malicious-tfidf-ml-feat-engg.ipynb

{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"!pip install python-whois","execution_count":1,"outputs":[{"output_type":"stream","text":"Collecting python-whois\n  Downloading python-whois-0.7.3.tar.gz (91 kB)\n\u001b[K     |████████████████████████████████| 91 kB 197 kB/s eta 0:00:011\n\u001b[?25hRequirement already satisfied: future in /opt/conda/lib/python3.7/site-packages (from python-whois) (0.18.2)\nBuilding wheels for collected packages: python-whois\n  Building wheel for python-whois (setup.py) ... \u001b[?25ldone\n\u001b[?25h  Created wheel for python-whois: filename=python_whois-0.7.3-py3-none-any.whl size=87701 sha256=517fba3afeee0303eb68941d2df4d1bb100d3977ad2e8361d200a03b7c6cf838\n  Stored in directory: /root/.cache/pip/wheels/11/05/f7/895ce5a73665f77c8274a7d55e34fb3e6b4abbb9a7637e215b\nSuccessfully built python-whois\nInstalling collected packages: python-whois\nSuccessfully installed python-whois-0.7.3\n\u001b[33mWARNING: You are using pip version 20.1; however, version 20.2.2 is available.\nYou should consider upgrading via the '/opt/conda/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\n","name":"stdout"}]},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport pandas as pd\nimport itertools\nfrom sklearn.metrics import mean_squared_error,confusion_matrix, precision_score, recall_score, auc,roc_curve\nfrom sklearn.model_selection import train_test_split\nimport pandas as pd\nimport numpy as np\nimport random\nimport math\nfrom collections import Counter\nfrom sklearn import metrics\nimport matplotlib.pyplot as plt\nfrom sklearn.metrics import classification_report,confusion_matrix,accuracy_score\nimport xgboost as xgb\nfrom lightgbm import LGBMClassifier\nfrom xgboost import XGBClassifier\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\nimport socket\nimport whois\nfrom datetime import datetime\nimport time\nfrom bs4 import BeautifulSoup\nimport urllib\nimport bs4","execution_count":52,"outputs":[{"output_type":"stream","text":"/kaggle/input/malicious/malicious_phish.csv\n","name":"stdout"}]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"df=pd.read_csv('/kaggle/input/malicious/malicious_phish.csv')\n\nprint(df.shape)\ndf.head()","execution_count":4,"outputs":[{"output_type":"stream","text":"(651191, 2)\n","name":"stdout"},{"output_type":"execute_result","execution_count":4,"data":{"text/plain":"                                                 url        type\n0                                   br-icloud.com.br    phishing\n1                mp3raid.com/music/krizz_kaliko.html      benign\n2                    bopsecrets.org/rexroth/cr/1.htm      benign\n3  http://www.garage-pirenne.be/index.php?option=...  defacement\n4  http://adventure-nicaragua.net/index.php?optio...  defacement","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>url</th>\n      <th>type</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>br-icloud.com.br</td>\n      <td>phishing</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>mp3raid.com/music/krizz_kaliko.html</td>\n      <td>benign</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>bopsecrets.org/rexroth/cr/1.htm</td>\n      <td>benign</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>http://www.garage-pirenne.be/index.php?option=...</td>\n      <td>defacement</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>http://adventure-nicaragua.net/index.php?optio...</td>\n      <td>defacement</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.type.value_counts()","execution_count":5,"outputs":[{"output_type":"execute_result","execution_count":5,"data":{"text/plain":"benign        428103\ndefacement     96457\nphishing       94111\nmalware        32520\nName: type, dtype: int64"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"import re\n#Use of IP or not in domain\ndef having_ip_address(url):\n    match = re.search(\n        '(([01]?\\\\d\\\\d?|2[0-4]\\\\d|25[0-5])\\\\.([01]?\\\\d\\\\d?|2[0-4]\\\\d|25[0-5])\\\\.([01]?\\\\d\\\\d?|2[0-4]\\\\d|25[0-5])\\\\.'\n        '([01]?\\\\d\\\\d?|2[0-4]\\\\d|25[0-5])\\\\/)|'  # IPv4\n        '((0x[0-9a-fA-F]{1,2})\\\\.(0x[0-9a-fA-F]{1,2})\\\\.(0x[0-9a-fA-F]{1,2})\\\\.(0x[0-9a-fA-F]{1,2})\\\\/)' # IPv4 in hexadecimal\n        '(?:[a-fA-F0-9]{1,4}:){7}[a-fA-F0-9]{1,4}', url)  # Ipv6\n    if match:\n        # print match.group()\n        return 1\n    else:\n        # print 'No matching pattern found'\n        return 0\ndf['use_of_ip'] = df['url'].apply(lambda i: having_ip_address(i))","execution_count":6,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from urllib.parse import urlparse\n\n\n\n\ndef abnormal_url(url):\n    hostname = urlparse(url).hostname\n    hostname = str(hostname)\n    match = re.search(hostname, url)\n    if match:\n        # print match.group()\n        return 1\n    else:\n        # print 'No matching pattern found'\n        return 0\n\n\ndf['abnormal_url'] = df['url'].apply(lambda i: abnormal_url(i))","execution_count":7,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from googlesearch import search","execution_count":16,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def google_index(url):\n    site = search(url, 5)\n    return 1 if site else 0\ndf['google_index'] = df['url'].apply(lambda i: google_index(i))","execution_count":17,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['count.'] = df['url'].apply(lambda i: i.count('.'))\ndf.head()","execution_count":20,"outputs":[{"output_type":"execute_result","execution_count":20,"data":{"text/plain":"                                                 url        type  use_of_ip  \\\n0                                   br-icloud.com.br    phishing          0   \n1                mp3raid.com/music/krizz_kaliko.html      benign          0   \n2                    bopsecrets.org/rexroth/cr/1.htm      benign          0   \n3  http://www.garage-pirenne.be/index.php?option=...  defacement          0   \n4  http://adventure-nicaragua.net/index.php?optio...  defacement          0   \n\n   abnormal_url  google_index  count.  \n0             0             1       2  \n1             0             1       2  \n2             0             1       2  \n3             1             1       3  \n4             1             1       2  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>url</th>\n      <th>type</th>\n      <th>use_of_ip</th>\n      <th>abnormal_url</th>\n      <th>google_index</th>\n      <th>count.</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>br-icloud.com.br</td>\n      <td>phishing</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>mp3raid.com/music/krizz_kaliko.html</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>bopsecrets.org/rexroth/cr/1.htm</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>http://www.garage-pirenne.be/index.php?option=...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>http://adventure-nicaragua.net/index.php?optio...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['count-www'] = df['url'].apply(lambda i: i.count('www'))\ndf['count@'] = df['url'].apply(lambda i: i.count('@'))\nfrom urllib.parse import urlparse\ndef no_of_dir(url):\n    urldir = urlparse(url).path\n    return urldir.count('/')\ndf['count_dir'] = df['url'].apply(lambda i: no_of_dir(i))\ndef no_of_embed(url):\n    urldir = urlparse(url).path\n    return urldir.count('//')\ndf['count_embed_domian'] = df['url'].apply(lambda i: no_of_embed(i))\ndef shortening_service(url):\n    match = re.search('bit\\.ly|goo\\.gl|shorte\\.st|go2l\\.ink|x\\.co|ow\\.ly|t\\.co|tinyurl|tr\\.im|is\\.gd|cli\\.gs|'\n                      'yfrog\\.com|migre\\.me|ff\\.im|tiny\\.cc|url4\\.eu|twit\\.ac|su\\.pr|twurl\\.nl|snipurl\\.com|'\n                      'short\\.to|BudURL\\.com|ping\\.fm|post\\.ly|Just\\.as|bkite\\.com|snipr\\.com|fic\\.kr|loopt\\.us|'\n                      'doiop\\.com|short\\.ie|kl\\.am|wp\\.me|rubyurl\\.com|om\\.ly|to\\.ly|bit\\.do|t\\.co|lnkd\\.in|'\n                      'db\\.tt|qr\\.ae|adf\\.ly|goo\\.gl|bitly\\.com|cur\\.lv|tinyurl\\.com|ow\\.ly|bit\\.ly|ity\\.im|'\n                      'q\\.gs|is\\.gd|po\\.st|bc\\.vc|twitthis\\.com|u\\.to|j\\.mp|buzurl\\.com|cutt\\.us|u\\.bb|yourls\\.org|'\n                      'x\\.co|prettylinkpro\\.com|scrnch\\.me|filoops\\.info|vzturl\\.com|qr\\.net|1url\\.com|tweez\\.me|v\\.gd|'\n                      'tr\\.im|link\\.zip\\.net',\n                      url)\n    if match:\n        return 1\n    else:\n        return 0\ndf['short_url'] = df['url'].apply(lambda i: shortening_service(i))","execution_count":21,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['count-https'] = df['url'].apply(lambda i : i.count('https'))\ndf['count-http'] = df['url'].apply(lambda i : i.count('http'))","execution_count":22,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['count%'] = df['url'].apply(lambda i: i.count('%'))\ndf['count?'] = df['url'].apply(lambda i: i.count('?'))