{"id":532,"date":"2018-05-18T00:24:34","date_gmt":"2018-05-17T15:24:34","guid":{"rendered":"http:\/\/www.typea.info\/blog\/index.php\/2018\/05\/18\/python_5\/"},"modified":"2018-05-18T00:24:34","modified_gmt":"2018-05-17T15:24:34","slug":"python_5","status":"publish","type":"post","link":"https:\/\/www.typea.info\/blog\/index.php\/2018\/05\/18\/python_5\/","title":{"rendered":"Python \u6570\u5b66\u3001\u7d71\u8a08\u3001\u6a5f\u68b0\u5b66\u7fd2\uff1a\u57fa\u672c\u7d71\u8a08\u91cf\u3001\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0"},"content":{"rendered":"

<\/a><\/p>\n

1.\u6e96\u5099<\/h2>\n

Python\u74b0\u5883\u306b\u3001\u4ee5\u4e0b\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3002<\/p>\n

pip install {\u30d1\u30c3\u30b1\u30fc\u30b8}\"\"<\/p>\n\n\n\n\n\n\n\n\n
\u30d1\u30c3\u30b1\u30fc\u30b8<\/td>\n\u5185\u5bb9<\/td>\n<\/tr>\n
numpy<\/a><\/td>\n\u6570\u5024\u8a08\u7b97\u3092\u52b9\u7387\u7684\u306b\u884c\u3046\u305f\u3081\u306e\u62e1\u5f35\u30e2\u30b8\u30e5\u30fc\u30eb<\/td>\n<\/tr>\n
matplotlib<\/a><\/td>\n\u30b0\u30e9\u30d5\u3092\u63cf\u753b\u3059\u308b\u3068\u304d\u306a\u3069\u306b\u4f7f\u308f\u308c\u308b\u6a19\u6e96\u7684\u306a\u30e9\u30a4\u30d6\u30e9\u30ea<\/td>\n<\/tr>\n
scikit-learn<\/a><\/td>\n\u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u6a5f\u68b0\u5b66\u7fd2\u30e9\u30a4\u30d6\u30e9\u30ea<\/td>\n<\/tr>\n
pandas<\/a><\/td>\n\u30c7\u30fc\u30bf\u89e3\u6790\u3092\u652f\u63f4\u3059\u308b\u6a5f\u80fd\u3092\u63d0\u4f9b\u3059\u308b\u30e9\u30a4\u30d6\u30e9\u30ea<\/td>\n<\/tr>\n
scipy<\/a><\/td>\n\u79d1\u5b66\u6280\u8853\u8a08\u7b97\u30eb\u30fc\u30c1\u30f3\u306e\u4e2d\u6838\u3068\u306a\u308b\u30d1\u30c3\u30b1\u30fc\u30b8<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

2.\u30b5\u30f3\u30d7\u30eb\u30bd\u30fc\u30b9<\/h2>\n

2.1 \u5bfe\u8c61\u30c7\u30fc\u30bf<\/h3>\n

scikit-learn \u30d1\u30c3\u30b1\u30fc\u30b8\u3092import\u3059\u308b\u3068\u3001Boston \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8<\/a> \u3068\u3044\u3046\u30dc\u30b9\u30c8\u30f3\u5e02\u306e\u4f4f\u5b85\u4fa1\u683c\u306e\u30c7\u30fc\u30bf\u3092\u5229\u7528\u3067\u304d\u308b\u3002\u4eba\u5de5\u77e5\u80fd\u30d7\u30ed\u30b0\u30e9\u30df\u30f3\u30b0\u306e\u305f\u3081\u306e\u6570\u5b66\u304c\u308f\u304b\u308b\u672c<\/a> \u304b\u3089\u3001\u5404\u30ab\u30e9\u30e0\u8aac\u660e\u3092\u5f15\u7528\u3002<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n

\u30ab\u30e9\u30e0<\/p>\n<\/td>\n

\n

\u30dc\u30b9\u30c8\u30f3\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u5185\u5bb9<\/p>\n<\/td>\n<\/tr>\n

