064 0.0845177 0.029532766 0.08595427 -0.09070171 0.15308212 0.18268627 0.09285502 -0.11547787 0.027929712 0.04343557 0.020659618 0.065448344 -0.07095329 0.089826666 -0.23045984 0.1349074 -0.031824708 0.18186863 -0.052882556 -0.05481395 -0.028017564 -0.04275991 -0.10879742 -0.031883456 0.091686316 -0.2179221 -0.036103304 0.22455652 0.060402837 0.09004586 0.04117303 0.051705968 0.0116814235 0.062016785 0.06862172 -0.261973 -0.053262074 0.053808287 0.20308614 0.3252411 0.05030674 -0.25896332 0.05991141 0.115146324 0.23390774 -0.12036977 -0.25397167 0.086755626 0.13543075 0.06880622 -0.10234195 -0.26424524 -0.0261892 -0.23767613 0.016381156 -0.21006028 -0.06903735 -0.05201222 0.28107873 0.1360687 -0.16828234
16 | b -0.104696065 -0.07081813 0.19224901 -0.20937015 0.05933781 -0.25705627 0.21469428 -0.18465026 -0.06288924 -0.0029247876 -0.15480913 0.23668596 0.2778858 0.11907464 -0.012278035 0.33900526 -0.01676178 -0.17450097 -0.12535073 -0.029635707 0.36625585 0.1223491 -0.14504716 0.22621936 0.022769647 0.15206039 -0.008880497 0.113687225 0.21841896 0.07502701 -0.11102672 0.065996744 -0.11003207 -0.18173969 0.103972204 -0.18277985 0.016486302 0.07203452 -0.012486204 -0.03755638 -0.02173766 -0.063362524 0.02049098 0.06001514 0.10021142 0.13000533 0.010651847 -0.025989538 0.018391406 -0.04659886 0.117142454 -0.0346855 -0.03494893 0.11306936 -0.038175322 -0.29995254 -0.03789865 0.10137964 -0.08394757 0.26574087 -0.020555396 0.040980563 0.06782771 -0.015778868 -0.19354145 -0.12546737 0.0937983 0.17950109 0.075521976 -0.19649467 -0.16502936 -0.16004442 0.20488039 0.17871623 -0.15691493 -0.18356697 -0.089240305 0.063799255 0.09714829 0.1172592 -0.22972311 -0.053596426 0.20857319 0.2280078 0.11659212 -0.09508251 0.09388824 0.011119666 0.1373837 -0.15320948 0.144054 0.08480063 -0.08527917 0.04873432 -0.20346624 -0.079291925 0.20172316 0.16906442 0.15233266 -0.16405763
17 | g 0.08890391 -0.25260854 0.26048005 -0.07638261 -0.003951663 -0.15013239 0.14083925 -0.007170578 0.03958312 0.074520186 0.1114248 0.27694714 0.2345191 0.27585545 0.09477804 0.11210592 -0.1670766 0.07934381 0.019158963 0.16975617 0.17504759 -0.045596484 0.0031231304 -0.0157248 -0.088738255 0.32086805 0.134235 -0.0036868881 0.11323218 -0.04317198 -0.20103405 -0.051753093 -0.031027427 0.007400271 0.034389198 -0.16018811 0.1913402 -0.1699395 -0.06616845 0.110246144 -0.02580875 0.0273535 0.16841222 0.04019162 0.22346082 0.023787402 0.12806514 0.0035551253 0.05098447 0.14723533 0.030489128 0.06401364 0.07712204 -0.13949172 0.057822578 -0.17066292 0.20283782 -0.05250878 -0.080606714 -0.057363413 -0.0042529893 0.035941854 -0.023998281 -0.22075866 -0.09693966 -0.062186863 0.0851893 0.0710045 0.025248092 0.060380198 0.033611633 0.17308885 0.10703289 -0.11702216 -0.25705183 -0.09604693 0.1314916 0.09430825 0.021011824 0.054558106 -0.20141174 0.06540118 0.11449064 0.06259666 -0.058402438 -0.17886415 0.18739349 0.094575256 -0.012264629 -0.13857798 -0.19965805 0.037693426 -0.09540148 0.016736949 -0.11653018 -0.11143459 -0.0803067 0.12832536 -0.027294705 0.028843135
