dj_pandas_tutoral.ipynb

{

 "metadata": {

  "name": "",

  "signature": "sha256:417b5f27bd29d6f386355063d8a8f870c73e9a747409dd9ce5541b138d881187"

 },

 "nbformat": 3,

 "nbformat_minor": 0,

 "worksheets": [

  {

   "cells": [

    {

     "cell_type": "markdown",

     "metadata": {},

     "source": [

      "\u0423\u0447\u0435\u0431\u043d\u044b\u0439 \u043d\u043e\u0443\u0442\u0431\u0443\u043a \u043f\u043e \u0442\u0435\u043c\u0435 \"\u041f\u0430\u043a\u0435\u0442 Pandas\"\n",

      "\n",

      "\u0430\u0432\u0442\u043e\u0440: \u0410\u043b\u0435\u043a\u0441\u0430\u043d\u0434\u0440 \u0414\u044c\u044f\u043a\u043e\u043d\u043e\u0432 (alexanderdyakonov.wordpress.com)"

     ]

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043f\u043e\u0434\u0433\u0440\u0443\u0436\u0430\u0435\u043c \u0432\u0441\u0435 \u043d\u0443\u0436\u043d\u044b\u0435 \u043f\u0430\u043a\u0435\u0442\u044b\n",

      "import pandas as pd\n",

      "import numpy as np\n",

      "\n",

      "\n",

      "# \u0434\u043b\u044f \u0432\u0441\u0442\u0440\u043e\u0435\u043d\u043d\u044b\u0445 \u043a\u0430\u0440\u0442\u0438\u043d\u043e\u043a\n",

      "%pylab inline\n",

      "# \u0447\u0443\u0442\u044c \u043f\u043e\u043a\u0440\u0430\u0441\u0438\u0432\u0448\u0435 \u043a\u0430\u0440\u0442\u0438\u043d\u043a\u0438:\n",

      "pd.set_option('display.mpl_style', 'default')\n",

      "figsize(10, 3)"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "Populating the interactive namespace from numpy and matplotlib\n"

       ]

      }

     ],

     "prompt_number": 2

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0437\u0430\u0433\u0440\u0443\u0437\u043a\u0430 \u0434\u0430\u043d\u043d\u044b\u0445\n",

      "\n",

      "# Excel\n",

      "data2 = pd.read_excel('D:\\\\filename.xlsx', sheetname='1')\n",

      "# csv-\u0444\u0430\u0439\u043b\n",

      "data = pd.read_csv('D:\\\\filename.csv', sep=';', decimal=',')\n",

      "data.to_csv('foo.csv') # \u0441\u043e\u0445\u0440\u0430\u043d\u0435\u043d\u0438\u0435\n",

      "# HDF5\n",

      "pd.read_hdf('foo.h5', 'df')\n",

      "df.to_hdf('foo.h5', 'df') # \u0441\u043e\u0445\u0440\u0430\u043d\u0435\u043d\u0438\u0435"

     ],

     "language": "python",

     "metadata": {},

     "outputs": []

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0435 \u0434\u0430\u0442\u0430-\u0444\u0440\u0435\u0439\u043c\u0430\n",

      "\n",

      "data = pd.DataFrame({ 'A' : [1., 4., 2., 1.],\n",

      "'B' : pd.Timestamp('20130102'),\n",

      "'C' : pd.Series(1,index=list(range(4)),dtype='float32'),\n",

      "'D' : np.array([3] * 4,dtype='int32'),\n",

      "'E' : pd.Categorical([\"test\",\"train\",\"test\",\"train\"]),\n",

      "'F' : 'foo' }, index=pd.period_range('Jan-2000', periods=4, freq='M'))\n",

      "print data\n",

      "\n",

      "# \u0432\u0442\u043e\u0440\u043e\u0439 \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "tmp = dict([('A',[1., np.nan, 2., 1.]), ('B',[2.2, np.nan, np.nan, 0.0])]) # \u0435\u0449\u0451 \u043e\u0434\u0438\u043d \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "data2 = pd.DataFrame(tmp)\n",

      "print data2"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "         A          B   C  D      E    F\n",