CRIM<\/td>\n\u4eba\u53e3\u4e00\u4eba\u5f53\u305f\u308a\u306e\u72af\u7f6a\u767a\u751f\u56de\u6570<\/td>\n<\/tr>\n
ZN<\/td>\n25,000\u5e73\u65b9\u30d5\u30a3\u30fc\u30c8\u4ee5\u4e0a\u306e\u4f4f\u5c45\u533a\u753b\u306e\u5360\u3081\u308b\u5272\u5408<\/td>\n<\/tr>\n
INDUS<\/td>\n\u5c0f\u58f2\u696d\u4ee5\u5916\u306e\u5546\u696d\u304c\u5360\u3081\u308b\u9762\u7a4d\u306e\u5272\u5408<\/td>\n<\/tr>\n
CHAS<\/td>\n\u30c1\u30e3\u30fc\u30eb\u30ba\u5ddd\u306b\u3088\u308b\u30c0\u30df\u30fc\u5909\u6570<\/td>\n<\/tr>\n
NOX<\/td>\nNOx\u306e\u6fc3\u5ea6<\/td>\n<\/tr>\n
RM<\/td>\n\u4f4f\u5c45\u306e\u5e73\u5747\u90e8\u5c4b\u6570<\/td>\n<\/tr>\n
AGE<\/td>\n1940\u5e74\u3088\u308a\u524d\u306b\u5efa\u3066\u3089\u308c\u305f\u7269\u4ef6\u306e\u5272\u5408<\/td>\n<\/tr>\n
DIS<\/td>\n5\u3064\u306e\u30dc\u30b9\u30c8\u30f3\u5e02\u306e\u96c7\u7528\u65bd\u8a2d\u304b\u3089\u306e\u8ddd\u96e2<\/td>\n<\/tr>\n
RAD<\/td>\n\u74b0\u72b6\u9ad8\u901f\u9053\u8def\u3078\u306e\u30a2\u30af\u30bb\u30b9\u306e\u3057\u3084\u3059\u3055(1-24\u306e\u9593\u9694\u5c3a\u5ea6)<\/td>\n<\/tr>\n
TAX<\/td>\n$10,000\u3042\u305f\u308a\u306e\u4e0d\u52d5\u7523\u7a0e\u7387\u306e\u7dcf\u8a08($)<\/td>\n<\/tr>\n
PTRATIO<\/td>\n\u6559\u5e2b1\u4eba\u304c\u6301\u3064\u751f\u5f92\u306e\u6570<\/td>\n<\/tr>\n
B<\/td>\n1,000\u00d7(\u9ed2\u4eba\u6bd4\u7387(%)-0.63) ^2\u306e\u5f0f\u3067\u8a08\u7b97\u3055\u308c\u308b\u6307\u6570<\/td>\n<\/tr>\n
LSTAT<\/td>\n\u7d66\u4e0e\u306e\u4f4e\u3044\u8077\u696d\u306b\u5f93\u4e8b\u3059\u308b\u4eba\u53e3\u306e\u5272\u5408<\/td>\n<\/tr>\n
MEDV<\/td>\n\u4f4f\u5b85\u4fa1\u683c\u306e\u4e2d\u592e\u5024<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

2.2 \u30bd\u30fc\u30b9\u30b3\u30fc\u30c9<\/h3>\n

scikit-learn \u306e Boston \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u30ed\u30fc\u30c9\u3001\u6307\u5b9a\u3057\u305f\u30ab\u30e9\u30e0\u306e\u57fa\u672c\u7d71\u8a08\u91cf<\/a>\u3068\u3001\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u3092\u8868\u793a\u3055\u305b\u308b\u3002<\/p>\n