18 | h -0.08521921 -0.43004262 -0.0021144233 -0.13937066 0.005849196 -0.08581845 0.13572668 -0.032554295 -0.06381067 0.047157533 0.25402844 0.1734149 0.09236836 0.2269508 0.13458976 0.20560902 -0.13976839 0.07258417 0.059327792 0.25786886 0.18601635 -0.030350339 -0.05299338 0.14910506 -0.058754727 0.31133848 0.20966767 -0.004724338 0.028380662 0.035884004 -0.09118581 -0.18648262 0.00580773 -0.13068683 0.19442606 -0.20906436 -0.07478274 0.11055508 0.0381673 0.03384025 -0.06269051 -0.022189246 -0.04817212 0.055856902 0.014001538 0.19950208 0.01544642 0.036800805 0.090606295 -0.025884375 0.1979778 -0.13444613 0.21743447 -0.21799368 0.059929416 -0.04718893 0.15091193 -0.0012309143 -0.18942167 -0.02376919 0.14182714 -0.08343895 -0.062540546 -0.085126966 -0.06445169 0.11243091 0.20191188 -0.02417418 -0.08138987 0.06261451 -0.02344868 0.09126904 0.08156025 0.15367097 -0.16060902 -0.08620386 -0.07366721 0.17839305 0.07410673 0.049605362 -0.108898334 -0.0040154723 0.108226225 0.05885973 -0.0034900478 -0.095353805 0.14464733 0.19016711 0.05588286 -0.22560573 -0.2297222 0.041668776 0.041907966 -0.09816554 -0.13897483 0.13410582 -0.25734228 0.15941775 0.12405148 -0.06881953
19 | k -0.09215546 -0.24799444 0.023569463 -0.07086082 0.035207786 -0.06996515 0.15231064 0.011192847 -0.08132726 0.08835065 0.12605287 0.18704653 0.169136 0.22775751 0.03709306 0.1858321 -0.06178588 0.07635303 -0.020812545 0.15504664 0.16563979 -0.004367251 -0.042248104 0.05405794 -0.0071436176 0.3036857 0.08139407 0.049530808 0.07760046 -0.013812358 -0.07118335 -0.051371977 -0.050560176 0.0016788488 0.17559248 -0.14872405 0.066718094 -0.041319244 -0.030021295 0.039744146 0.023707373 -0.030719196 0.09004138 -0.0082294755 0.12048634 0.1302152 0.10688799 0.025753187 0.052864775 0.034978934 0.12825412 -0.015645921 0.13837712 -0.19301565 0.022279933 -0.14168285 0.051956333 0.039628316 -0.16015643 0.07203872 0.0875609 0.036005765 -0.027894925 -0.1381202 -0.08372142 0.0018328312 0.14934878 0.07644203 -0.076377496 0.016239043 -0.050764423 0.08129476 0.18336894 0.05499936 -0.21597047 -0.14055434 0.0052526584 0.13710757 0.030599942 0.064412616 -0.12328008 0.09640631 0.063444495 0.08447667 0.03470835 -0.14717673 0.12075886 0.15262917 0.031789325 -0.19177091 -0.15516984 0.088593245 -0.06020588 -0.06793137 -0.12204909 -0.0470094 -0.155208 0.17068525 0.059937343 -0.019082757
20 | y -0.03304683 -0.24198124 0.10167503 -0.13710599 0.04410612 -0.08346686 0.11977929 -0.068828836 -0.082509235 0.14051656 0.09369337 0.22345321 0.20914675 0.27898887 0.044384174 0.10451268 -0.13117087 0.040781535 -0.15194896 0.21966021 0.21917139 0.18355171 -0.15484741 0.049834482 -0.090931945 0.18305424 -0.024018737 -0.10375332 0.24082468 0.024596108 -0.26619908 -0.064383775 -0.11641643 -0.115010984 0.1428091 -0.20327704 0.11970474 -0.0404396 -0.14467633 -0.028246716 -0.05099949 -0.01716773 0.08336503 -0.055336736 0.18651785 0.02662152 0.013132311 0.0017866433 0.029918425 0.064123385 0.012834428 -0.0156074185 -0.022298945 -0.13634722 -0.03345883 -0.19295157 0.051964726 0.064791106 -0.11086943 0.036432005 0.009650755 -0.016004488 -0.10274735 -0.14189826 -0.09968405 -0.060301706 0.13413294 0.07714425 -0.07793404 -0.11132174 -0.086361594 0.11761092 0.21430156 -0.029785164 -0.18466778 -0.11574724 -0.032593075 0.068989605 -0.0028003578 0.17539974 -0.17489694 0.06118515 0.20741248 0.12677209 0.016121365 -0.074178405 0.16773063 0.19094107 0.069364324 -0.23906209 -0.15187147 0.0029179696 -0.020240748 0.08374939 -0.060633272 -0.08256757 -0.001965709 0.26923022 -0.0027070218 -0.039023764