\ndf['count-'] = df['url'].apply(lambda i: i.count('-'))\ndf['count='] = df['url'].apply(lambda i: i.count('='))\n#Length of URL\ndf['url_length'] = df['url'].apply(lambda i: len(str(i)))\n#Hostname Length\ndf['hostname_length'] = df['url'].apply(lambda i: len(urlparse(i).netloc))\n\ndf.head()","execution_count":23,"outputs":[{"output_type":"execute_result","execution_count":23,"data":{"text/plain":"                                                 url        type  use_of_ip  \\\n0                                   br-icloud.com.br    phishing          0   \n1                mp3raid.com/music/krizz_kaliko.html      benign          0   \n2                    bopsecrets.org/rexroth/cr/1.htm      benign          0   \n3  http://www.garage-pirenne.be/index.php?option=...  defacement          0   \n4  http://adventure-nicaragua.net/index.php?optio...  defacement          0   \n\n   abnormal_url  google_index  count.  count-www  count@  count_dir  \\\n0             0             1       2          0       0          0   \n1             0             1       2          0       0          2   \n2             0             1       2          0       0          3   \n3             1             1       3          1       0          1   \n4             1             1       2          0       0          1   \n\n   count_embed_domian  short_url  count-https  count-http  count%  count?  \\\n0                   0          0            0           0       0       0   \n1                   0          0            0           0       0       0   \n2                   0          0            0           0       0       0   \n3                   0          0            0           1       0       1   \n4                   0          0            0           1       0       1   \n\n   count-  count=  url_length  hostname_length  \n0       1       0          16                0  \n1       0       0          35                0  \n2       0       0          31                0  \n3       1       4          88               21  \n4       1       3         235               23  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>url</th>\n      <th>type</th>\n      <th>use_of_ip</th>\n      <th>abnormal_url</th>\n      <th>google_index</th>\n      <th>count.</th>\n      <th>count-www</th>\n      <th>count@</th>\n      <th>count_dir</th>\n      <th>count_embed_domian</th>\n      <th>short_url</th>\n      <th>count-https</th>\n      <th>count-http</th>\n      <th>count%</th>\n      <th>count?</th>\n      <th>count-</th>\n      <th>count=</th>\n      <th>url_length</th>\n      <th>hostname_length</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>br-icloud.com.br</td>\n      <td>phishing</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>16</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>mp3raid.com/music/krizz_kaliko.html</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>35</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>bopsecrets.org/rexroth/cr/1.htm</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>3</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>31</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>http://www.garage-pirenne.be/index.php?option=...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>4</td>\n      <td>88</td>\n      <td>21</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>http://adventure-nicaragua.net/index.php?optio...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>235</td>\n      <td>23</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"def suspicious_words(url):\n    match = re.search('PayPal|login|signin|bank|account|update|free|lucky|service|bonus|ebayisapi|webscr',\n                      url)\n    if match:\n        return 1\n    else:\n        return 0\ndf['sus_url'] = df['url'].apply(lambda i: suspicious_words(i))","execution_count":24,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.head()","execution_count":25,"outputs":[{"output_type":"execute_result","execution_count":25,"data":{"text/plain":"                                                 url        type  use_of_ip  \\\n0                                   br-icloud.com.br    phishing          0   \n1                mp3raid.com/music/krizz_kaliko.html      benign          0   \n2                    bopsecrets.org/rexroth/cr/1.htm      benign          0   \n3  http://www.garage-pirenne.be/index.php?option=...  defacement          0   \n4  http://adventure-nicaragua.net/index.php?optio...  defacement          0   \n\n   abnormal_url  google_index  count.  count-www  count@  count_dir  \\\n0             0             1       2          0       0          0   \n1             0             1       2          0       0          2   \n2             0             1       2          0       0          3   \n3             1             1       3          1       0          1   \n4             1             1       2          0       0          1   \n\n   count_embed_domian  short_url  count-https  count-http  count%  count?  \\\n0                   0          0            0           0       0       0   \n1                   0          0            0           0       0       0   \n2                   0          0            0           0       0       0   \n3                   0          0            0           1       0       1   \n4                   0          0            0           1       0       1   \n\n   count-  count=  url_length  hostname_length  sus_url  \n0       1       0          16                0        0  \n1       0       0          35                0        0  \n2       0       0          31                0        0  \n3       1       4          88               21        0  \n4       1       3         235               23        0  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>url</th>\n      <th>type</th>\n      <th>use_of_ip</th>\n      <th>abnormal_url</th>\n      <th>google_index</th>\n      <th>count.</th>\n      <th>count-www</th>\n      <th>count@</th>\n      <th>count_dir</th>\n      <th>count_embed_domian</th>\n      <th>short_url</th>\n      <th>count-https</th>\n      <th>count-http</th>\n      <th>count%</th>\n      <th>count?</th>\n      <th>count-</th>\n      <th>count=</th>\n      <th>url_length</th>\n      <th>hostname_length</th>\n      <th>sus_url</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>br-icloud.com.br</td>\n      <td>phishing</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>16</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>mp3raid.com/music/krizz_kaliko.html</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>35</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>bopsecrets.org/rexroth/cr/1.htm</td>\n      <td>benign</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>3</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>31</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>http://www.garage-pirenne.be/index.php?option=...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>4</td>\n      <td>88</td>\n      <td>21</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>http://adventure-nicaragua.net/index.php?optio...</td>\n      <td>defacement</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>235</td>\n      <td>23</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.columns","execution_count":26,"outputs":[{"output_type":"execute_result","execution_count":26,"data":{"text/plain":"Index(['url', 'type', 'use_of_ip', 'abnormal_url', 'google_index', 'count.',\n       'count-www', 'count@', 'count_dir', 'count_embed_domian', 'short_url',\n       'count-https', 'count-http', 'count%', 'count?', 'count-', 'count=',\n       'url_length', 'hostname_length', 'sus_url'],\n      dtype='object')"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"!pip install tld","execution_count":27,"outputs":[{"output_type":"stream","text":"Collecting tld\n  Downloading tld-0.12.2-py37-none-any.whl (329 kB)\n\u001b[K     |████████████████████████████████| 329 kB 193 kB/s eta 0:00:01\n\u001b[?25hInstalling collected packages: tld\nSuccessfully installed tld-0.12.2\n\u001b[33mWARNING: You are using pip version 20.1; however, version 20.2.2 is available.