from sklearn import datasets\r\nimport pandas as pd\r\nimport matplotlib.pyplot as plt\r\nimport math\r\n\r\ntitle_dict = {'CRIM':      'per capita crime rate by town'\r\n              ,'ZN':       'proportion of residential land zoned for lots over 25,000 sq.ft.'\r\n              ,'INDUS':    'proportion of non-retail business acres per town'\r\n              ,'CHAS':     'Charles River dummy variable (= 1 if tract bounds river; 0 otherwise)'\r\n              ,'NOX':      'nitric oxides concentration (parts per 10 million)'\r\n              ,'RM':       'average number of rooms per dwelling'\r\n              ,'AGE':      'proportion of owner-occupied units built prior to 1940'\r\n              ,'DIS':      'weighted distances to five Boston employment centres'\r\n              ,'RAD':      'index of accessibility to radial highways'\r\n              ,'TAX':      'full-value property-tax rate per $10,000'\r\n              ,'PTRATIO':  'pupil-teacher ratio by town'\r\n              ,'B':        '1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town'\r\n              ,'LSTAT':    '% lower status of the population'\r\n              ,'MEDV':     'Median value of owner-occupied homes in $1000\\'s'}\r\n\r\ndef basic_statistics(feature_index):\r\n    \"\"\"\r\n    \u57fa\u672c\u7d71\u8a08\u91cf\u3068\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\r\n    :param feature_index:feature_index:['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT']\r\n    :return:\r\n    \"\"\"\r\n    boston = datasets.load_boston()\r\n    df = pd.DataFrame(boston.data)\r\n    df.columns = boston.feature_names\r\n\r\n    x = df[feature_index]\r\n    print(x.describe())\r\n\r\n    from pprint import  pprint\r\n    print(pprint(dir(x)))\r\n\r\n    desc = \"\"\"\r\n    count={0}\r\n    mean={1:f}\r\n    median={2:f}\r\n    std={3:f}\r\n    min={4:f}\r\n    25%={5:f}\r\n    50%={6:f}\r\n    75%={7:f}\r\n    max={8:f}\r\n    \"\"\".format(\r\n        x.count(),\r\n        x.mean(),\r\n        x.median(),\r\n        x.std(),\r\n        x.min(),\r\n        x.quantile(.25),\r\n        x.quantile(.5),\r\n        x.quantile(.75),\r\n        x.max())\r\n    \r\n    # \u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u306e\u67f1\u306e\u6570\u3092\u30c7\u30fc\u30bf\u6570\u306e\u5e73\u65b9\u6839\u3068\u3059\u308b\r\n    plt.hist(x, bins=int(round(math.sqrt(x.count()),0)),\r\n             rwidth=0.8, label=desc)\r\n\r\n    plt.title(title_dict[feature_index])\r\n    plt.legend()\r\n    plt.show()\r\n\r\n\r\nif __name__ == \"__main__\":\r\n    basic_statistics('RM')<\/pre>\n
2.3 \u51fa\u529b\u3055\u308c\u305f\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0<\/h3>\n
\"hist01\"<\/a><\/p>\n
3.\u78ba\u8a8d<\/h2>\n
3.1 \u57fa\u672c\u7d71\u8a08\u91cf<\/h3>\n
<\/h2>\n
http:\/\/www.trunktools.jp\/services\/analysis\/basis.html<\/a><\/p>\n
\u4e0a\u8a18\u3001x.describe() \u306b\u3066\u57fa\u672c\u7d71\u3046\u3088\uff42\u8a08\u91cf\u3092\u4e00\u62ec\u3067\u51fa\u529b\u3002\u500b\u5225\u306b\u53d6\u5f97\u3059\u308b\u5834\u5408\u306b\u306f\u3001\u30e1\u30bd\u30c3\u30c9\u3092\u547c\u3076\u3002<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n
\u9805\u76ee<\/td>\n\u610f\u5473<\/td>\n\u500b\u5225\u306b\u53d6\u5f97<\/td>\n<\/tr>\n
count<\/td>\n\u6a19\u672c\u6570<\/td>\ncount()<\/td>\n<\/tr>\n
mean<\/td>\n\u5e73\u5747<\/td>\nmean()<\/td>\n<\/tr>\n
std<\/td>\n\u6a19\u6e96\u504f\u5dee<\/td>\nstd()<\/td>\n<\/tr>\n