21 | f 0.014668905 -0.107221834 0.16447137 -0.26872578 0.017281312 -0.299774 0.10093116 -0.057283532 -0.08774917 0.12902893 0.12182817 0.23479006 0.2242874 0.24042939 0.18308444 0.075035445 -0.01800861 0.1616582 -0.05378137 -0.046837535 0.22277457 0.10698119 -0.03475638 0.1396006 0.16319972 0.19215883 0.07374807 0.027177125 0.05407101 -0.023786176 -0.083753444 0.039258666 0.09059716 -0.0067064073 0.043450657 -0.18125819 0.08067583 -0.06522509 -0.094865985 0.067238435 0.01954378 0.028229948 0.089013405 0.031953923 0.0092187105 0.12823088 0.03614065 0.044994846 0.007025591 -0.021992633 0.14171796 -0.02309716 0.18528858 -0.20936075 0.13768436 -0.07939527 0.15905213 -0.015559245 -0.058800597 0.04461056 0.07462226 0.04248818 -0.049244817 -0.026953027 -0.21038863 0.017589383 0.17225964 0.052272584 -0.009062194 -0.10318616 0.040749926 -0.00905727 0.14774683 -0.058274984 -0.3124769 -0.12755331 0.04914195 0.16960758 0.15783292 0.07482018 -0.2075715 0.16547477 0.060000326 0.1114354 0.017000498 -0.1389587 0.015033586 0.21574683 -0.01623313 -0.21697626 -0.25328428 0.032935794 -0.17343922 -0.020569284 -0.24502872 -0.032603648 -0.08584067 0.282617 0.14275417 -0.004787791
22 | w -0.032410894 -0.22031659 0.031349532 -0.069770545 0.040971268 -0.0416638 0.27467644 -0.024408266 0.00839984 0.06603507 0.06471432 0.16886689 0.1117572 0.23479089 0.03883411 0.052291762 -0.10167857 -0.022046534 -0.08103379 0.25402072 0.26447392 -0.1496173 -0.011135708 0.11832451 -0.24274136 0.3273302 0.19808583 0.0500611 -0.02722104 0.17388767 -0.27209273 -0.11162222 -0.058076236 -0.1652996 -0.041349445 -0.15083158 0.10840928 -0.021151204 0.0009584433 0.008162052 -0.024023416 0.11074994 0.063289516 0.011398587 0.1524441 0.22021641 0.011773423 0.04346426 0.1301035 0.22987771 0.033823915 -0.03388422 0.08163596 -0.045810744 0.046392746 -0.20378843 0.26473555 -0.035665125 0.050475992 -0.06917049 -0.081634745 -0.017139712 -0.049723003 -0.14858711 0.005105913 -0.06717927 0.23580621 0.1280029 -0.019613322 0.032897525 -0.10110632 0.20694834 0.0799768 -0.051035676 -0.22436418 -0.08272468 0.108925454 0.013583037 -0.0075046574 0.08920874 -0.20445602 0.018522073 0.03331418 0.18777837 -0.10154185 -0.20173359 0.23025793 0.06616148 -0.04055833 -0.011391936 -0.08417345 0.06425156 -0.10467157 -0.061598323 -0.20367981 -0.10008356 -0.10821719 0.0958135 -0.03779756 0.023209685
23 | v -0.08403122 -0.16335614 0.08348566 -0.19038562 0.04106152 -0.32183415 0.18678573 0.029350026 -0.042176303 -0.017399698 0.19469562 0.22517149 0.094950974 0.24112675 0.11056021 0.11713126 -0.04177587 0.114533246 -0.03556987 0.16446503 0.34059963 -0.24100105 -0.04950593 0.26012352 -0.012684174 0.3080495 0.15489182 0.068389244 -0.01437787 0.15035848 -0.1740834 -0.06955549 -0.018047847 -0.15902206 0.042318694 -0.1373041 0.064717636 0.028410342 0.07732284 0.07361216 0.0446535 0.070599206 0.115606144 -0.039243985 -0.0059865364 0.37689275 0.09760193 0.108329736 0.07596615 0.094623886 0.23176576 -0.06295208 0.24806896 -0.15328482 0.15757313 -0.12515256 0.31348795 -0.03184009 0.024533885 -0.021165079 0.032174032 0.084867835 -0.028839177 -0.11145576 -0.095508486 -0.012098987 0.2173698 0.010073884 -0.053183615 -0.0449915 -0.027361298 0.10048844 -0.025326002 -0.15006424 -0.27550948 -0.036014616 0.09190896 0.023400182 -0.011278867 0.09508371 -0.07554621 0.008000784 0.003185623 0.0996707 -0.08002428 -0.19750333 0.10825231 0.026937768 -0.0031855034 -0.13091448 -0.14388014 0.06288034 -0.10000085 -0.12880467 -0.3304057 0.0855 -0.12943777 0.16789454 0.10248381 0.13539939