\nYou should consider upgrading via the '/opt/conda/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Importing dependencies\nfrom urllib.parse import urlparse\nfrom tld import get_tld\nimport os.path\n\n#First Directory Length\ndef fd_length(url):\n    urlpath= urlparse(url).path\n    try:\n        return len(urlpath.split('/')[1])\n    except:\n        return 0\n\ndf['fd_length'] = df['url'].apply(lambda i: fd_length(i))\n\n#Length of Top Level Domain\ndf['tld'] = df['url'].apply(lambda i: get_tld(i,fail_silently=True))\ndef tld_length(tld):\n    try:\n        return len(tld)\n    except:\n        return -1\n\ndf['tld_length'] = df['tld'].apply(lambda i: tld_length(i))","execution_count":28,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def digit_count(url):\n    digits = 0\n    for i in url:\n        if i.isnumeric():\n            digits = digits + 1\n    return digits\ndf['count-digits']= df['url'].apply(lambda i: digit_count(i))","execution_count":29,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def letter_count(url):\n    letters = 0\n    for i in url:\n        if i.isalpha():\n            letters = letters + 1\n    return letters\ndf['count-letters']= df['url'].apply(lambda i: letter_count(i))","execution_count":30,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df = df.drop(\"tld\",1)","execution_count":31,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.columns","execution_count":32,"outputs":[{"output_type":"execute_result","execution_count":32,"data":{"text/plain":"Index(['url', 'type', 'use_of_ip', 'abnormal_url', 'google_index', 'count.',\n       'count-www', 'count@', 'count_dir', 'count_embed_domian', 'short_url',\n       'count-https', 'count-http', 'count%', 'count?', 'count-', 'count=',\n       'url_length', 'hostname_length', 'sus_url', 'fd_length', 'tld_length',\n       'count-digits', 'count-letters'],\n      dtype='object')"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['type'].value_counts()","execution_count":33,"outputs":[{"output_type":"execute_result","execution_count":33,"data":{"text/plain":"benign        428103\ndefacement     96457\nphishing       94111\nmalware        32520\nName: type, dtype: int64"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.preprocessing import LabelEncoder\n\nlb_make = LabelEncoder()\ndf[\"type_code\"] = lb_make.fit_transform(df[\"type\"])\ndf[\"type_code\"].value_counts()\n\n","execution_count":34,"outputs":[{"output_type":"execute_result","execution_count":34,"data":{"text/plain":"0    428103\n1     96457\n3     94111\n2     32520\nName: type_code, dtype: int64"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Predictor Variables\nX = df[['use_of_ip','abnormal_url', 'google_index', 'count.', 'count-www', 'count@',\n       'count_dir', 'count_embed_domian', 'short_url', 'count-https',\n       'count-http', 'count%', 'count?', 'count-', 'count=', 'url_length',\n       'hostname_length', 'sus_url', 'fd_length', 'tld_length', 'count-digits',\n       'count-letters']]\n\n#Target Variable\ny = df['type_code']","execution_count":35,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X.head()","execution_count":36,"outputs":[{"output_type":"execute_result","execution_count":36,"data":{"text/plain":"   use_of_ip  abnormal_url  google_index  count.  count-www  count@  \\\n0          0             0             1       2          0       0   \n1          0             0             1       2          0       0   \n2          0             0             1       2          0       0   \n3          0             1             1       3          1       0   \n4          0             1             1       2          0       0   \n\n   count_dir  count_embed_domian  short_url  count-https  ...  count?  count-  \\\n0          0                   0          0            0  ...       0       1   \n1          2                   0          0            0  ...       0       0   \n2          3                   0          0            0  ...       0       0   \n3          1                   0          0            0  ...       1       1   \n4          1                   0          0            0  ...       1       1   \n\n   count=  url_length  hostname_length  sus_url  fd_length  tld_length  \\\n0       0          16                0        0          0          -1   \n1       0          35                0        0          5          -1   \n2       0          31                0        0          7          -1   \n3       4          88               21        0          9           2   \n4       3         235               23        0          9           3   \n\n   count-digits  count-letters  \n0             0             13  \n1             1             29  \n2             1             25  \n3             7             63  \n4            22            199  \n\n[5 rows x 22 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>use_of_ip</th>\n      <th>abnormal_url</th>\n      <th>google_index</th>\n      <th>count.</th>\n      <th>count-www</th>\n      <th>count@</th>\n      <th>count_dir</th>\n      <th>count_embed_domian</th>\n      <th>short_url</th>\n      <th>count-https</th>\n      <th>...</th>\n      <th>count?</th>\n      <th>count-</th>\n      <th>count=</th>\n      <th>url_length</th>\n      <th>hostname_length</th>\n      <th>sus_url</th>\n      <th>fd_length</th>\n      <th>tld_length</th>\n      <th>count-digits</th>\n      <th>count-letters</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>16</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-1</td>\n      <td>0</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>35</td>\n      <td>0</td>\n      <td>0</td>\n      <td>5</td>\n      <td>-1</td>\n      <td>1</td>\n      <td>29</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>3</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>31</td>\n      <td>0</td>\n      <td>0</td>\n      <td>7</td>\n      <td>-1</td>\n      <td>1</td>\n      <td>25</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1</td>\n      <td>1</td>\n      <td>4</td>\n      <td>88</td>\n      <td>21</td>\n      <td>0</td>\n      <td>9</td>\n      <td>2</td>\n      <td>7</td>\n      <td>63</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>235</td>\n      <td>23</td>\n      <td>0</td>\n      <td>9</td>\n      <td>3</td>\n      <td>22</td>\n      <td>199</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 22 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X.columns","execution_count":37,"outputs":[{"output_type":"execute_result","execution_count":37,"data":{"text/plain":"Index(['use_of_ip', 'abnormal_url', 'google_index', 'count.', 'count-www',\n       'count@', 'count_dir', 'count_embed_domian', 'short_url', 'count-https',\n       'count-http', 'count%', 'count?', 'count-', 'count=', 'url_length',\n       'hostname_length', 'sus_url', 'fd_length', 'tld_length', 'count-digits',\n       'count-letters'],\n      dtype='object')"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X.head()","execution_count":38,"outputs":[{"output_type":"execute_result","execution_count":38,"data":{"text/plain":"   use_of_ip  abnormal_url  google_index  count.  count-www  count@  \\\n0          0             0             1       2          0       0   \n1          0             0             1       2          0       0   \n2          0             0             1       2          0       0   \n3          0             1             1       3          1       0   \n4          0             1             1       2          0       0   \n\n   count_dir  count_embed_domian  short_url  count-https  ...  count?  count-  \\\n0          0                   0          0            0  ...       0       1   \n1          2                   0          0            0  ...       0       0   \n2          3                   0          0            0  ...       0       0   \n3          1                   0          0            0  ...       1       1   \n4          1                   0          0            0  ...       