min<\/td>\n\u6700\u5c0f\u5024<\/td>\nmin()<\/td>\n<\/tr>\n
25%<\/td>\n\u7b2c1\u56db\u5206\u4f4d<\/td>\nquantile(.25)<\/td>\n<\/tr>\n
50%<\/td>\n\u4e2d\u592e\u5024(\u7b2c2\u56db\u5206\u4f4d)<\/td>\nquantile(.5)<\/td>\n<\/tr>\n
75%<\/td>\n\u7b2c3\u56db\u5206\u4f4d<\/td>\nquantile(.75)<\/td>\n<\/tr>\n
max<\/td>\n\u6700\u5927\u5024(\u7b2c4\u56db\u5206\u4f4d)<\/td>\nmax()<\/td>\n<\/tr>\n
<\/td>\n<\/td>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
\u56db\u5206\u4f4d<\/p>\n
3.2 \u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u306e\u51fa\u529b<\/h3>\n
bins \u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u306e\u68d2\u306e\u6570\u306b\u3064\u3044\u3066\u3001EXCEL\u30d3\u30b8\u30cd\u30b9\u7d71\u8a08\u5206\u6790<\/a><\/p>\n
\n
\u7d76\u5bfe\u3053\u3046\u3057\u306a\u3051\u308c\u3070\u3044\u3051\u306a\u3044\u3068\u3044\u3046\u6c7a\u3081\u4e8b\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u4e00\u822c\u7684\u306b\u5b9f\u52d9\u306e\u4e16\u754c\u3067\u306f\u3001\u30c7\u30fc\u30bf\u306e\u500b\u6570\u306e\u5e73\u65b9\u6839\u3092\u4f7f\u3063\u3066\u533a\u9593\u306e\u6570\u3092\u6c7a\u3081\u307e\u3059\u3002\u5c02\u9580\u7684\u306b\u306f\u30b9\u30bf\u30fc\u30b8\u30a7\u30b9\u306e\u516c\u5f0f\u306a\u3069\u304c\u3042\u308a\u307e\u3059\u304c\u3001\u3053\u3053\u3067\u306f\u3088\u308a\u7c21\u5358\u3067\u4fbf\u5229\u306a\u65b9\u6cd5\u3092\u4f7f\u3063\u3066\u3044\u307e\u3059\u3002<\/p>\n<\/blockquote>\n
\u306b\u5f93\u3046\u3002\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u306e\u7bc4\u56f2\u306e\u6c7a\u3081\u65b9<\/a><\/p>\n
https:\/\/pythondatascience.plavox.info\/matplotlib\/%E3%83%92%E3%82%B9%E3%83%88%E3%82%B0%E3%83%A9%E3%83%A0<\/a><\/p>\n
matplotlib.pyplot.hist(x, bins=10, range=None, normed=False, weights=None,\r\n                       cumulative=False, bottom=None, histtype='bar',\r\n                       align='mid', orientation='vertical', rwidth=None,\r\n                       log=False, color=None, label=None, stacked=False,\r\n                       hold=None, data=None, **kwargs)<\/pre>\n<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
x (\u5fc5\u9808)<\/td>\n\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u3092\u4f5c\u6210\u3059\u308b\u305f\u3081\u306e\u751f\u30c7\u30fc\u30bf\u306e\u914d\u5217\u3002<\/td>\n<\/tr>\n
bins<\/td>\n\u30d3\u30f3 (\u8868\u793a\u3059\u308b\u68d2) \u306e\u6570\u3002\u968e\u7d1a\u6570\u3002(\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: 10)<\/td>\n<\/tr>\n
range<\/td>\n\u30d3\u30f3\u306e\u6700\u5c0f\u5024\u3068\u6700\u5927\u5024\u3092\u6307\u5b9a\u3002(\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: (x.min(), x.max()))<\/td>\n<\/tr>\n
normed<\/td>\nTrue \u306b\u8a2d\u5b9a\u3059\u308b\u3068\u6b63\u898f\u5316 (\u5408\u8a08\u5024\u304c 1 \u306b\u306a\u308b\u3088\u3046\u306b\u5909\u63db) \u3092\u5b9f\u65bd\u3002 (\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: False)<\/td>\n<\/tr>\n
cumulative<\/td>\nTrue \u306b\u8a2d\u5b9a\u3059\u308b\u3068\u3001\u7d2f\u7a4d\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u3092\u51fa\u529b\u3002 (\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: False)<\/td>\n<\/tr>\n
bottom<\/td>\n\u5404\u68d2\u306e\u4e0b\u5074\u306e\u4f59\u767d\u3092\u6570\u5024\u307e\u305f\u306f\u914d\u5217\u3067\u6307\u5b9a\u3002<\/td>\n<\/tr>\n