24 | z 0.04188836 -0.25111985 0.1850754 -0.12807718 -0.02802349 -0.21350771 0.18805392 -0.029564725 -0.08257791 -0.00015464751 0.19024524 0.2679126 0.12381543 0.23373671 0.0101947505 0.19389139 -0.14537172 0.006021071 -0.080028825 0.24746802 0.24905951 -0.040212262 -0.19987018 0.058766637 -0.080590755 0.2596077 0.04789264 -0.0037278347 0.14630868 0.1496338 -0.13641201 -0.046185575 -0.09200375 -0.09024734 0.13797255 -0.09111029 0.060843553 0.025527883 -0.028364537 0.019342938 -0.054642387 0.042264167 0.04133608 -0.15996055 0.07232787 0.20200618 0.029764902 0.107544094 0.06997305 -0.004130755 0.00482789 -0.07609163 0.04558279 -0.10975031 0.008419858 -0.24843715 0.1489674 0.11128129 -0.28686857 0.18100539 0.06082298 0.17388417 -0.060372423 -0.10895816 -0.08935944 0.026322888 0.12919392 -0.017875263 -0.1479803 -0.024543645 -0.02115278 0.09387045 0.087826714 -0.02891458 -0.16351114 -0.09143618 0.07506968 0.03647387 -0.055852633 0.10994188 -0.17940353 -0.10731756 0.11150333 0.034309812 0.024537541 -0.15771544 0.17959473 0.061527524 0.12555854 -0.1380637 -0.15631582 0.056146108 0.096366204 0.060364068 -0.114323206 -0.0599136 0.020379739 0.10167063 -0.082642525 0.018131457
25 | x -0.07195969 -0.10384891 0.042889006 -0.18151927 0.058367416 -0.15749595 0.18476525 0.0072446307 -0.13384558 0.099668965 0.07753233 0.1848738 0.018788917 0.16755322 -0.020258257 0.10696196 -0.027385132 0.056112878 -0.24158946 0.19012064 0.19634788 -0.040901687 -0.19519567 0.15179941 -0.13403729 0.31337887 0.032950733 0.015265252 -0.0333566 0.2214465 -0.15526561 -0.025386147 -0.009238737 -0.13649464 0.034074273 -0.04941042 -0.024661332 -0.060978446 -0.041046545 -0.018040199 -0.11073333 0.070340194 0.0476421 -0.16103448 -0.070560925 0.22872673 -0.047350116 0.094769955 0.13895798 0.069424845 0.053643215 -0.07156436 0.1256253 -0.07155963 0.021339232 -0.30526027 0.15205444 0.04749555 -0.114309244 0.1608413 0.1001255 0.18149936 -0.098815694 -0.009411489 -0.060829867 0.08313096 0.25675836 0.080985606 -0.22605076 -0.08516066 -0.08880107 0.09695035 0.14704269 -0.0479482 -0.21674848 -0.18541498 0.07627307 0.086630195 -0.090367906 0.07555553 -0.26950395 0.0037829576 0.0062968684 0.16311868 0.046255223 -0.18305984 0.12323938 0.17274247 0.022727482 -0.14139862 -0.11391154 0.14688766 0.020521311 -0.052950837 -0.23878042 -0.14168318 -0.16787106 0.114123955 -0.06992749 0.0010571572
26 | j -0.11344891 -0.18714751 0.07144669 -0.15941001 -0.030485662 -0.053612985 0.122495055 -0.05351344 -0.16589901 0.13324116 -0.029648086 0.17084533 0.17010163 0.17602259 0.09005554 0.11342317 -0.12043303 0.14998966 -0.19111405 0.08457623 0.23994367 0.13032182 -0.25174272 0.2295933 -0.13076717 0.52357566 0.107177734 -0.21178748 0.12441281 0.080550276 -0.19739726 -0.09788506 -0.09045887 -0.1540241 0.05031309 -0.22327968 0.044423066 -0.08468933 -0.0853702 -0.10075579 -0.14919342 -0.075450525 0.096049815 -0.07740641 -0.09985836 0.044179592 -0.022841303 0.07250386 0.10696339 -0.040484484 0.11261889 -0.063154705 0.049994186 -0.070493095 -0.02597128 -0.16760316 0.19162714 -0.0047643785 -0.07365538 0.018324474 -0.017169995 0.019090423 -0.0963545 -0.039931536 -0.10197194 0.06765312 0.18561673 0.08432698 -0.097390726 -0.14222638 -0.06651065 -0.0027934965 0.21004462 0.12962951 -0.14926945 -0.18246944 -0.063895136 0.074719615 -0.021651028 -0.019405955 -0.2441432 0.028069695 0.15518633 0.08285815 0.15891278 -0.09701887 0.17385258 0.13913815 0.04025835 -0.13942638 -0.12102695 0.16578546 -0.06316902 0.09375757 -0.13349275 -0.15166284 -0.0199566 0.16333106 0.024514064 -0.047834076