1       1   \n\n   count=  url_length  hostname_length  sus_url  fd_length  tld_length  \\\n0       0          16                0        0          0          -1   \n1       0          35                0        0          5          -1   \n2       0          31                0        0          7          -1   \n3       4          88               21        0          9           2   \n4       3         235               23        0          9           3   \n\n   count-digits  count-letters  \n0             0             13  \n1             1             29  \n2             1             25  \n3             7             63  \n4            22            199  \n\n[5 rows x 22 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>use_of_ip</th>\n      <th>abnormal_url</th>\n      <th>google_index</th>\n      <th>count.</th>\n      <th>count-www</th>\n      <th>count@</th>\n      <th>count_dir</th>\n      <th>count_embed_domian</th>\n      <th>short_url</th>\n      <th>count-https</th>\n      <th>...</th>\n      <th>count?</th>\n      <th>count-</th>\n      <th>count=</th>\n      <th>url_length</th>\n      <th>hostname_length</th>\n      <th>sus_url</th>\n      <th>fd_length</th>\n      <th>tld_length</th>\n      <th>count-digits</th>\n      <th>count-letters</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>16</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-1</td>\n      <td>0</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>35</td>\n      <td>0</td>\n      <td>0</td>\n      <td>5</td>\n      <td>-1</td>\n      <td>1</td>\n      <td>29</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>3</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>31</td>\n      <td>0</td>\n      <td>0</td>\n      <td>7</td>\n      <td>-1</td>\n      <td>1</td>\n      <td>25</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1</td>\n      <td>1</td>\n      <td>4</td>\n      <td>88</td>\n      <td>21</td>\n      <td>0</td>\n      <td>9</td>\n      <td>2</td>\n      <td>7</td>\n      <td>63</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1</td>\n      <td>1</td>\n      <td>3</td>\n      <td>235</td>\n      <td>23</td>\n      <td>0</td>\n      <td>9</td>\n      <td>3</td>\n      <td>22</td>\n      <td>199</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 22 columns</p>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, test_size=0.2,shuffle=True, random_state=5)","execution_count":39,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"lgb = LGBMClassifier(objective='multiclass',boosting_type= 'gbdt',n_jobs = 5, \n          silent = True, random_state=5)\nLGB_C = lgb.fit(X_train, y_train)\n\n\ny_pred = LGB_C.predict(X_test)\nprint(classification_report(y_test,y_pred))\n\nscore = metrics.accuracy_score(y_test, y_pred)\nprint(\"accuracy:   %0.3f\" % score)","execution_count":40,"outputs":[{"output_type":"stream","text":"              precision    recall  f1-score   support\n\n           0       0.97      0.99      0.98     85621\n           1       0.96      0.99      0.97     19292\n           2       0.96      0.89      0.92      6504\n           3       0.90      0.81      0.85     18822\n\n    accuracy                           0.96    130239\n   macro avg       0.95      0.92      0.93    130239\nweighted avg       0.95      0.96      0.95    130239\n\naccuracy:   0.956\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"def plot_confusion_matrix(cm, classes,\n                          normalize=False,\n                          title='Confusion matrix',\n                          cmap=plt.cm.Blues):\n    \"\"\"\n    See full source and example: \n    http://scikit-learn.org/stable/auto_examples/model_selection/plot_confusion_matrix.html\n    \n    This function prints and plots the confusion matrix.\n    Normalization can be applied by setting `normalize=True`.\n    \"\"\"\n    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n    plt.title(title)\n    plt.colorbar()\n    tick_marks = np.arange(len(classes))\n    plt.xticks(tick_marks, classes, rotation=45)\n    plt.yticks(tick_marks, classes)\n\n    if normalize:\n        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n        print(\"Normalized confusion matrix\")\n    else:\n        print('Confusion matrix, without normalization')\n\n    thresh = cm.max() / 2.\n    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n        plt.text(j, i, cm[i, j],\n                 horizontalalignment=\"center\",\n                 color=\"white\" if cm[i, j] > thresh else \"black\")\n\n    plt.tight_layout()\n    plt.ylabel('True label')\n    plt.xlabel('Predicted label')\n    ","execution_count":41,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cm = metrics.confusion_matrix(y_test, y_pred, labels=[0,1,2,3])\nplot_confusion_matrix(cm,classes=['benign', 'defacement','phishing','malware'])\n","execution_count":42,"outputs":[{"output_type":"stream","text":"Confusion matrix, without normalization\n","name":"stdout"},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 2 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"lgb_feature = lgb.feature_importances_\nlgb_feature","execution_count":43,"outputs":[{"output_type":"execute_result","execution_count":43,"data":{"text/plain":"array([  31,   57,    0,  992,  529,   78, 1168,   41,  124,  310,  162,\n        274,  191,  591,  445,  995, 1496,  299, 1765,  636, 1101,  715],\n      dtype=int32)"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = xgb.XGBClassifier(n_estimators= 100)\nmodel.fit(X_train,y_train)\ny_pred = model.predict(X_test)\nprint(classification_report(y_test,y_pred))\n\n\nscore = metrics.accuracy_score(y_test, y_pred)\nprint(\"accuracy:   %0.3f\" % score)","execution_count":44,"outputs":[{"output_type":"stream","text":"              precision    recall  f1-score   support\n\n           0       0.97      0.99      0.98     85621\n           1       0.97      0.99      0.98     19292\n           2       0.98      0.92      0.94      6504\n           3       0.91      0.83      0.87     18822\n\n    accuracy                           0.96    130239\n   macro avg       0.96      0.93      0.94    130239\nweighted avg       0.96      0.96      0.96    130239\n\naccuracy:   0.961\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"CM=confusion_matrix(y_test,y_pred,labels=[0,1,2,3])\n\nplot_confusion_matrix(cm,classes=['benign', 'defacement','phishing','malware'])","execution_count":45,"outputs":[{"output_type":"stream","text":"Confusion matrix, without normalization\n","name":"stdout"},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 2 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAWUAAAEmCAYAAAC6SYF5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOydd3xUVdOAn0kCCApIL6FL7xAICIIoIqhIkS4KKor62hURRcWGNF8LL8JnF7EAotIEFEGkSEcQkKp0QpWOlMB8f9yTsAnJJkCS3Y3z+Du/vXdOuXM3Mnvu3HNmRFUxDMMwgoOwQCtgGIZhnMOMsmEYRhBhRtkwDCOIMKNsGIYRRJhRNgzDCCLMKBuGYQQRZpSNgCAi2UVkkogcEpGvL2GcriLyY1rqFihEpJGIrAu0HkZgEVunbPhDRG4HngQqAkeA5UB/VZ17iePeCTwCNFDV2EtWNMgREQXKqerGQOtiBDc2UzaSRUSeBN4GXgcKASWA4UDrNBi+JLD+32CQU4OIRARaByNIUFUrVs4rQG7gKNDBT5tseEZ7pytvA9lcXRNgO/AUsAeIAe52dS8Dp4DT7ho9gJeAz33GLgUoEOHO7wL+wputbwK6+sjn+vRrACwGDrnPBj51s4BXgXlunB+B/MncW5z+vX30bwPcDKwH/gae82kfDcwHDrq2w4Csrm62u5dj7n47+Yz/DLALGBUnc32ucteo7c6LAvuAJoH+f8NK+habKRvJcTVwGfCdnzZ9gfpATaAGnmF63qe+MJ5xj8QzvO+KSB5V7Yc3+x6jqleo6kf+FBGRy4GhwE2qmhPP8C5Pol1e4HvXNh/wJvC9iOTzaXY7cDdQEMgK9PJz6cJ430Ek8CLwAXAHEAU0Al4UkTKu7RngCSA/3nfXFPgPgKo2dm1quPsd4zN+Xrynhp6+F1bVP/EM9hcikgP4BPhUVWf50dfIBJhRNpIjH7BP/bsXugKvqOoeVd2LNwO+06f+tKs/rapT8GaJFS5Sn7NAVRHJrqoxqro6iTa3ABtUdZSqxqrqV8Ba4FafNp+o6npV/QcYi/eDkhyn8fznp4HReAb3HVU94q6/GqgOoKpLVXWBu+5m4D3g2lTcUz9VPen0SYCqfgBsABYCRfB+BI1MjhllIzn2A/lT8HUWBbb4nG9xsvgxEhn148AVF6qIqh7De+R/AIgRke9FpGIq9InTKdLnfNcF6LNfVc+44zijudun/p+4/iJSXkQmi8guETmM9ySQ38/YAHtV9UQKbT4AqgL/U9WTKbQ1MgFmlI3kmA+cwPOjJsdOvEfvOEo42cVwDMjhc17Yt1JVf1DVZngzxrV4xiolfeJ02nGROl0II/D0KqequYDnAEmhj9+lTyJyBZ6f/iPgJeeeMTI5ZpSNJFHVQ3h+1HdFpI2I5BCRLCJyk4gMds2+Ap4XkQIikt+1//wiL7kcaCwiJUQkN/BsXIWIFBKRVs63fBLPDXImiTGmAOVF5HYRiRCRTkBlYPJF6nQh5AQOA0fdLP7BRPW7gTLn9fLPO8BSVb0Xz1f+f5espRH0mFE2kkVV38Rbo/w8sBfYBjwMjHdNXgOWAL8DK4FlTnYx15oOjHFjLSWhIQ3DW8WxE29FwrW4l2iJxtgPtHRt9+OtnGipqvsuRqcLpBfeS8QjeLP4MYnqXwJGishBEemY0mAi0hpogeeyAe/vUFtEuqaZxkZQYptHDMMwggibKRuGYQQRZpQNwzCCCDPKhmEYQYQZZcMwjCDCgqCkMxKRXSVrzkCrcdHUqlQi0Cr868kMr+J/W7Z0n6oWSKvxwnOVVI09bxNkAvSfvT+oaou0umZGYUY5nZGsOclWIcUVUEHLvIXDAq3Cv54zZ0PfLF+RLSzxTstLQmP/SfHf1Ynl76a0ozIoMaNsGEboIQJh4YHWIl0wo2wYRmgimfOVmBllwzBCE0kptEhoYkbZMIwQxNwXhmEYwYOQad0XmfOuDMPI5IjnvvBXUjOKyBMislpEVonIVyJymYjkFZHpIrLBfebxaf+siGwUkXUi0txHHiUiK13dUBFPARHJJiJjnHyhiJRKSSczyoZhhCZh4f5LCohIJPAoUEdVqwLhQGegDzBDVcsBM9w5IlLZ1VfBi+A3XETiLjQCL6VXOVfi1kf3AA6oalngLWBQireVurs3DMMIJsRzX/grqSMCyO4y7OTACw/bGhjp6kdyLtFDa2C0S9+1CdgIRItIESCXqs5XL+zmZ4n6