histtype<\/td>\n\u2018bar\u2019 (\u901a\u5e38\u306e\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0), \u2018barstacked\u2019 (\u7a4d\u307f\u4e0a\u3052\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0), \u2018step\u2019 (\u7dda), \u2018stepfilled \u2018 (\u5857\u308a\u3064\u3076\u3057\u3042\u308a\u306e\u7dda) \u304b\u3089\u9078\u629e\u3002 (\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: \u2018bar\u2019)<\/td>\n<\/tr>\n
align<\/td>\n\u5404\u68d2\u306e\u4e2d\u5fc3\u3092 X \u8ef8\u76ee\u76db\u4e0a\u306e\u3069\u306e\u6a2a\u4f4d\u7f6e\u3067\u51fa\u529b\u3059\u308b\u304b\u3002 \u2018left\u2019, \u2018mid\u2019, \u2018right\u2019 \u304b\u3089\u9078\u629e\u3002(\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: \u2018mid\u2019)<\/td>\n<\/tr>\n
orientation<\/td>\n\u68d2\u306e\u65b9\u5411\u3002\u2019horizontal\u2019 (\u6c34\u5e73\u65b9\u5411), \u2018vertical\u2019 (\u5782\u76f4\u65b9\u5411) \u304b\u3089\u9078\u629e\u3002(\u30c7\u30d5\u30a9\u30eb\u30c8\u5024: \u2018vertical\u2019)<\/td>\n<\/tr>\n
rwidth<\/td>\n\u5404\u68d2\u306e\u5e45\u3092\u6570\u5024\u307e\u305f\u306f\u3001\u914d\u5217\u3067\u6307\u5b9a\u3002<\/td>\n<\/tr>\n
log<\/td>\nTrue \u306b\u8a2d\u5b9a\u3059\u308b\u3068\u3001\u7e26\u8ef8\u3092\u5bfe\u6570\u76ee\u76db\u3067\u8868\u793a\u3057\u307e\u3059\u3002<\/td>\n<\/tr>\n
color<\/td>\n\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u306e\u8272\u3002\u914d\u5217\u3067\u6307\u5b9a\u3057\u3001\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u5358\u4f4d\u3067\u8272\u3092\u6307\u5b9a\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/td>\n<\/tr>\n
label<\/td>\n\u51e1\u4f8b\u3092\u8f09\u305b\u308b\u969b\u306b\u4f7f\u7528\u3057\u307e\u3059\u3002<\/td>\n<\/tr>\n
stacked<\/td>\nTrue \u306b\u8a2d\u5b9a\u3059\u308b\u3068\u7a4d\u307f\u4e0a\u3052\u30d2\u30b9\u30c8\u30b0\u30e9\u30e0\u3067\u51fa\u529b\u3057\u307e\u3059\u3002False \u306b\u8a2d\u5b9a\u3059\u308b\u3068\u3001\u6a2a\u306b\u4e26\u3079\u3066\u51fa\u529b\u3057\u307e\u3059\u3002<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
<\/p>\n
<\/a>\"\"<\/p>\n","protected":false},"excerpt":{"rendered":"
1.\u6e96\u5099 Python\u74b0\u5883\u306b\u3001\u4ee5\u4e0b\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3002 pip install {\u30d1\u30c3\u30b1\u30fc\u30b8} \u30d1\u30c3\u30b1\u30fc\u30b8 \u5185\u5bb9 numpy \u6570\u5024\u8a08\u7b97\u3092\u52b9\u7387\u7684\u306b\u884c\u3046\u305f\u3081\u306e\u62e1\u5f35\u30e2\u30b8\u30e5\u30fc\u30eb matplotlib \u30b0\u30e9\u30d5\u3092\u63cf\u753b\u3059\u308b\u3068\u304d […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"sns_share_botton_hide":"","vkExUnit_sns_title":"","_vk_print_noindex":"","sitemap_hide":"","_veu_custom_css":"","veu_display_promotion_alert":"","vkexunit_cta_each_option":"","footnotes":""},"categories":[35,86],"tags":[],"class_list":["post-532","post","type-post","status-publish","format-standard","hentry","category-math","category-statistics"],"veu_head_title_object":{"title":"","add_site_title":""},"_links":{"self":[{"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/posts\/532","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=532"}],"version-history":[{"count":0,"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/posts\/532\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=532"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=532"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.typea.info\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=532"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}