27 | q 0.08517465 -0.18712334 0.1849958 -0.47358313 -0.07627318 -0.39042506 0.117762946 -0.18213409 -0.098908044 0.11216427 0.080832936 0.16040829 0.1665408 0.14162172 0.16428104 0.10678668 -0.11821475 0.064479135 -0.19709395 0.07359679 0.31345415 0.16939752 -0.34538767 0.20504838 0.12362731 0.29442075 0.03173975 -0.035841852 0.25504336 0.08791239 -0.15142202 -0.010957913 0.048224624 -0.09918974 0.007594991 -0.07608966 0.057423815 -0.012034671 -0.015818508 -0.030550223 -0.16893914 0.022423409 0.004079535 -0.11861544 -0.0067758868 0.21070537 -0.08492238 0.05501255 0.10049278 -0.080453895 -0.0008096878 -0.08249105 0.16347167 0.026842758 0.13851844 -0.22064017 0.34080777 0.054017182 -0.20052475 0.15195817 0.06476622 0.069980755 -0.19358444 0.06422441 -0.056049485 0.039506163 0.18571527 -0.10324722 -0.17132294 -0.19490315 0.016792819 -0.07952712 0.15284674 0.09837393 -0.113749884 -0.15974198 -0.014858386 0.07640661 0.09495512 0.12663317 -0.28643838 -0.1061391 0.18800278 0.15553932 0.05046154 -0.1310227 0.21158104 0.12388787 0.088728495 -0.2424983 -0.14745952 0.06996265 -0.07326424 0.12150764 -0.15159217 -0.15829095 -0.009338701 0.06766454 0.06282346 -0.101689786
28 | - -0.11717634 -0.1511274 -0.06912113 0.110410266 0.21339259 -0.19827528 0.222841 0.27835524 -0.071811914 0.3743894 -0.07306852 0.41545373 -0.09557201 0.09840342 0.07054609 0.35558453 -0.041078966 0.36687267 -0.28313217 -0.007988353 0.24761288 0.27538925 -0.4406256 0.43250605 -0.63512784 0.5778142 -0.01618778 -0.2395711 -0.12694097 0.28449214 -0.19538042 -0.027429234 0.18842965 -0.29880726 -0.19393098 -0.051112987 -0.085430816 -0.29605362 -0.07441005 0.30162883 -0.45688003 -0.0719094 0.13268863 -0.025598688 -0.2654899 -0.022221206 0.06752689 -0.0013421029 0.078556836 -0.028370688 0.48761392 -0.16265678 0.1966587 -0.30011684 0.009538995 -0.55096006 0.2632322 0.0041703377 -0.45863858 0.0184351 0.27998433 0.28525308 -0.15600659 0.019671477 -0.38906148 0.19986287 0.18396893 0.0697025 -0.38594452 -0.16953602 -0.08621148 0.003648659 -0.001522291 -0.019719683 -0.24366117 -0.2920398 0.017696729 0.31041032 -0.15943745 -0.12703171 -0.61213446 -0.06025422 0.37902242 -0.038015876 0.3569995 -0.28402287 0.11751555 0.2463137 0.21479304 -0.36196494 -0.3637389 0.5037049 -0.19460398 -0.05049795 -0.50023526 -0.12637198 -0.39716318 0.29907042 0.18811636 -0.014217993
29 | 1 -0.10578755 -0.79974693 1.559121 -0.2144091 1.060053 -0.16257158 0.17087193 -1.1313504 -0.26788226 -0.12751733 -1.6388621 1.5777588 0.04599483 0.9014916 -0.20720752 1.2500955 0.11656526 -0.62581056 0.23003334 -0.18279476 -0.011284278 0.8369775 0.47037265 0.29919824 -0.9729427 -0.70086855 0.9217705 -0.2889931 -0.668005 0.6214847 0.007966269 -0.25697592 -0.28041705 0.102639906 -0.29690814 -0.60025036 0.09799304 0.6236977 -0.59013313 0.19110909 -0.1142138 0.7525327 -0.7196503 0.852028 0.39560872 -0.50143874 0.34653726 -0.35126776 0.20730746 -0.67594475 0.45674738 -0.8853343 -0.5178775 -0.553561 -0.37661672 -0.881321 -0.7450074 0.7587129 -0.0968869 1.0821803 -0.14401203 0.39256543 0.12500045 -0.034077596 -1.4817541 -0.28491715 0.0375107 0.6038841 -0.24854523 0.4059825 -0.7501047 0.18315165 0.059982356 -0.28362542 -0.92431307 -0.19872396 0.51872116 -0.07960054 0.34360674 -0.11446986 -0.508664 -0.31784967 0.6231811 0.57192177 -0.62342876 -0.4277829 -0.44520327 0.84439784 0.45644072 -0.50048 -0.23198098 -1.2178423 0.96614736 -0.3221664 -0.93311435 1.0341275 0.033835806 0.9761267 -0.39482442 -0.57526934