xI01DmgaN4tODjPKhmGEHkJq3Bf5RWSJT0mcnHYH8AawFS8D+SFV/REopKoxrk0MXpJd8NKKbfMZYruTRbrjxPIEfVxqtEN4+S+TxV70GYYRmqQ8G96nqnWS7e75ilsDpYGDwNcicoe/KyYhUz9yf32SxWbKhmGEIALh4f5LytwAbFLVvS5j+bdAA2C3c0ngPve49tuB4j79i+G5O7a748TyBH2ciyQ3XvacZDGjbBhG6BG3JO7SfMpbgfou/6QATYE1wESgu2vTHZjgjicCnd2KitJ4L/QWORfHERGp78bplqhP3FjtgZmaQronc18YhhGaXOKOPlVdKCLj8HJLxgK/Ae8DVwBjRaQHnuHu4NqvFpGxwB+u/UOqGpfA90HgUyA7MNUV8DKRjxKRjXgz5M4p6WVG2TCMECRtdvSpaj+gXyLxSbxZc1Lt+wP9k5AvAaomIT+BM+qpxYyyYRihie3oM9KTR7pex9JxfVny9XOMHHAX2bKe+718/M6m/PPbMPJdeTkAdaqUZMHoPiwY3YeFY/rQ6rrqAFyRI1u8fMHoPmybOZAhvdoB0LD2Vfz65TMcWfwObW+omfE3CGzbto3mN1xHzWqVqF2jCsOGvgPAa6+8RJmSkdSLqkm9qJpMmzolIPqlhvvvvYcSRQsSVfPcpOibcV9Tu0YVcmQNY+mSJQHU7hwP9ryHUsUKUbdWtXhZ3z5PU6taJepF1aBzh9s4ePAgAPv37+emG6+nUN6cPPnYw0mO1/G21gnGCjgpLYcL4WBFZpSDgKIFcvOfLtfSsOtg6nR4nfCwMDo0jwKgWKErub5+RbbGnHthu/rPnTTsOpj6nQfS+qHh/O/5LoSHh3H0+Enqdx4YX7bG/M34mcsB2BZzgJ79RjFmWuCMRkREBAMH/5flK9fwy9wFvPd/77Lmjz8AeOSxJ1i4dDkLly6nxU03B0zHlLiz+11MmDwtgaxKlaqMHvst1zRqHCCtzqfrnXcxftLUBLLrmzZj8W8rWbh0BeXKleO/gwcAcNlll/FCv1foP3BIkmNNGP8tl19xRbrrfMFc4o6+YMWMcpAQER5O9mxZCA8PI/tlWYnZewiAwb3a0fed8fi+sP3nxGnOnDkLQLasWUjqZe5VJQpQMG9O5i37E4CtMX+zasNOzgYwi0WRIkWoVbs2ADlz5qRixUrs3LkjYPpcDNc0akzevHkTyCpWqkT5ChUCpFHSXNOoMXnyJNSzabMbiYjwnsDq1qvPjh3ed3/55ZfToOE1XHbZZeeNc/ToUYa98xa9n+2b/kpfEGm2oy/oCF3NMxE79x7i7c9msH7qq2ya3p/DR/9hxoK13HJtNXbuOcjK9ecbrrpVS8a7Ox7tPzreSMfRsUUU435cllG3cMFs2byZ5ct/o250PQD+b/gw6taqzv333sOBAwcCrF3mZ9Snn3Bj85TT17360gs88viT5MieIwO0ukDMfRGciEgpEVmVBuPUEZGhaaHThXJlzuy0bFKNSi37UebGvlyePSu3t4zmmR7NeWXE90n2WbxqC1Ht+3PNHYN5+p4bE/igATo0j2JsAF0V/jh69ChdOrZjyH/fJleuXNx3/4P8se5PFi5dTuEiRejz9FOBVjFTM3hgf8IjIujUpavfdr+vWM5ff/5Jq9ZtM0izC0AEwiL8lxAl5I1yWqGqS1T10UBc+/p6Fdm8cz/7DhwlNvYs42euoFur+pSMzMeiMc+y9vuXiSx4JfO/fIZC+RJmxl63aTfH/jlFlbJF42XVykcSER7Ob2u2Jb5UwDl9+jRdOrajU5eutGl7GwCFChUiPDycsLAw7ulxH0uWLAqwlpmXL0aNZNqU7/l45OekEBeHhQvm89tvS6lcvjTNrm/Exg3radHsugzSNBXYTDmoiRCRkSLyu4iMczt0okTkFxFZKiI/+GybnCUig0RkkYisF5FGTt5ERCa74wIujuoyEXlPRLaISH43K18jIh+4GKw/ikj2S1V+266/ia5WmuyXZQHguugKTJi5gpJNn6XiLf2oeEs/duw5yNW3D2L3/iOULJqP8HDvT1eiSB7KlyrElp3748fr2CI4Z8mqygP39aBCxUo89sST8fKYmJj44wnjv6NylfOWexppwPQfpvHmG4MZ880EcuRI2R1x3/0PsnHzDv5Yv4npM+dQtlx5pk3/OQM0TSWZ1KccunP8hFQAeqjqPBH5GHgIaAu0VtW9ItIJb8H3Pa59hKpGi8jNeAvHb0g0Xj+87ZADRKQFXpzUOMoBXVT1Pre7px3w+aUov3jVFr776Tfmf/kMsWfOsmLtdj76Zl6y7RvUKkOvu2/kdOwZzp5VHnt9DPsPHouvb9esNm0eGZGgT1TlEox58z6uzJWDmxtX4/kHbiGq/Xlr4NOVX+fN48svRlG1ajXqRXnL8l5+7XXGjv6K31csR0QoWaoU/xv+XobqdSF0u6MLc36Zxb59+7iqVDFeePFl8uTNy5OPP8K+vXu5rfUtVK9Rk0lTfgionnfdeTtzZs9i/759lC9TnL4vvMR/Bw/k5KmTtLr5RgDqRtdj6Lv/B0Dl8qU5cvgwp06dYvKkCUz4/gcqVaocyFvwTybOZi0pbMMOelwk/9mqWsKdXw88B0QDf7lm4UCMqt4oIrOAvs6AFwLmqWpZEWkC9FLVliKyHGjrYqYiIn8D5fG2X053wa8RkWeALKr6WiKdehJnyLNcEXVZle6EKgcWDwu0Cv96zgRwxUxacUW2sKX+IrZdKGF5Smm2617w2+bEd/em6TUziswyU078f+0RYLWqXp1M+5Pu8wxJfwf+HFInfY7P4O11T6iM6vt4e+gJy1Ew9P9FGUYQkpJPPFQJXcdLQkqISJwB7gIsAArEyUQki4hUuYDx5gIdXd8bgTz+mxuGkZGIgISJ3xKqZBajvAboLiK/A3mB/+GFyRskIiuA5XhxUlPLy8CNIrIMuAkvK8GRtFXZMIyLRxDxX0KVkHdfqOpmIKk3EsuB8/a9qmoTn+N9QCl3PAuY5aoOAc1VNdbNtq9T1ZPAZnwiQanqG5d+B4ZhXAyhbHj9kVlmymlNCWCxm2UPBe4LsD6GYSQiLCzMb0kJEakgIst9ymEReVxE8rolsRvcZx6fPs+KyEYRWScizX3kUSKy0tUNjUuO6gLij3HyhW5hgv/7uqhvI5OjqhtUtZaq1lDVuqq6ONA6GYbhg6SipICqrlPVmqpaE4gCjgPfAX2AGW6V1Qx3johUxgtSXwVoAQwXkbh1eSPwVlyVcyVuD3sP4ICqlgXeAgalpJcZZcMwQg5Je59yU+BPVd2Cl0x1pJOPBNq449bAaFU96ZbLbgSi3ca0XKo636V6+ixRn7ixxgFNJQXlQt6nbBjGv5NUuCjyi4jv1tb33XLVpOgMfOWOC7m8e6hqjIgUdPJIvJVdcWx3stPuOLE8rs82N1asiBwC8gH7klPajLJhGCFJKmbD+1KzeUREsgKtgGdTapqETP3I/fVJFnNfGIYReqSBT9mHm4Blqrrbne/2iZVTBNjj5NuB4j79igE7nbxYEvIEfUQkAsiNl0A1WcwoG4YRkqShT7kL51wXABOBuNgI3YEJPvLObkVFabwXeoucq+OIiNR3/uJuifrEjdUeL6aO35myuS8Mwwg5BEnVsrcUxxHJATQD7vcRDwTGikgPYCsuG7WqrnZByP4AYoGHVPWM6/Mg8Cle2IWprgB8BIwSkY14M+TOKelkRtkwjNAkDfaOqOpxvBdvvrL9eKsxkmrfHy/iZGL5Enw2lvnIT+CMemoxo2wYRughmXdHnxllwzBCkrRwXwQjZpQNwwg54jaPZEbMKBuGEZpkTptsRtkwjBBEzH1hGIYRVJj7wjAMI5jInDbZjLJhGKGHSNpsHglGzCgbhhGSmPvCuChqVSrBvIXDAq3GRbNq26FAq3DJVC2eO9AqXBLhIZwEND0xo2wYhhFEhHLGan+YUTYMI/SwbdaGYRjBgwCZ1CabUTYMIxQRwjKp+yJzrikxDCPTkxZB7kXkShEZJyJrRWSNiFwtInlFZLqIbHCfeXzaPysiG0VknYg095FHichKVzc0LjmqC4g/xskXikiplHQyo2wYRughnvvCX0kl7wDTVLUiUANYA/QBZqhqOWCGO0dEKuMFqa8CtACGi0i4G2cE0BMvG0k5Vw/QAzigqmWBt4BBKSlkRtkwjJBDgPBw8VtSHEMkF9AYLzsIqnpKVQ8CrYGRrtlIoI07bg2MVtWTqroJ2AhEuzx+uVR1vkv19FmiPnFjjQOaSgrTeDPKhmGEJKlwX+QXkSU+pWeiIcoAe4FPROQ3EflQRC4HCrm8e7jPgq59JLDNp/92J4t0x4nlCfqoaixwiESZThJjL/oMwwg9Uuei2KeqdfzURwC1gUdUdaGIvINzVSR/1fNQP3J/fZLFZsqGYYQccYlT/ZVUsB3YrqoL3fk4PCO927kkcJ97fNoX9+lfDNjp5MWSkCfoIyIRQG68BKrJYkbZMIyQ5FJf9KnqLmCbiFRwoqZ4maonAt2drDswwR1PBDq7FRWl8V7oLXIujiMiUt/5i7sl6hM3VntgpvM7J4u5LwzDCEnSaEffI8AXIpIV+Au4G2+yOlZEegBbcdmoVXW1iIzFM9yxwEOqesaN8yDwKZAdmOoKeC8RR4nIRrwZcueUFDKjbBhG6HFhy96SRVWXA0n5nZsm074/0D8J+RKgahLyEzijnlrMKBuGEXIIZNodfWaUDcMISSwgkWEYRhCRSW2yGWXDMEIPkczrvrAlcSHGsKHvEFWzKrVrVOF/77wdLx8+7H9Ur1KB2jWq8Fyf3hmu1yu9H+LGumXp1OLqeNn6NSu5p10zOrdowBP3duLokcPxdZ8Mf5O219WiXdM6zJ89I16+ZuVyOrdoQNvravHGy72JWz20a8c2Hri9JV1bNqLLTQ2Y9/OPGXdziThz5gz169TittYtAVixfDmNG9anXlRNGtarw+JFiwKmW1Lcf+89lChakKia572H4q033yB7FmHfvn3xsiGDBlClYlmqV6nA9B9/yEhVLwD/u/lC2bVhRjmEWL1qFZ98/AFzfl3EoqUrmDplMhs3bOCXWT8zedIEFi/7nWUrVvP4k70yXLeW7W9n6CfjEshe6/MoD/Xux+hpv3LdjS0Z9cFQAP7asJbpk79hzLQFDP10HINefIozZ7yVRQNfeJLnXn+bb2cuY+vmv/j1l58A+OjdN7jh5rZ8MXkO/Yd+zKAXn8rYG/Rh2NB3qFCpUvx532d70/eFfixcupwXXnqFvs9m/I+iP+7sfhcTJk87T75t2zZm/jSd4iVKxMvW/PEHX48ZzbIVq5k4eRqPPfKf+L9NsJFGAYmCDjPKIcTatWuIjq5Pjhw5iIiIoFHja5kw4Tvef28EvXr3IVu2bAAULFgwhZHSntrRDcl1ZZ4Esq2bNlI7uiEA0ddcx8/TJgHwy/QpNGvZjqzZshFZvBTFS5Zh9Yql7Nuzi2NHj1C9djQiwi1tO/PL9O8B76XOsaNHADh65DD5CxXJwLs7x/bt25k29XvuvufeeJmIcPiw9xRw6NAhihQtGhDdkuOaRo3JmzfvefLevZ6g/4DBCWaVkydNoEOnzmTLlo1SpUtz1VVlg27mD4BzX/groYoZ5RCiSpWqzJ07m/3793P8+HGmTZ3C9m3b2Lh+PfPmzqFRg3o0u/5alixeHGhVAShTvhKzf5oCwIwp49kdswOAvbtjKFQ0Mr5dwcJF2bsrhj27YihYuOh5coCej/Vh6vix3NKgMo/f04Gn+w3OwDs5x9NPPU7/AYMTbOMd8t+3ea7P05QtXZxnn+nFK68NCIhuF8LkSRMpWjSS6jVqJJDv2LGDYsXO7SSOjCzGzp07Mlq9FPEyj5j74pIQkZdEJNnnahEp4IJA/yYijTJKr4vBBcb+T0Zft2KlSjzV6xlatmhGq1taUL16DSIiIog9E8uBAweYPW8Brw8cwh23dySFnZwZwouDhvH1qA+5s9W1HD92lCxZsgAkqZtI0vK4cC4/TBxHy/Zd+P7XP3j746/p99T9nD17Nj3VP48p30+mYIGC1I6KSiB//70RDH7jLTZu2sbgN97iwZ49MlSvC+X48eMMGtCfF1965fzKJP82wWngzCinP02BtapaS1XnBFqZFLgSyHCjDHDXPT2Yv3gZP/08mzx581K2bDkiI4vRpu1tiAh1o6MJCwtL8OImUJS6qjzDPvuOURN/4cZb2xNZojTgzYB3+8y+9uzaSf5CRShUpCh7du1MIC/g3BQTvv6cG25uC0D12tGcPHmCg3/vz8C7gfm/zmPy5IlUKFuKbl07M+vnmdzd7Q6+GDWSNm1vA6Bd+w4sWRyEj/s+/PXnn2zZvInoqBpUKFuKHdu3c3V0bXbt2kVksWJs334uOuWOHdspUiS43DFxmPviIhCRvi5tyk9ABSe7SkSmichSEZkjIhVFpCYwGLhZRJaLSHYRGeFioK4WkZd9xqwrIr+KyAoRWSQiOUUkXESGiMhiEfldRO53bZuIyC8iMlZE1ovIQBHp6vqtFJGrXLsCIvKN679YRBo6+Usi8rGIzBKRv0TkUafGQOAqp+uQ9PwOE7NnjxewauvWrUwY/y0dO3fh1lZtmPXzTAA2rF/PqVOnyJ8/f0aqlSR/79sLwNmzZ/n43SG0u/1uABrfcBPTJ3/DqZMn2bFtM1s3/0mVGlHkL1iYHJdfwcrfFqOqfP/daK694WYAChctxuJffwFg08Z1nDp5kjz5MvYeX+0/gD83b2fdxs189sVomlx3PZ989jlFihZlzmxPt1k/z6Rs2XIZqteFUrVaNbbu3MO6jZtZt3EzkcWKMX/RMgoXLswtLVvx9ZjRnDx5ks2bNrFx4wbqRkcHWuXzSbvMI0FHuq1TFpEovOAbtdx1lgFLgfeBB1R1g4jUA4ar6vUi8iJQR1Ufdv37qurfLt3KDBGpDqwFxgCdVHWxyxzwD17KlUOqWldEsgHzRCRuzVQNoBJeMJC/gA9VNVpEHsMLRvI4XkqYt1R1roiUAH5wfQAqAtcBOYF1IjICL+ZqVVWtmcy998RLDZPgzXZa0KVjO/7+ez9ZIrLw9tB3yZMnD93vvof7772HqJpVyZolKx9+PDLDH9/6PtqDpQvncvDAfm5pUJmej/Xh+PFjjBv1IQBNmt/KrR3uAOCq8pW44Za2dGxej/DwCHq//Abh4V5WnT6vvsnLvf/DyRP/0ODaZjRo0gyAx597jf7PPcZXHw8HEfoNGR40j6jvjviAp598jNjYWLJddhnDRrwfaJUS0O2OLsz5ZRb79u3jqlLFeOHFl7nrnqRdLJWrVKFdh47Uql6ZiIgI3h76bvzfJpgQQttF4Q9JL9+jiDwO5FXVF935m3iGsS+wzqdpNlWtJCJ3kdAoP4Bn2CKAIngGdDXwf6raMNG1xgHVgeNOlBu4HzgF9FXVZq7dbOBZVZ0nItcDj6pqGxHZw7n4pwAF8IzxU8BpF4QEEVkDNHM6TVbV8xd+JiIqqo7OW7gkxe8rWFm17VCgVbhkqhbPHWgV/vVkzyJLUwg4f0HkKlFJ6/X+xG+bnx65Ok2vmVEkO1MWkf/hJ0K+qj6aXJ1vs0TnYcDB5GaYPtcuDfQC6qrqARH5FLgM77VPUjoJXvaABCvdRaQJcNJHdNbn/Czn7j8MuFpV/0nUn0T9z2C7IA0jKMikE2W/PuUleO6G5EpKzAbaOv9wTuBWvJnsJhHpACAeNZLomws4BhwSkULATU6+FigqInVd/5ziRfP/AXhQRLI4eXnxcm2llh+Bh+NOnI/bH0fw3BmGYQQAz2986asvRGSze7+0XESWOFleEZkuIhvcZx6f9s+KyEb3rqy5jzzKjbNRRIaKU0C8gPhjnHyhiJRKSadkZ32qOtL3XEQuV9VjqbpTr/8yERkDLAe2AHErKroCI0TkeSALMBpYkajvChH5Dc9d8Rcwz8lPiUgn4H8ikh3Pn3wD8CFQCljmvoy9nMsmmxoeBd4Vkd/xvpPZwAN+7m2/iMwTkVXAVFV9+gKuZRhGGhCedissrlNV3+VKfYAZqjpQRPq482dEpDLee7IqQFHgJxEp7wLdj8Bzty4ApgAt8ALd9wAOqGpZEekMDAI6+VMmRZ+yiFyNFz3/ClUt4Wa296tqQJaEhRrmUw485lMOPGntU85dspJe89xIv22mPFAvxWuKyGa8d1n7fGTrgCaqGiNejr5ZqlpBRJ4FUNUBrt0PwEvAZuBnVa3o5F1c//vj2qjqfPdUvwso4C8lVGqWxL0NNAf2O4VWAI1T0c8wDCNdENwKDD//AfnFW1YbV3omMZQCP4q3RDeuvpDLu4f7jItbEAls8+m73cki3XFieYI+qhoLHALy+bu3VL20UtVtiXw0wRmhxDCMfwciqXFf7EvF7Lyhqu4UkYLAdBFZ6++qScjUj9xfn2RJzUx5m4g0AFREsoq3VXpNKvoZhmGkG2mxeURVd7rPPcB3QDSw27ktcJ97XPPtQHGf7sXwltJud8eJ5Qn6OPdFbrylwcmSGqP8APAQ3jR8B1DTnRuGYQQEAcJE/JYUxxC53K0Mw63WuhFYBUwEurtm3YEJ7ngi0NmtqCgNlAMWORfHERGp7xYadEvUJ26s9sBMf/5kSIX7wjnAu6Z4h4ZhGBlIGsS3KAR851yzEcCXqjpNRBYDY0WkB7AVl41aVVeLyFjgDyAWeMitvAB4EPgUyI636mKqk38EjBKRjXgz5M4pKZWiURaRMnjbkOvj+ULmA0+o6l+puGnDMIw0Jy3iWzgbdt4+CVXdjxcgLak+/YH+SciXAOft8FXVEzijnlpS4774EhiLt9W5KPA18NWFXMQwDCOtuVT3RbCSGqMsqm+p8s8AACAASURBVDpKVWNd+ZwU3h4ahmGkN5JCCVX8xb6Iyx/zs9vVMhrPGHcCvs8A3QzDMJJESNMdfUGFP5/yUhKuwbvfp06BV9NLKcMwDL+EeHYRf/iLfVE6IxUxDMO4EDKpTU7djj4RqQpUxgufCYCqfpZeShmGYfjj3+q+AEBE+gFN8IzyFLwwmnMBM8qGYQSMzOq+SM3qi/Z4a/Z2qerdeOv6sqWrVoZhGCnwr1t94cM/qnpWRGLFy4m3ByiTznoZhmEki8i/2H0BLBGRK4EP8FZkHAWCO4e6YRiZnszqvkhN7Iu4YPb/JyLTgFyq+nv6qmUYhuGfTGqT/W4eqe2vTlWXpY9KhmEY/pHUxVMOSfzNlP/rp06B69NYFyMIyQyplA7/czrQKlwSOS+zBOpJ8a9zX6jqdRmpiGEYxoWQmqVjoUhmvS/DMDIxcZtH/JVUjyUSLiK/ichkd55XRKaLyAb3mcen7bMislFE1olIcx95lIisdHVDXbB7XED8MU6+UERKpaSPGWXDMEKSMPFfLoDHSJjirg8wQ1XLATPcOSJSGS9IfRWgBTBcRMJdnxFAT7xsJOVcPUAP4ICqlgXeAgaleF8XpLphGEYQ4AW5F78ldeNIMeAW4EMfcWtgpDseCbTxkY9W1ZOqugnYCES7PH65VHW+S/X0WaI+cWONA5pKCsqlaJTF4w4RedGdlxCR6JT6GYZhpCfhYf4LkF9ElviUnkkM8zbQGzjrIyvk8u7hPgs6eSSwzafddieLdMeJ5Qn6qGoscAjI5+++UvNad7hT+HrgFeAI8A1QNxV9DcMw0py4xKkpsE9V6yQ7hkhLYI+qLhWRJqm8bGLUj9xfn2RJjVGup6q1ReQ3AFU9ICJZU9HPMAwj3UgD32tDoJWI3IwXATOXiHwO7BaRIqoa41wTe1z77UBxn/7FgJ1OXiwJuW+f7SISAeTGS6CaLKm5r9POma0AIlKAhFN9wzCMDCVu88ilrL5Q1WdVtZiqlsJ7gTdTVe8AJgLdXbPuwAR3PBHo7FZUlMZ7obfIuTiOiEh95y/ulqhP3Fjt3TUueaY8FPgOKCgi/d3Az6ein2EYRrqRjntHBgJjRaQHsBWXjVpVV4vIWOAPIBZ4SFXPuD4PAp8C2YGprgB8BIwSkY14M+TOKV08NbEvvhCRpXjhOwVoo6prUuhmGIaRrqTlLmtVnQXMcsf78exdUu36A/2TkC8BqiYhP4Ez6qklNUHuSwDHgUm+MlXdeiEXMgzDSCtS+aIvJEmN++J7zr1hvAwoDazDW0BtGIaR8Uj8srdMR2rcF9V8z130uPuTaW4YhpEhSEjnF0meCw4/parLRMTWKBuGETA890WgtUgfUuNTftLnNAyoDexNN40MwzBSwb8xnnIcOX2OY/F8zN+kjzqGYRgp86+dKbtNI1eo6tMZpI/hh/Xr1nHn7Z3izzdt+osX+r3CwoXz2bBuHQAHDx3kytxXsnDp8kCpmYBt27Zx793d2L17F2FhYdzToycPP/pYfP1bb77Bc888zbaYveTPn5+vvvyCt/87JL5+5crfmb9oGTVq1sxQvetULcflV1xBeHg4ERER/PjLAu6763b+3LAegMOHDpErd25mzlvCuDFfMnzom/F9/1i1kp/mLKRU6ato1eJcWPKYHTto1+l2XhvkL39E+nHmzBka1q9L0chIvh0/ib///ptuXTuzZctmSpYsxagvx5AnTx4WL17Eww+610aqPPdCP1q3aRsQnZNFMm86KEluc4mIRKhqrIjMUNUk1+wZKRMVVUfnLVyS5uOeOXOGq0pG8su8hZQsWTJe/szTT5E7d26ee/7FNL/mxRATE8OumBhq1a7NkSNHaFAvirHjxlOpcmW2bdvGf+6/l3Xr1vLrwqXkz58/Qd9VK1fSoV1r1qz/65J0uJjMI3WqluOHX+aTL1/+JOv7PdebXLly8VSfhPuo/li9ku5d2rP493Xn9WnWuB6vDHiDqxs2uiBd0irzyNC332TZ0qUcPnKYb8dPom+f3uTJm5devfvwxuCBHDxwgNcGDOL48eNkzZqViIgIYmJiqF+nJn9u2UFExMXrkSNr2FJ/cSgulBIVq2mvDyf6bfNYozJpes2Mwt+ikriM1ctFZKKI3Ckit8WVjFDOSJ6fZ86gdJmrEhhkVeWbcWPp2KlLADVLSJEiRahV20v3mDNnTipWrMTOnTsA6N3rCfoPGJxsmMWxY74KqnuJQ1WZ+N042rbvdF7dd+PG0LZ9x/Pkf23cwL69e6nf4JqMUPE8tm/fzrSpU7jrnh7xssmTJtL1Tm8HcNc7uzNporczOEeOHPEG+OSJE0GbdknEfwlVUrPSLy+wHy9KXEvgVvdpBJCvx4w+z2DNmzuHQgULUbZcuQBp5Z8tmzezfPlv1I2ux+RJEylaNJLqNWok237c12MCZ5RF6NTmZpo1rsdnn3yYoGrBr3MpULAgZcqe/z1P+CZ5Y936tg4BM3C9n3qC1wYMIizs3D/5PXt2U6RIEcD78dy7d0983aJFC4mqUZW6tavzzrARlzRLTh+EsBRKqOLPKBd0Ky9WASvd52r3uSoDdEsWEdksIuc9V4pIKxHp46ffXSIyLJm6KSJyZVrqmV6cOnWK7ydP5Lb2CXdvjh39FR06B9/MEuDo0aN06diOIf99m4iICAYN6M+LL72SbPtFCxeSI3sOqlQ9b+dqhjD5x1n8NGcRX34ziU8+GMH8eXPi67zZ8PmGd+niRWTPkZ1Klc/Xefw3Y5PskxFM+X4yBQoWoHbtqFT3iY6ux9IVq5jz6yLeGDyQEydOpKOGF45IquIphyT+fv7CgSu4iHiggUJVJ+JFZbqYvjensTrpxg/TplKzVm0KFSoUL4uNjWXC+G+Zt3BpADVLmtOnT9OlYzs6delKm7a3sWrlSrZs3kR0lDdL3rF9O1dH12bOr4soXLgwAF+PHU3HAP7AFC5SFIACBQpyc8vW/LZ0MVc3bERsbCzfTxzP9NkLzuuTnOFdvXIFsbGx1KhVO931TooFv87j+8mT+GHaVE6cOMGRw4e5p/udFCxYiJiYGIoUKUJMTAwFChQ8r2/FSpW4/PLLWb16FVFRweWezazbrP39nsSo6iuq+nISJfkpThoiIqVEZK2IjBSR30VknIjkcNWPiMgyl6ywomsfPxMWkQ4iskpEVojIbJ9hi4rINJcUcbDPtTaLSH53zTUi8oGIrBaRH0Uku2tT1+kxX0SGiEhAnhiS8rXOnPET5StUpFixYsn0CgyqygP39aBCxUo89oS35L1qtWps3bmHdRs3s27jZiKLFWP+omXxBvns2bN8+83XdOiYYkCtdOHYsWMcPXIk/njWzJ+oWMmLKjD75xmUK1+BopEJv+ezZ88yafw3tGl3vj/522Rm1hnFK/0HsHHTNtZu2MRnn3/Ftdddz8cjR3HLrbfyxSgvU9EXo0bS8tZWAGzetInY2FgAtm7Zwvr16yhZslSg1E8S4d/pUw6W26oAvK+q1YHDwH+cfJ+q1sZLWNgriX4vAs1VtQbQykdeE+gEVAM6iUjxJPqWA95V1SrAQaCdk38CPKCqVwNnkuiX7hw/fpyZP02ndduE71qT8jEHA7/Om8eXX4zil59nUi+qJvWiajJt6hS/febOmU1kZDFKlymTQVomZO+e3dzavAnXNYjipusacEPzm7i+mZe4OLnZ8Px5cyhSNJJSpc/XeeJ339C2Q+CMcnI89XQfZs74iWqVyzNzxk881dvz/P06b673t6pTi04dbuPtoe+etzImGEirbNbBhr8lcXlV1W+E/PTGpeOeraol3Pn1wKN4hrWhqu4QkXpAf1W9QUTuAuqo6sMi8n/AVcBY4FtV3e/qG6rqfW68qa7vXBHZDNTBc9lMd5lsEZFngCzAMGCFqpZ08urAl6p6ngPR5QLrCVC8RImo9X9uSfsvx0g1F7MkLphIqyVxgSStl8SVrlRd+3022W+bu6NL+r2miFwGzAay4blyx6lqPxHJC4wBSgGbgY6qesD1eRYvQ/UZ4FFV/cHJozgXT3kK8Jiqqohkw0ukGoW3YKKTqm72p3eyM+VAG2QfEv9qxJ2fdJ9nSMI3rqoP4AXjL463rC9fon7J9k2mTap/elX1fVWto6p1CuQvkNpuhmGklrTJZn0SuN49TdcEWohIfaAPMMNNzGa4c0SkMl6Q+ipAC2C422AH3hN7T7yn7HKuHjwDfkBVywJvAYNSUioU3lGWEJGr3XEXYG5qOonIVaq6UFVfBPaRMLfWBeN+KY+4PxqkIoOAYRjpgwDhIn5LSqjHUXeaxRUFWgMjnXwk0MYdtwZGq+pJVd0EbASiXR6/XKo636V6+ixRn7ixxgFNJYVfjFAwymuA7iLyO96a6RGp7DfEvQRchfeIsiINdOkBvC8i8/H+vziUBmMahnERSAoFyC8iS3xKz/PGEAkXkeV4yVGnq+pCoJDLu4f7jFuWEgls8+m+3cki3XFieYI+qhqLZzPy4YdQcFadda4IX0rFHbg0LE3c8ad4fh1UNaldh/H1rk1Ln+O4Mffhk9ZFVd/w6b/avXDErYdO+/3ThmGkilRMhvel5Md2OfZquj0K34mIv4XxyS0P9rds+IKXFIeCUQ4mbnGO/ghgC3BXYNUxjH8nQupcFKlFVQ+KyCw8X/BuESmiqjHONRG31XE7Cd2gxYCdTl4sCblvn+0iEgHkxkugmixB7b5Q1c1JrW4IFKo6RlVrqmpVVb1FVS2utGEEiEt90SciBeJ28bq9CDcAa/E2oHV3zboDE9zxRKCziGQTkdJ4L/QWORfHERGp7/zF3RL1iRurPTBTk1vy5rCZsmEYIUkazJOLACPdCoowYKyqTnbvjMaKSA9gKy4btaquFpGxwB94seUfcu4PgAc5tyRuqisAHwGjRGQj3gw5xQUCZpQNwwg5RLhk94Wq/g7USkK+H0gyXLGq9gf6JyFfgs+7KB/5CZxRTy1mlA3DCEmCNaTopWJG2TCMkCRzmmQzyoZhhCiZdKJsRtkwjNAjbkdfZsSMsmEYIYggmdSBYUbZMIyQJJNOlM0oG4YReqTFkrhgxYyyYRghSSa1yWaUDcMITcynbBiGESTY6gvDMIwgI5PaZDPKhmGEJua+MAzDCBLSOp5yMGFG2TCM0EPMfWFcJGcVTp4+k3LDICVrRFDnQUgVubJnCbQKl8TGXUdTbvQv5FJtsogUx0tyWhg4C7yvqu+ISF5gDF7auc1AR5c4GZd5qAdelvtHVfUHJ4/iXDzlKcBjqqoiks1dIwrYD3RS1c3+9Ar9f3GGYfzrSIts1niB6p9S1UpAfeAhEakM9AFmqGo5YIY7x9V1BqrgpY0a7gLkg5fQuSdeNpJyrh48A35AVcsCbwGDUlLKjLJhGKFJKtJZ+0NVY1R1mTs+AqzByz7dGhjpmo0E2rjj1sBoVT2pqpuAjUC0y+OXS1Xnu1RPnyXqEzfWOKCppBAI2oyyYRghiaTwH5BfRJb4lJ7JjiVSCi8LyUKgkMu7h/ss6JpFAtt8um13skh3nFieoI+qxgKHgHz+7st8yoZhhCRhKc+G96lqnZQaicgVwDfA46p62M9ENqkK9SP31ydZbKZsGEZoconuCwARyYJnkL9Q1W+deLdzSeA+9zj5dqC4T/diwE4nL5aEPEEfEYkAcuMlUE0WM8qGYYQcnt1N0X3hfwxvSvwRsEZV3/Spmgh0d8fdgQk+8s4ikk1ESuO90FvkXBxHRKS+G7Nboj5xY7UHZjq/c7KY+8IwjNBDUuW+SImGwJ3AShFZ7mTPAQOBsSLSA9iKy0atqqtFZCzwB97KjYdUNW6964OcWxI31RXwjP4oEdmIN0PunJJSZpQNwwhNLtEoq+pcP6M0TaZPf6B/EvIlQNUk5CdwRj21mFE2DCMEsXRQhmEYQYOQJu6LoMSMsmEYoYkZZcMwjODB3BeGYRhBhEWJMwzDCBYsdKdhGEZwYe4LwzCMIEGwmbJhGEZQkVmNssW+CDK2b99GyxZNia5VlfpR1Rnx7lAA7r6zC9fUi+KaelFUq3gV19SLiu+zauXvNGvSkPpR1WlQtyYnTpwA4NSpUzz20ANEVa9E3ZpVmDD+2ySvmd5ULFeaurWqU69OLRrWrwvA7ytW0KRRA+rWqk67Nq04fPhwgj7btm6lQJ6cvP3mG4FQOQH333sPJYoWJKpmwg1bw4f9j+pVKlC7RhWe69MbgMWLFlEvqib1omoSXbsGE8Z/l2F6vvDUg1xbszRtm0af0/HN12lapzztmzegffMGzJ75AwC/zp5Jx5sb0faGenS8uREL5/0S32faxG+4rVl92jSty5v9n4+X79y+lXs7t+S2ZvW5u8NN7IrZkWH3lhSXGvsiWLGZcpARER7BawOGULNWbY4cOUKThtFcd/0NfDLqq/g2ffv0Ileu3ADExsbSs0d33vvwU6pVr8Hf+/eTJYuX/uiNQa9ToEABlv6+hrNnz3Lgb7/BqdKVqdNnkj9//vjz/zxwHwMGDaFR42sZ+enHvPXfIfR7+dX4+t69nuTG5jcFQtXzuLP7XTzwn4e5955u8bJfZv3M5EkTWLzsd7Jly8aePV4gsSpVqzJv4RIiIiKIiYmhXlQNbml5KxER6f9PrXWHrnS56376Pp4wbPCd9z7EXQ88lkCWJ28+hn08loKFi7Bh7R88cEcbZixZz8ED+/lv/+cZM2U2efMVoO8TPVkwdxb1r2nCG6/15dZ2XWjdoSsL5/3COwNfYsA7H6T7fSWHzZSNDKFwkSLUrFUbgJw5c1K+QkVidp6bkagq478ZR/uOXlyTmT/9SNWq1ahWvQYAefPlIzzcy1DzxWef8sTTfQAICwsjn49RDDQb1q/jmkaNAWjatBkTvjs3i584YTyly5SmUuXKgVIvAdc0akzevHkTyN5/bwS9evchW7ZsABQs6MVBz5EjR7wBPnniBCkkmUhT6tS/htxX5klV20pVa1CwcBEAylaoxMmTJzh18iTbt2ymZJmy5M1XAID611zHT1O8gGd/bVhLvWuaABDdoDE///h92t/EBSDiv4QqZpSDmC1bNrNyxXKi6taLl/06bw4FChbiqrLlANi4cQOIcFurm2h8dV3eeXMIAAcPHgSg/ysv0vjqunTv2ok9u3dn/E0AIsKtNzenQb06fPTh+wBUrlKVyZMmAvDtN1+zfbuX0OHYsWO8+cZgnnu+X0B0TS0b169n3tw5NGpQj2bXX8uSxYvj6xYtXEjtGlWoU6saQ9/9vwyZJfvjq5Hvc1uz+rzw1IMcOnjgvPrpUyZQsWoNsmbLRvFSZdi0cT07tm0hNjaWmT9MZleMl1SjfKVq8QZ6xrSJHDt6hIMH9mfovcSRFqE7g5VMa5RFpImITA60HhfL0aNH6dalI68PfpNcuXLFy78ZO4Z2HTvFn5+JjWXBr/P44ONRTJvxC5MnjueXn2dwJjaWHTu2U+/qhsyev5i69erz/HO9A3ErzJg1l/mLljJ+0hTeHzGcuXNm83/vf8T7/zecBvXqcOTIEbJmzQrAa6/045FHH+eKK64IiK6pJfZMLAcOHGD2vAW8PnAId9zekbgwudH16rFsxWrmzl/MkEED4n38gaDjnfcyZe7vjPvhVwoULMwbrz6XoH7jujW89fqL9BvwDgC5r8zDC6+/xdP/uYvu7W6kaLEShId7Pyq9nu/PkgVz6dCiIUsWzKNg4aLxdRlOCrPkUJ4pm085lYhIuE/s1HTl9OnTdLu9Ax06d6FVm7bx8tjYWCZN/I5ZcxfFy4pGFqNho8bxrolmzW9ixfLfaNzkenLkyMGtrbz8jW1ua8/nIz/JCPXPo2jRooD3iH9r6zYsWbyIx5/sxaQp3kunDevXM23qFMB7Ufbdt9/Q97lnOHTwIGFhYWS77DIe/M/DAdE9OSIji9Gm7W2ICHWjowkLC2Pfvn0UKFAgvk3FSpW4/PLLWb1qFVF1UsxKlC7kL1Aw/rjd7Xfx8F3nokjuitnB4/d14fW336N4qTLx8ibNbqZJs5sB+PqLj+PdYQULF+HtD74E4Pixo0yfMoGc7t1GIEgLwysiHwMtgT2qWtXJ8gJjgFLAZqCjqh5wdc/iZag+Azyqqj84eRTn4ilPAR5TVRWRbHiJVKOA/