        "2000-01  1 2013-01-02 NaN  3   test  foo\n",

        "2000-02  4 2013-01-02 NaN  3  train  foo\n",

        "2000-03  2 2013-01-02 NaN  3   test  foo\n",

        "2000-04  1 2013-01-02 NaN  3  train  foo\n",

        "    A    B\n",

        "0   1  2.2\n",

        "1 NaN  NaN\n",

        "2   2  NaN\n",

        "3   1  0.0\n"

       ]

      }

     ],

     "prompt_number": 50

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043f\u0440\u043e\u0441\u0442\u0435\u0439\u0448\u0438\u0435 \u043e\u043f\u0435\u0440\u0430\u0446\u0438\u0438\n",

      "# \u0441\u0442\u043e\u043b\u0431\u0446\u044b\n",

      "print data.columns\n",

      "# \u0441\u0442\u0440\u043e\u043a\u0438 - \u043d\u043e \u0442\u0443\u0442 \u0432\u0440\u0435\u043c\u0435\u043d\u043d\u0430\u044f \u0438\u043d\u0434\u0435\u043a\u0441\u0430\u0446\u0438\u044f\n",

      "print data.index\n",

      "# \u0441\u0430\u043c\u0438 \u0434\u0430\u043d\u043d\u044b\u0435\n",

      "print data\n",

      "# \u0441\u043e\u0440\u0442\u0438\u0440\u043e\u0432\u043a\u0430\n",

      "print data.sort(columns='A')\n",

      "# \u0442\u0438\u043f\u044b\n",

      "print data.dtypes\n",

      "# \u0441\u0442\u0430\u0442\u0438\u0441\u0442\u0438\u043a\u0430   \u0442\u0440\u0430\u043d\u0441\u043f\u043e\u043d\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435\n",

      "print data.describe().T"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "Index([u'A', u'B', u'C', u'D', u'E', u'F'], dtype='object')\n",

        "<class 'pandas.tseries.period.PeriodIndex'>\n",

        "[2000-01, ..., 2000-04]\n",

        "Length: 4, Freq: M\n",

        "         A          B   C  D      E    F\n",

        "2000-01  1 2013-01-02 NaN  3   test  foo\n",

        "2000-02  4 2013-01-02 NaN  3  train  foo\n",

        "2000-03  2 2013-01-02 NaN  3   test  foo\n",

        "2000-04  1 2013-01-02 NaN  3  train  foo\n",

        "         A          B   C  D      E    F\n",

        "2000-01  1 2013-01-02 NaN  3   test  foo\n",

        "2000-04  1 2013-01-02 NaN  3  train  foo\n",

        "2000-03  2 2013-01-02 NaN  3   test  foo\n",

        "2000-02  4 2013-01-02 NaN  3  train  foo\n",

        "A           float64\n",

        "B    datetime64[ns]\n",

        "C           float32\n",

        "D             int32\n",

        "E            object\n",

        "F            object\n",

        "dtype: object\n",

        "   count  mean       std  min  25%  50%  75%  max\n",

        "A      4     2  1.414214    1    1  1.5  2.5    4\n",

        "C      0   NaN       NaN  NaN  NaN  NaN  NaN  NaN\n",

        "D      4     3  0.000000    3    3  3.0  3.0    3\n"

       ]

      }

     ],

     "prompt_number": 60

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0438\u043d\u0434\u0435\u043a\u0441\u0430\u0446\u0438\u044f\n",

      "data.at['2000-01','A'] = 10.\n",

      "print data.loc['2000-01':'2000-02',['D','B','A']]\n",

      "data.iat[0,1] = pd.Timestamp('19990101') # \u043f\u0440\u043e\u0441\u0442\u043e = '1999/01/01' \u043d\u0435 \u0440\u0430\u0431\u043e\u0442\u0430\u0435\u0442\n",

      "print data.iloc[0:2,1:3]\n",

      "# \u0432\u044b\u0431\u043e\u0440 \u0441 \u043f\u0440\u043e\u0432\u0435\u0440\u043a\u043e\u0439 \u043d\u0430 \u0432\u0445\u043e\u0436\u0434\u0435\u043d\u0438\u0435\n",

      "print data[data['E'].isin(['test','valid'])]\n",

      "# \u0435\u0441\u043b\u0438 \u043d\u0435 \u0434\u0435\u043b\u0430\u0442\u044c \u043f\u0440\u0438\u043d\u0442 - \u0432\u044b\u0433\u043b\u044f\u0434\u0438\u0442 \u043a\u0440\u0443\u0447\u0435\n",