30 | 2 -0.2119582 -0.5973319 0.691543 -0.1126316 0.9476311 0.42381942 0.42022592 -1.0399522 -0.5305956 0.19444431 -1.1866939 0.9496192 0.43689403 1.1287516 -0.47035024 0.54501605 0.27803057 -0.39775372 0.01168412 -0.2989614 0.17850615 0.8408311 0.14007647 0.18836303 -0.46075502 -0.91193557 0.90066636 -0.16927077 -0.7027362 0.846808 -0.05345008 0.1645284 -0.53604585 0.13150117 -0.2524094 -0.34147793 0.46316516 0.37626457 -0.6525121 0.24587427 0.2505632 0.7811102 -0.14881226 0.6237348 0.40854985 -0.1101658 0.27478778 -0.11403407 0.31892878 -0.32594323 0.1511594 -1.1061523 -0.13929796 -0.84624654 -0.332029 -0.37457997 -0.6753213 1.0793401 -0.11087725 1.2135605 -0.4985108 0.564934 0.09151458 -0.0075201932 -1.068123 -0.60166216 0.3580061 0.14513935 -0.45512918 0.049329467 -0.8236405 0.16156751 0.55231977 -0.16450316 -0.89239633 0.04749353 0.7303606 -0.6431723 0.48771316 0.34656194 -0.3058962 -0.063466854 0.5117423 0.6061756 -0.553581 -0.21688451 -0.015675532 0.82723856 0.3247579 -0.18717809 -0.4916446 -1.1227667 0.6335212 -0.23666503 -0.89298034 0.7245715 0.48824292 1.0768734 -0.34752417 -0.2766324
31 | 0 -0.18527375 -0.8670408 1.6205827 -0.18734664 1.2826948 -0.22097297 0.13478945 -1.106813 -0.24750154 -0.15805838 -1.995448 1.7905121 -0.08404253 1.0178026 -0.22427627 1.4397779 0.14725792 -0.6481898 0.26091656 -0.1556429 0.03553785 0.8309699 0.5262068 0.6023852 -1.377152 -0.5866043 1.0382582 -0.40093744 -0.84557486 0.7472878 -0.05261586 -0.4544033 -0.29529428 -0.031994455 -0.44514507 -0.55628115 0.15283474 0.6404921 -0.5768324 0.22272944 -0.25326023 0.80527467 -0.77449626 0.8885105 0.3703965 -0.5545779 0.4389142 -0.382733 0.36354133 -0.6903851 0.7231771 -1.0087689 -0.5564704 -0.615143 -0.478802 -1.0259113 -0.77030504 0.80119437 0.09451197 1.103789 -0.2975916 0.4148677 0.12933245 -0.024785155 -1.6952379 -0.36185893 0.07150217 0.6909213 -0.39120224 0.45117205 -0.91021466 0.20190828 -0.13610604 -0.49261656 -1.0866657 -0.13018823 0.64359486 -0.15528797 0.23335019 -0.23228072 -0.4674538 -0.46080324 0.7446005 0.6683394 -0.8222074 -0.6140208 -0.5443572 0.79438955 0.5253872 -0.4667565 -0.22347988 -1.3115349 1.0814224 -0.4537901 -1.2065952 1.245113 -0.102546416 1.0353687 -0.42008886 -0.6167847
32 | 3 0.062847674 -0.84112215 1.3454363 0.14378679 0.90019566 0.08388339 0.10082598 -0.7911489 -0.2739435 -0.017696038 -1.1061938 1.5360744 -0.08778637 0.95053583 -0.2798377 1.0097679 0.04561945 -0.48954812 0.23978157 0.20211041 -0.08889525 0.59537923 0.54449475 -0.0024111732 -1.1060934 -0.8095228 0.6695251 -0.26606962 -0.6658734 0.5257598 -0.17803638 -0.33210376 -0.33385703 -0.02369803 -0.11986225 -0.5787453 -0.036332157 0.5480639 -0.5612368 0.2449682 -0.06208074 0.77734745 -0.6516375 0.6626688 0.44793725 -0.5225205 0.26436916 -0.2544453 0.15684128 -0.39837372 0.2264045 -0.6934933 -0.631495 -0.5207933 -0.3889559 -0.78397036 -0.77215785 0.62013745 -0.11695783 0.723455 0.013630283 0.336641 0.16599578 -0.2533914 -1.1231133 -0.1711439 0.041626543 0.5505953 -0.08258548 0.48348165 -0.5360852 0.48903716 -0.12527129 -0.4869888 -0.8330187 -0.052322492 0.48560733 -0.026142318 0.039476298 0.061190367 -0.36688527 -0.18523495 0.41112128 0.294946 -0.50734186 -0.26942933 -0.37223637 0.6686348 0.31825653 -0.40790707 -0.21185242 -0.9285066 0.97913027 -0.22532257 -0.72771454 0.9804896 -0.042439267 0.89283705 -0.51591766 -0.25512812