UAnVd3sT6egnimLSCkRWSsiH4rIKhH5QkRuEJF5IrJBRKJd+VVEfnOfFZIYZ6WIXCke+0Wkm5OPcuOVEpE5IrLMlQauvomI/CwiX+IFwg4XkSEislhEfheR+9P6nlWVhx+8j/IVKvHwo08kqJs18yfKla9AZLFzmWea3nAjq1eu5Pjx48TGxjJv7mwqVKyEiNDi5pbMmT0LgF9+nkmFipXSWt0UOXbsGEeOHIk/nvHTdCpXqRr/Yuzs2bMMGtCfe3t6X+VPP89m7YZNrN2wiYceeYynn3k26AwywK2t2jDr55mA96Ny6tQp8ufPz+ZNm4iNjQVgy5YtrF+/jpKlSgVMz727d8Ufz5g2ibIVPD/94UMHeah7ex7r8zK16l6doM/+fXsBOHTwAGM++5DbuniJMw78vY+zZ88C8OGw/9K2050ZcQvJkJLzItUW+1OgRSJZH2CGqpYDZrhzRKQyXpD6Kq7PcBEJd31GAD3xspGU8xmzB3BAVcsCbwGDUlIoFGbKZfGCRPcEFgO3A9cArfCyBHQDGqtqrIjcALwOtEs0xjy8LANbgL+ARni/XvXxMgacBZqp6gkRKQd8BcRNbaKBqqq6yWXDPaSqdd0v4DwR+dGlG4/HtesJULx4iQu62QXz5zHmy8+pXLVa/LK3F19+lRtb3Mw348bSvkPCxAVX5snDQ48+zvWN6iMiNGveguY33QLAS68N4P4e3Xm291Pkz5+fd9/76IJ0SQv27N5N5w63Ad5Mv2PnLtzYvAXv/u8d3hsxHIDWbdrSrfvdGa5baul2Rxfm/DKLffv2cVWpYrzw4st0v/se7r/3HqJqViVrlqx8+PFIRIRf583ljSEDyRKRhbCwMN753/AEq07Sk94P3c3iBXM4+Pd+mtatwENPPcfi+XNZu/p3RITIYiV4caC3xPKrT99n2+a/eO+dQbz3jmcn3vtiAvnyF2BQv96sW7MSgAce60OpMt77i8Xz5/LOwJcQgah6Den72ptJK5JBpMVMWVVnu0zWvrQGmrjjkcAs4BknH62qJ4FNLptItIhsBnKp6nxPL/kMaIOXfaQ18JIbaxwwTETEX0ooSSFdVEBxX9Z094sVd7M/qOoXIlIG+Ba4FRiK9+ukQBZVrSgiTYBeqtpSRLoC1fGM8gk8g9kO+FZV64lIbmAYUBPvsaS8quZwY/RT1evc9ce5cY47FXMD96vqj8ndQ63adXTWvIVp9ZVkOFkjgvphKlVk5AqI9GDjrqOBVuGSqVY859LUZJZOLdVrRunEGfP8timdP/sWYJ+P6H1VfT9xO2dnJvu4Lw6q6pU+9QdUNY+IDAMWqOrnTv4RnuHdDAxU1RucvBHwjLM9q4AWqrrd1f0J1FNVX70SEAoz5ZM+x2d9zs/i6f8q8LOqtnVf7qwkxpgNPASUAPoCbfGSGM5x9U8Au4EaeC4d3zczx3yOBXgkzo9kGEbgSIWLYl9a/hCQdARn9SP31ydZQn8a5M1W4xby3pVUA1XdBuQHyqnqX8BcoBfnjHJuIEZVz+IlUgxPahzgB+BBl5YcESkvIpenxU0YhnFhpOPqi90iUsS7hhQB9jj5dqC4T7tiwE4nL5aEPEEfEYnAszV+d3FlBqM8GBggIvNI3pgCLATWu+M5QCSecQYYDnQXkQVAeRLOjn35EC+T7TL3WPIeofG0YRiZC5fN2l+5BCYC3d1xd2CCj7yziGQTkdJ4LtNFqhoDHBGR+uL5yrol6hM3Vntgpj9/MgS5TzkzYD7lwGM+5cCT5j7lWlE6ZeZ8v22K582W4jVF5Cu8l3r58VyY/YDxwFg8d+dWoIOq/u3a9wXuAWKBx1V1qpPX4dySuKl4bk4VkcuAUUAtvBlyZ/e0niw2yzMMI+RIq9CdqtolmaqmybTvD/RPQr4EqJqE/ATe6rFUY0bZMIyQxLJZG4ZhBBGhHN/CH2aUDcMITTKnTTajbBhG6CGXvsIiaDGjbBhGSGLuC8MwjGAic9pkM8qGYYQmmdQmm1E2DCMUEcJCfFNQcphRNgwj5EirzSPBSOjvoTUMw8hE2EzZMIyQxNwXhmEYwUKIJ0f1hxllwzBCDsFWXxiGYQQVoR6SNTnMKBuGEZJkUptsRtkwjNAkk9pkM8qGYYQmmdV9Yemg0hkR2QtsScdL5CdhGvVQJNTvIdT1h/S/h5KqWiCtBhORaXg6+2OfqrZIq2tmFGaUQxwRWZLGadQznFC/h1DXHzLHPWQWbEefYRhGEGFG2TAMI4gwoxz6vB9oBdKAUL+HUNcfMsc9ZArMp2wYhhFE2EzZMAwjiDCjbBiGEUSYUTYMwwgizCgbmQoJgW1eoaCjETjMKBshi4gUFJGC7vhqEcmmofHmunigFbgY7MckYzCjnIn5F/wjKgOMFJGBQC/g8gDrkyxxfwsRKQ9MF5HHA6zSBSEiEveDJyI1RaT2v+D/r4BgAYkyMaqqItIUaALMAFar6t7AapV2qOoCETkMPAR0UNW/RSSLqp4OtG6JcX+LlkAHYBnQw+k6JMCqpQofg9wLuBk4ApwQkWdV9a+AKpfJsJlyJsRnVlYHGAwUBroD94pIZCB1u1REJPH/sxOBt4FXRaROnEEWkfAMV84PIpIPeBV4D7gLuBfoICKPBlKvlPCdDYtIA+B6Vb0eWAlcAWwKlG6ZFZspZ0LcrKw2MBR4WFXni8hNwHVANxH5XFW3BVbLC8c9Qp91x7fgGYXvVPW4iBwA3heRVkB5oBLwbuC09fB57D8LxABrVPWkiCwGxgCPiMg/qvpBQBVNgkQui/LAemCOiAwBqgGt4p7GVHVGIHXN6Mk8rgAAEMFJREFUTNhMORORyMd3GiiHNytDVafiuTD+v70zD7hrvNb47zEk0QSNoW7MSnENpWqepxJqDELE0BpDUGNNEYJyb82tmiq5RUrNXGLINeTimoKaZ4KqGmoOiUxP/1jv4fgu8iXE3uezfn+db599zl7fPuesvfazhndu4JeSOn/rBn5NmhzE7sBJwLbASEmL2T4VuAi4HTiV+F8ro+mzmAvA9rtEVHll+XsS8AxwLbBpcXq1oul8bwtcRXyfVgWWBXrbHlc+ixMkzVadpR2LdModiBK1rCGpr+3HgJ8By0kaUJ6/GbgBuMz2J1XaOrWUW+iNgTVs9wKuAc4vjvk0oBewse2nK7RR5bPYCLhV0n9J6g38GnhK0kOSfkVcPK4F3gRqWTUiaW1gHyIqvge4CXgHOELS8cB+wC6236nOyo5FOuUOQJOGvBJwIHCRpL1sP0xolz3LDwjbN1bpsKaUNprmLMCmwCJATwDbRwIjgCslLWr7MduvVWFrsbHhkJcnIvl9gfuICHMn2/0JDXwscQGZAPwEGFORyZ+jzfmeAehCVLn0B7D9B+ACQsr4AOhl+4kKTO2w5ECiDoKkNYCzgV2BJYD/AI62fU7Rl88DtgNeaJFa3raa5uxExn8icDihJ99i+5by/ABgqO2XKrJ1EaC77ZGSuhMXimdtb1Oc21bAKsCLwPlFB1+R0L13s/1IFXY30+Z8dwfG2x4taX0iIh5RZKJkGpKJvhZFUg9gW9unl00LArfZvg+4T9LTwAhJ420PlrSu7Q+qsndqaHIQBxJlfbMClwHnA3sA60nqZPsG28dXZmiwGPCupK6235U0CDhH0g62h0q6DJgRWAn4AfAS8DYhC/yjMqubaDrfBxBJ4c6SLrB9cQmg+5UGnROrtLOjk/JF6zILcHNxzgB/A3pImknSdEX/GwwcL2mLVnPIDUqSqSewBZEo29T268AQ4vu7sqTvVWgiALaHAU8BT0jqafsq4JfAIUXjN3AxcEIjmrf9Ql0ccgNJ/YDNibuqfwJDixR2C3HOlyxRdDKNyEi5RbH9jKQuwLmS3rF9gKRdiDrYMyTNSURmJwGbSLq2VWSLNnwCDCI69noQmnKDE4EZbH9chWEQ9dC2J0JUWEg6kTj/e9seJmkS8Iey34VEWVwtKd+nNwktfE9gemAtIlk50fZ5kkZUeb6/C2Sk3GI0J2JsjwVOA2aXNMD2TsBrQD/geEKvfJlI1tS+JbZNSV+DrsBQ4Ke2NyxlWHsBvwHG2K5kFWlJMwPYnqiYu7GnpB62zyUuIudJWqeUIu5HzZssSpXFWiXCnwHYEDjS9p3A9cAxkmZJhzztyUi5xWhqnV6IWEL9Gkn/CRxZHPNh8GmlwmrAQGCHRtNFnWnSNPsT9dTP2L6wlMEtVRKWaxIVJb2rKusr5/YYSfcCrxK39U8SHZOHFQ12EnCppO1tX19ep7rcrXyBLYsCO0u6zfbfJb0KrCTpZ4T+3b9VJbBWI51yi9BUarUSoRVfBOyuaC0eUErejpN0uu39iSqFHwI7lprl2tIm678K0RI+FFhD0jK2+0s6BtgJmIOYc/FUZQbH7+ZZ4gKxANDX9oOKIUP7lP/nL4pW7wmNF9XFIcP/awwZXaSJuYkk6s7AA8CKRIJ1h7pp3x2ZLIlrISStQOh9d9j+b0kLEJ1WN9g+StKShMZaeXlVe2njkJcFVgZetD1c0mJE3fWHwBFFuuhke1yF9k5ne1KRL7YC9geuaFR/KGZZbAKc4mjWqVWEDJ/KRNMV6WUk0XV4O/B7YEvgekdr/vTA92x/WKG53zkyUm4tVgI2A14rpUkvS9qSGAXZyfahUD8n8FU0OeR+wO5AdyKxdGdJZp4KHAmcVCLRyibAlfM6SdKGwDjgkvLUao3SN9u/K87sU627Lp+FpHltv1ruuJYhptVtR1S2zEVo4d2Idup7SgIzHfK3TEbKNaZJsvgh8HppONiKyIwPAu63Pb5EzPPZvqtSg6cSxRChHWz3lrQU0X58HdFkMUbRmDG6lMJVSrH1eCJyv76Uh21OXDAfsD24UgO/BEXzzRCiAeQ0SX8lWrxfAOYBbiQccF9gb2AZ4K26XFC+S2T1RY0pDnkjomHicEnDiB/SMKKrbXXFTN6XW8khN1dZFBlgXWCtUr3wOJGc3BTYT1IX289X5ZAl/Zukxcvj7wMHAFsDwyT9tNh+MxF1rl502TryEdHVubqkLYgk8CvA4sCOhHTRyfZxwI9sv5kOuRpSvqgxkpYgSr+2JuYkzAF0sX2GYq7wgPLcu9VZOWW00ZBntv1hqe2dEThF0oGO4fXHAocQjmRsRbZ2AjYC7i41vO8TIzh7E85sErABcCyRfL3BFc7d+CIa59v2WEm3EQngfYBZbA+R1I0IzvYDflU08Y8qNPk7T8oXNUNNzQiSFibqRUcRckUf2y9IWtX23ZLmt/1KlfZOLUUfXoZwbIOJyWO/IErhDrP9WomSK3HIDRTdgl0JyeIEYn29TYDhtm8rdzJ9gF1dsxVP2lwAuwDjiia+ESFR3GD77PJ8H+BO269WZ3EC6ZRrQyNqLI9XJ8rZPgHOJJJGKziGw6wJHEoMsWnJMiVJ2xG6eE/gcaJ64XBJCwEHAZ2JBhhXUV9d5JXmgfr/DuwAfB84w/azZfsGhP59iKNJpJaU6HcV4GPgYtu3Fse8O3Cv7d9WamDyOdIp14ASjd0EnEEss3MF8AjwOqFZdgJOJsY7Hg4cY/vaaqydcpoSlo1yskOAR4ka317EUJ5xRV/uBky0/WaVNgMULfkDQh7qRlwo5gb+CDwNXA6c5Zh7UUsUjTjbEHM4TiMG1B9o+ypJmxMllv2B91JDrgfplGtCKW07jMiADyi66sLErfIqRKv088Cttm9slbK3NrfQi9h+XlJfwsGNJgbSW9IRQGfbR1do64LA1rZPLlHwhcBw4qJ4AFGOtzsxEW4QUZ0wuk6fRePCVx53Jpzx5YQ0tBrwF2Ks64GObtCutlNDrhGZ6KsJtq+WNJqIktcH7iWy46OAeW0f0ti3Tk5gcjQ55H2BPorZvE8SyburgaVL08s2hERQJQYOVgxzGk8kUZ8F9iKi492Ijrf+QFfbo6A+dcjw6TJTlCh4DPAEcfHblOiEfEvSHkT35y22R1dnbfJFZKRcM0rEfBJwlO1LioZ8MvGjaskyJUm9gCMIpzCqbFuP0JQXIqaRHVXK4aqyccZS892DqJF+n6jZfQOYjXDEaxMXjrdqntTbjljd5E/AesBZxEKyQ4HliMqR02y/UY21yVeRkXLNKBHzBOACSdsA7wHHtvgPaCZgsO1Rjdvlkmwa0Xi+6oitOOS1iEqL3xPyxNa2zwTelnQWUbY3d13L3srjBYiIf7VSqbM9IYvNSCSO+wBbtPj3qUOTzSM1xPZ1xK3yIsDZpXOs9qM34UvHb04P7CtproZ+KWlXYF3bE6t0yA17Ja1MOOO+RFT5GjCwVC7gGBF6rO0HqrL1i2jjkPsDlwJHA+uUksKLiXr2bkSDy3JV3pEkkycj5ZpSsuMjXFYJbhXZoslBbAbMSaxV92eiauECxbS3JYhmhe2qsfIzSpJxRaJJZ3fb9ynauv9GLHZ6mKQ5bA+sm2QBnzvfmxMLsO5IJCOXJlZlucsxvKoL8Ihz/GbtyUi5xriFlm2X1F0xiKdRF3soUWt9CVF2dSlwS9n+c2B7Vzt+s5lZCb14vfL3y4RTfoGoWBhejVntQ9I8RJSP7eeINvUPiCl260iawfZltl+s0MyknaRTTr42JSo+BehWos4NiFnDbxC3zT2BVW2fbHtLYsHX2ixLb/t/iHrpXST1KRHxe0Q54ju276qzfGT778QI0Y2L/WMJTXw80RHaqUr7kikj5Yvka1FmJxxIzKiYnWh66UdEw5vZXkLSb4BBJYF5he0JX/qGFWH7WsVqIX9WDOz5GBho+/3yfK3loyJ3fQKcKIlSufNroLtzCaeWIiPl5GtRknTnEY74bqIb71VgQaLrDaJL8X7gdpe5HnWkJFh3IOYJP9ZIsNY5Sm6mdBYeCvxW0ja2J9h+q2q7kikjI+Xkm6ATsCRwAzGb92VCP95T0mWEk+tdh9bpyVGSYmOBIZJeciwk2jKUbs9dCD08aUGyeSSZYtp2FJZEU1dgY6Kc7DzHmnWLEXryMNvPV2Pt1KFYMPSFTI4l3zbplJOpprROz090vA0ERJRkzQNcZPu+Cs1LkpYkNeWk3Uiau0y0azQqbA6cTbTuHlq05CuJFuXekrq0ih6bJHUhNeWkXRSJ4jDgcUlDiNbpPsBORPfbwYqVOkYRjnqMKx5QnyStSMoXSbsoEe/ORKfYE8Sc5/mAfxCLnk6QtA8w3va51VmaJK1NyhfJZGlK7E0iZgn3JrL7SwJ3FIf8C2KJodsrMzRJOgAZKSftogym3x/YFdgDeJNYHmkT4Fbgx8TsiCcrMzJJOgDplJN2oVhd+kPbJxXteG9iYM99wAWEbPF+lTYmSUcg5YukvTwErCZpSdvjbJ8OzEu0Vo9Lh5wk3wxZfZG0lxHACsSSTrcR1Rf/BH6X4yCT5Jsj5Yuk3Uiam5im1guYABxk+7FqrUqSjkU65WSKkdSV+O7koptJ8g2TTjlJkqRGZKIvSZKkRqRTTpIkqRHplJMkSWpEOuUkSZIakU45SZKkRqRTTr5xJE2U9LCkxyVd3pjBPJXv9SdJW5fH50ta4iv2XVvSqlNxjJckzdHe7W32maKyQEnHSDp4Sm1MvjukU06mBWNsL2t7KWAcsajqp0iafmre1PZukxl4tDYxjyNJWpZ0ysm05k5gkRLF3i7pYuAxSdNLOknSSEmPStoTYkyopDMlPSlpGPCDxhtJGiFp+fK4p6SHJD0i6VZJCxLO/4ASpa8haU5JV5ZjjJS0Wnnt7JKGS/qrpHOJZay+EknXSHpQ0hOS9mjz3CnFllslzVm2LSzppvKaOyUt/k2czKTjk7MvkmmGpBmAjYCbyqYVgaVsjyqO7X3bK0jqDPyfpOHAT4iZzUsDcwFPAkPavO+cwB+BNct7zWb7HUnnAKNtn1z2uxg4zfZdkuYHbiYWdj0auMv2sZJ+TowinRy7lGPMBIyUdKXtt4kFYx+yfZCkgeW99wHOA/rZfk7SSsBZxMIASfKVpFNOpgUzSXq4PL4TGEzICvfbHlW2bwD8uKEXA7MCPwLWBC6xPRF4rQw/asvKxHD9UQC23/kSO9YHlmhaJnAWSTOXY/Qqrx0m6d12/E/7SdqyPJ6v2Po2Mfj/0rJ9KHCVpG7l/7286did23GMJEmnnEwTxthetnlDcU4fNW8C9rV9c5v9NgYm1/uvduwDIc+tYnvMF9jS7vkCktYmHPwqtj+WNALo8iW7uxz3vbbnIEnaQ2rKSVXcDOwlaUYASYuWQUd3ANsVzbkHsM4XvPYeYC1JC5XXzla2fwjM3LTfcEJKoOzXcJJ3AH3Lto2A7pOxdVbg3eKQFyci9QbTAY1of3tCFvkAGCVpm3IMSVpmMsdIEiCdclId5xN68UOSHgfOJe7crgaeAx4jVsX+37YvtP0WoQNfJekRPpMPrgO2bCT6gP2A5Usi8Uk+qwIZBKwp6SFCRnllMrbeBMwg6VHgOODepuc+ApaU9CChGR9btvcFdi32PQFs3o5zkiQ5JS5JkqROZKScJElSI9IpJ0mS1Ih0ykmSJDUinXKSJEmNSKecJElSI9IpJ0mS1Ih0ykmSJDXiX5DXEgJxDVRRAAAAAElFTkSuQmCC\n"},"metadata":{"needs_background":"light"}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"gbdt = GradientBoostingClassifier(n_estimators=100,max_features='sqrt')\ngbdt.fit(X_train,y_train)\ny_pred = gbdt.predict(X_test)\nprint(classification_report(y_test,y_pred))\n\nscore = metrics.accuracy_score(y_test, y_pred)\nprint(\"accuracy:   %0.3f\" % score)","execution_count":46,"outputs":[{"output_type":"stream","text":"              precision    recall  f1-score   support\n\n           0       0.95      0.99      0.97     85621\n           1       0.90      0.97      0.93     19292\n           2       0.93      0.79      0.85      6504\n           3       0.88      0.73      0.80     18822\n\n    accuracy                           0.94    130239\n   macro avg       0.92      0.87      0.89    130239\nweighted avg       0.93      0.94      0.93    130239\n\naccuracy:   0.936\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"CM=confusion_matrix(y_test,y_pred,labels=[0,1,2,3])\n\nplot_confusion_matrix(cm,classes=['benign', 'defacement','phishing','malware'])","execution_count":47,"outputs":[{"output_type":"stream","text":"Confusion matrix, without normalization\n","name":"stdout"},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 2 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"gbdt_feature = gbdt.feature_importances_\ngbdt_feature","execution_count":56,"outputs":[{"output_type":"execute_result","execution_count":56,"data":{"text/plain":"array([0.01422164, 0.05206413, 0.        , 0.02868465, 0.13249984,\n       0.00071255, 0.12090048, 0.00017663, 0.0003583 , 0.02346279,\n       0.09642648, 0.00730826, 0.01089888, 0.02327172, 0.06502678,\n       0.04349845, 0.15375502, 0.0091799 , 0.03902921, 0.1139888 ,\n       0.02779176, 0.03674371])"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"xgb_feature = model.feature_importances_\nxgb_feature","execution_count":57,"outputs":[{"output_type":"execute_result","execution_count":57,"data":{"text/plain":"array([0.18995441, 0.00864374, 0.        , 0.01276796, 0.12045407,\n       0.01123682, 0.0595661 , 0.00428362, 0.0046364 , 0.05147118,\n       0.1236386 , 0.02176145, 0.00992822, 0.02163003, 0.01570353,\n       0.0258724 , 0.08584396, 0.0270134 , 0.01704087, 0.16727157,\n       0.01070208, 0.01057967], dtype=float32)"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"lgb_features = [  31,   57,    0,  992,  529,   78, 1168,   41,  124,  310,  162,\n        274,  191,  591,  445,  995, 1496,  299, 1765,  636, 1101,  715]\n\nxgb_features = [0.18995441, 0.00864374, 0.        , 0.01276796, 0.12045407,\n       0.01123682, 0.0595661 , 0.00428362, 0.0046364 , 0.05147118,\n       0.1236386 , 0.02176145, 0.00992822, 0.02163003, 0.01570353,\n       0.0258724 , 0.08584396, 0.0270134 , 0.01704087, 0.16727157,\n       0.01070208, 0.01057967]\n\ngbdt_features = [0.01422164, 0.05206413, 0.        , 0.02868465, 0.13249984,\n       0.00071255, 0.12090048, 0.00017663, 0.0003583 , 0.02346279,\n       0.09642648, 0.00730826, 0.01089888, 0.02327172, 0.06502678,\n       0.04349845, 0.15375502, 0.0091799 , 0.03902921, 0.1139888 ,\n       0.02779176, 0.03674371]","execution_count":58,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cols = X_train.columns\n# Create a dataframe with features\nfeature_dataframe = pd.DataFrame( {'features': cols,\n    \n    'Gradient Boost feature importances': gbdt_features,\n    'XG Boost feature importances': xgb_features,\n    'LGBM feature importances': lgb_features\n                                   \n    })\nfeature_dataframe['mean'] = feature_dataframe.mean(axis= 1) # axis = 1 computes the mean row-wise\nfeature_dataframe.head(3)","execution_count":59,"outputs":[{"output_type":"error","ename":"ValueError","evalue":"arrays must all be same length","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)","\u001b[0;32m<ipython-input-59-f0a943df1292>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0;34m'Gradient Boost feature importances'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mgbdt_features\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m     \u001b[0;34m'XG Boost feature importances'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mxgb_features\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m     \u001b[0;34m'LGBM feature importances'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mlgb_features\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      8\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m     })\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pandas/core/frame.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, data, index, columns, dtype, copy)\u001b[0m\n\u001b[1;32m    433\u001b[0m             )\n\u001b[1;32m    434\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 435\u001b[0;31m             \u001b[0mmgr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0minit_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolumns\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    436\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mma\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMaskedArray\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    437\u001b[0m             \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mma\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmrecords\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mmrecords\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36minit_dict\u001b[0;34m(data, index, columns, dtype)\u001b[0m\n\u001b[1;32m    252\u001b[0m             \u001b[0marr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_datetime64tz_dtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marr\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0marr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0marr\u001b[0m \u001b[0;32min\u001b[0m \u001b[0marrays\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    253\u001b[0m         ]\n\u001b[0;32m--> 254\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0marrays_to_mgr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_names\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolumns\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    255\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    256\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36marrays_to_mgr\u001b[0;34m(arrays, arr_names, index, columns, dtype)\u001b[0m\n\u001b[1;32m     62\u001b[0m     \u001b[0;31m# figure out the index, if necessary\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     63\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mindex\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 64\u001b[0;31m         \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mextract_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     65\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m         \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mensure_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36mextract_index\u001b[0;34m(data)\u001b[0m\n\u001b[1;32m    363\u001b[0m             \u001b[0mlengths\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mraw_lengths\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    364\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlengths\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 365\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"arrays must all be same length\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    366\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    367\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhave_dicts\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mValueError\u001b[0m: arrays must all be same length"]}]},{"metadata":{"trusted":true},"cell_type":"code","source":"import seaborn as sns\nimport matplotlib.pyplot as plt\n%matplotlib inline\ndef plot_feature_importance():\n    tmp = pd.DataFrame({'Feature': X_test.columns, 'Feature importance': feature_dataframe['mean'].values})\n    tmp = tmp.sort_values(by='Feature importance',ascending=False).head(20)\n    plt.figure(figsize = (10,12))\n    plt.title('Average Feature Importance Top 20 Features',fontsize=14)\n    s = sns.barplot(y='Feature',x='Feature importance',data=tmp, orient='h')\n    s.set_xticklabels(s.get_xticklabels(),rotation=90)\n    plt.show()\nplot_feature_importance()","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":4}