      "data.tail(3).T"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "         D          B   A\n",

        "2000-01  3 2013-01-02  10\n",

        "2000-02  3 2013-01-02   4\n",

        "                 B   C\n",

        "2000-01 1999-01-01 NaN\n",

        "2000-02 2013-01-02 NaN\n",

        "          A          B   C  D     E    F\n",

        "2000-01  10 1999-01-01 NaN  3  test  foo\n",

        "2000-03   2 2013-01-02 NaN  3  test  foo\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>2000-02</th>\n",

        "      <th>2000-03</th>\n",

        "      <th>2000-04</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>A</th>\n",

        "      <td>                   4</td>\n",

        "      <td>                   2</td>\n",

        "      <td>                   1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>B</th>\n",

        "      <td> 2013-01-02 00:00:00</td>\n",

        "      <td> 2013-01-02 00:00:00</td>\n",

        "      <td> 2013-01-02 00:00:00</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>C</th>\n",

        "      <td>                 NaN</td>\n",

        "      <td>                 NaN</td>\n",

        "      <td>                 NaN</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>D</th>\n",

        "      <td>                   3</td>\n",

        "      <td>                   3</td>\n",

        "      <td>                   3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>E</th>\n",

        "      <td>               train</td>\n",

        "      <td>                test</td>\n",

        "      <td>               train</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>F</th>\n",

        "      <td>                 foo</td>\n",

        "      <td>                 foo</td>\n",

        "      <td>                 foo</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 61,

       "text": [

        "               2000-02              2000-03              2000-04\n",

        "A                    4                    2                    1\n",

        "B  2013-01-02 00:00:00  2013-01-02 00:00:00  2013-01-02 00:00:00\n",

        "C                  NaN                  NaN                  NaN\n",

        "D                    3                    3                    3\n",

        "E                train                 test                train\n",

        "F                  foo                  foo                  foo"

       ]

      }

     ],

     "prompt_number": 61

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043e\u043f\u0435\u0440\u0430\u0446\u0438\u0438 \u0441 \u041d\u0430\u041d\u0430\u043c\u0438\n",

      "# \u043c\u0430\u0441\u043a\u0430 \u041d\u0430\u043d\u043e\u0432\n",

      "print data2.isnull()\n",

      "# nan \u0430\u0432\u0442\u043e\u043c\u0430\u0442\u0438\u0447\u0435\u0441\u043a\u0438 \u043d\u0435 \u0443\u0447\u0438\u0442\u044b\u0432\u0430\u044e\u0442\u0441\u044f\n",

      "print data2.mean()\n",

      "# \u0442\u043e\u0436\u0435 \u043e\u0431\u0445\u043e\u0434\u044f\u0442\u0441\u044f nan:\n",

      "print data2.apply(np.cumsum)\n",

      "# \u0443\u0434\u0430\u043b\u0435\u043d\u0438\u0435 \u041d\u0430\u043d\u043e\u0432\n",

      "data3 = data2.dropna()\n",

      "print data3\n",

      "# \u0437\u0430\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u0435 \u041d\u0430\u043d\u043e\u0432\n",

      "print data2.fillna(value=5.5)\n",

      "# \u0437\u0430\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u0435 \u0441\u043e\u0441\u0435\u0434\u043d\u0438\u043c\u0438 \u0437\u043d\u0430\u0447\u0435\u043d\u0438\u044f\u043c\u0438\n",

      "print data2.ffill()\n",

      "# \u043d\u0435 \u0437\u0430\u0431\u044b\u0432\u0430\u0442\u044c data2 = data2.fillna(value=5.5)\n",

      "data2"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "       A      B\n",

        "0  False  False\n",

        "1   True   True\n",

        "2  False   True\n",

        "3  False  False\n",

        "A    1.333333\n",

        "B    1.100000\n",

        "dtype: float64\n",

        "    A    B\n",

        "0   1  2.2\n",

        "1 NaN  NaN\n",

        "2   3  NaN\n",

        "3   4  2.2\n",

        "   A    B\n",

        "0  1  2.2\n",

        "3  1  0.0\n",

        "     A    B\n",

        "0  1.0  2.2\n",

        "1  5.5  5.5\n",

        "2  2.0  5.5\n",

        "3  1.0  0.0\n",

        "   A    B\n",

        "0  1  2.2\n",

        "1  1  2.2\n",

        "2  2  2.2\n",

        "3  1  0.0\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td>  1</td>\n",

        "      <td> 2.2</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td>NaN</td>\n",