33 | 4 0.00468706 -1.0366284 1.2883334 -0.4798393 0.4383445 0.13344511 -0.028325299 -1.0928019 -0.26281327 0.3106049 -0.71297044 1.2307802 0.29020584 0.98492116 -0.1635529 0.56921816 0.16802558 -0.16155148 -0.27380836 0.15147895 -0.018917507 0.5108951 0.8168674 -0.23355083 -0.4986309 -0.905729 0.42783597 -0.37056005 -1.0829473 0.2864259 -0.1976177 -0.07908319 -0.10199122 -0.04355054 0.056332745 -0.720148 -0.0594648 0.35440856 -0.58174866 0.4277542 0.42070016 1.1244248 -0.5810994 0.9974493 0.21742354 -0.5391329 -0.060008522 -0.27107915 0.68256146 -0.3986535 0.1617751 -0.773057 -0.5722275 -0.39685985 -0.027803674 -0.60980535 -0.40361875 0.79063207 -0.39847037 0.332445 0.3368173 0.44571272 -0.0075224885 -0.55091435 -0.85924476 -0.07796531 0.45671955 0.11916632 0.24308507 0.52268827 -0.38877058 0.27384362 0.13443767 -0.40788114 -0.45745534 0.16501386 0.36642653 0.0901955 0.13025373 0.2774515 -0.62307096 0.023438789 0.5302661 0.30725887 -0.42780307 0.19273157 -0.12761152 0.6025221 0.079928465 -0.45667216 -0.3463965 -0.4816256 0.86562353 -0.14449407 -0.92035246 0.9734452 0.45318875 0.9554819 -0.73955345 -0.20014778
34 | 5 -0.005032818 -0.7781706 1.4226191 0.06405355 1.1067717 0.10779251 0.21539834 -1.0204651 -0.48189715 -0.07491173 -1.4666692 1.5709033 -0.14816223 1.0187638 -0.3603628 1.0310539 0.087484054 -0.74374855 0.33237508 0.09817871 -0.07967052 0.6104667 0.49548024 0.13625494 -1.1690906 -0.77182686 1.0388092 -0.13282073 -0.70297956 0.7754905 -0.092928186 -0.42058605 -0.33529508 0.020570952 -0.28775483 -0.5469688 -0.018758215 0.6932561 -0.60522246 0.023756873 -0.1779853 0.8395464 -0.6669379 0.62055516 0.44010413 -0.4444806 0.12259648 -0.19491188 0.22355281 -0.41482475 0.16077879 -0.8051566 -0.5869736 -0.51294285 -0.4892163 -0.8434315 -0.859976 0.7076145 0.0440562 1.085674 -0.1903859 0.33440152 0.24055889 0.07401071 -1.2608249 -0.21133333 0.13056475 0.69625086 -0.20974842 0.4509427 -0.74159163 0.51521146 0.019427033 -0.40685734 -1.0376592 -0.13387787 0.57423323 -0.13206597 0.15672483 -0.026370829 -0.3772253 -0.28300375 0.29117596 0.53216404 -0.645979 -0.53478175 -0.42702666 0.7083997 0.3765964 -0.2341346 -0.11925427 -1.2300587 1.045041 -0.303245 -0.84494776 0.9000159 -0.02968162 0.82049495 -0.58474034 -0.45008466
35 | 8 -0.11772376 -0.5146414 1.4955081 0.04355441 1.2883295 0.026199475 0.23120478 -1.2118528 -0.5272542 -0.27237874 -1.7301478 1.520999 -0.34002015 0.81796956 -0.5026194 1.1326714 0.15935992 -1.0122119 0.19612251 0.058093335 -0.23888437 0.75305307 0.3992287 0.10506654 -1.0875624 -0.83064264 0.78108793 0.0738999 -0.5010049 0.7562395 -0.006979045 -0.3423749 -0.27438608 0.1415028 -0.23825347 -0.47866672 -0.12753329 0.78758276 -0.6618211 -0.41830954 -0.3052862 0.6261187 -0.7985945 0.42728058 0.35282427 -0.3339027 -0.088733666 -0.14766943 0.14878711 -0.3660214 0.019439582 -0.5818568 -0.63112086 -0.11184055 -0.54370004 -0.93436843 -1.0174274 0.5462382 0.1945454 1.3505502 -0.21495982 0.23158312 0.19583423 0.38498375 -1.1587154 -0.11429195 0.039105777 1.1361954 -0.33168572 0.3843917 -0.83810264 0.3423863 0.21759064 -0.282915 -0.98912877 -0.5207382 0.41913113 0.02419232 0.17478113 -0.044722807 -0.39086026 -0.16546845 -0.039121136 0.7039395 -0.51655895 -0.64217424 -0.69311446 0.8524855 0.10851195 -0.15787767 0.16233203 -1.2057325 1.0458709 -0.3452589 -0.56027067 0.5204993 -0.1731191 0.5285086 -0.50803804 -0.68300027
36 | 6 0.17245592 -0.7199714 1.6009655 0.15690728 1.2253128 0.021912763 0.20462431 -1.1564372 -0.50963986 -0.19317782 -1.5705637 1.6547011 -0.28627658 1.016359 -0.3310901 0.97582567 0.09487652 -0.87657994 0.5553576 0.0018962288 -0.25903538 0.8220062 0.433533 0.023524612 -1.1668776 -0.95637566 1.0939369 -0.066335954 -0.508871 0.7290709 0.035139903 -0.45055026 -0.4079707 0.20629144 -0.34372455 -0.52988285 0.039947726 0.8729469 -0.7124255 -0.10298069 -0.31452507 0.7904807 -0.82943606 0.46017474 0.5262078 -0.51821834 0.079301685 -0.07488236 -0.008229525 -0.49097675 -0.013020426 -0.7009903 -0.695532 -0.4711951 -0.5739869 -0.77941895 -1.0186204 0.7256061 -0.034765556 1.3786511 -0.31296852 0.35135293 0.24635518 0.2988869 -1.2838235 -0.26282957 -0.01470941 0.8816571 -0.28509644 0.50407404 -0.70612615 0.57151276 -0.021951443 -0.33143616 -1.2777596 -0.27859822 0.638587 -0.15704577 0.26393893 -0.017465644 -0.32295647 -0.2729996 0.08112599 0.50081843 -0.6731042 -0.6599375 -0.5343704 0.8533517 0.30609488 -0.23363757 -0.07598908 -1.4460158 1.1007018 -0.19235013 -0.74753904 0.80143493 -0.06346256 0.7562959 -0.60821736 -0.46561015
37 | 9 0.05661885 -0.5062355 1.6503258 0.2005339 1.275088 0.079824075 0.2472423 -1.1692607 -0.51271623 -0.34454554 -1.5294156 1.5498766 -0.2816564 0.9140771 -0.40763146 1.0401801 0.017245593 -1.1274711 0.4713442 0.15678748 -0.18688859 0.44345793 0.6224141 -0.047491662 -1.0093501 -0.94634724 1.0673369 0.2145874 -0.54511803 0.7770872 -0.05160536 -0.4648128 -0.2783119 0.06906855 -0.16311786 -0.52054477 -0.12935816 1.0162395 -0.5736352 -0.3069493 -0.06148502 0.76896125 -0.8847129 0.46152678 0.51252526 -0.26402187 -0.09567986 -0.10584917 -0.05996748 -0.33022857 -0.1609085 -0.62264407 -0.6926096 -0.34317094 -0.51221687 -0.82318157 -1.1225966 0.5147303 0.32921666 1.240932 -0.20747437 0.13436456 0.3337363 0.39129224 -1.0788952 -0.104668885 0.060663465 1.0127969 -0.07769072 0.44601035 -0.7086737 0.58602536 0.018301127 -0.3658797 -1.074878 -0.2630878 0.4259342 0.022973524 0.25972196 -0.023578119 -0.2987108 -0.097568884 -0.14100713 0.68171376 -0.6408449 -0.61703795 -0.7200228 0.74578357 0.24280709 -0.063050464 0.13178428 -1.3898183 1.0034199 -0.34892216 -0.4854974 0.6592025 -0.11598933 0.6701115 -0.524995 -0.58901393
38 | 7 -0.001357535 -0.70570815 1.3918632 -0.05598001 1.1051824 0.15009719 0.13817857 -1.0679485 -0.49593243 -0.059180796 -1.3596262 1.5565625 -0.1243726 1.0269536 -0.3840184 0.8593582 0.11610363 -0.82093424 0.17629775 0.07351425 -0.1068838 0.6766813 0.579601 -0.016796844 -0.94547176 -0.90848094 0.7461384 -0.06185071 -0.6749005 0.73682475 -0.05872918 -0.2867745 -0.2419598 0.002631954 -0.106298536 -0.5555586 -0.070898615 0.70616484 -0.63972837 -0.13670534 -0.0021926344 0.8390174 -0.6339976 0.4940183 0.37186047 -0.31405008 -0.0036604763 -0.09196432 0.16012383 -0.4613792 0.040675342 -0.69383705 -0.6871208 -0.4222366 -0.4728121 -0.76215607 -1.0156205 0.81852 -0.072944306 1.1687529 -0.033617277 0.33075568 0.21260236 0.076672666 -1.1232945 -0.20462091 0.23087479 0.7885465 -0.10128734 0.35012114 -0.7988299 0.45895463 0.26459086 -0.34275994 -0.9913349 -0.15678576 0.39356905 0.020381063 0.21201086 0.06981004 -0.49053597 -0.11826196 0.14861524 0.6000743 -0.5789016 -0.37677523 -0.50228566 0.8074118 0.3415648 -0.21900114 -0.055723283 -1.0253109 0.9882148 -0.20394799 -0.791333 0.7683678 0.08031362 0.8333682 -0.6310874 -0.53310263
39 | 


--------------------------------------------------------------------------------