        "      <td> NaN</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td>  2</td>\n",

        "      <td> NaN</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>3</th>\n",

        "      <td>  1</td>\n",

        "      <td> 0.0</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 63,

       "text": [

        "    A    B\n",

        "0   1  2.2\n",

        "1 NaN  NaN\n",

        "2   2  NaN\n",

        "3   1  0.0"

       ]

      }

     ],

     "prompt_number": 63

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043a\u0430\u043a \u0447\u0430\u0441\u0442\u043e \u0432\u0441\u0442\u0440\u0435\u0447\u0430\u044e\u0442\u0441\u044f \u043f\u0430\u0440\u044b \u0437\u043d\u0430\u0447\u0435\u043d\u0438\u0439\n",

      "d = pd.DataFrame({'A': [1,2,2,1,2,3,2,1,3], 'B': [1,2,3,4,1,2,3,3,4]})\n",

      "pd.crosstab(d['A'], d['B'])"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th>B</th>\n",

        "      <th>1</th>\n",

        "      <th>2</th>\n",

        "      <th>3</th>\n",

        "      <th>4</th>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>A</th>\n",

        "      <th></th>\n",

        "      <th></th>\n",

        "      <th></th>\n",

        "      <th></th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 1</td>\n",

        "      <td> 0</td>\n",

        "      <td> 1</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 1</td>\n",

        "      <td> 1</td>\n",

        "      <td> 2</td>\n",

        "      <td> 0</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>3</th>\n",

        "      <td> 0</td>\n",

        "      <td> 1</td>\n",

        "      <td> 0</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 64,

       "text": [

        "B  1  2  3  4\n",

        "A            \n",

        "1  1  0  1  1\n",

        "2  1  1  2  0\n",

        "3  0  1  0  1"

       ]

      }

     ],

     "prompt_number": 64

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0432\u0441\u0435 \u0441\u0442\u0440\u043e\u043a\u0438, \u0432 \u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u0441\u0442\u043e\u043b\u0431\u0435\u0446 \u043d\u0430\u0447\u0438\u043d\u0430\u0435\u0442\u0441\u044f \u0441 \u043e\u043f\u0440\u0435\u0434\u0435\u043b\u0451\u043d\u043d\u043e\u0439 \u0431\u0443\u043a\u0432\u044b\n",

      "d = pd.DataFrame({'A': [1,2,2,1,2,3,2,1,3], 'B': ['as','bs','e','qq','aaa','a','e','qwr','www']})\n",

      "d[d['B'].map(lambda x: x.startswith('a'))]"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 1</td>\n",

        "      <td>  as</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>4</th>\n",

        "      <td> 2</td>\n",

        "      <td> aaa</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>5</th>\n",

        "      <td> 3</td>\n",

        "      <td>   a</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 65,

       "text": [

        "   A    B\n",

        "0  1   as\n",

        "4  2  aaa\n",

        "5  3    a"

       ]

      }

     ],

     "prompt_number": 65

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0437\u0430\u043c\u0435\u043d\u0430 \u043e\u043f\u0440\u0435\u0434\u0435\u043b\u0451\u043d\u043d\u044b\u0445 \u0437\u043d\u0430\u0447\u0435\u043d\u0438\u0439\n",

      "d = pd.DataFrame({'A': [1,2,2,1,2,3,2,1,3], 'B': [1,2,3,4,1,2,3,3,4]})\n",

      "d['B'][d['A']==1] = 0\n",

      "# \u0432\u0442\u043e\u0440\u043e\u0439 \u0441\u043f\u043e\u0441\u043e\u0431: d.ix[d['A']==1, 'B'] = 0\n",

      "d"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 1</td>\n",

        "      <td> 0</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 2</td>\n",

        "      <td> 2</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 2</td>\n",

        "      <td> 3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>3</th>\n",

        "      <td> 1</td>\n",

        "      <td> 0</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>4</th>\n",

        "      <td> 2</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>5</th>\n",

        "      <td> 3</td>\n",

        "      <td> 2</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>6</th>\n",

        "      <td> 2</td>\n",

        "      <td> 3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>7</th>\n",

        "      <td> 1</td>\n",

        "      <td> 0</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>8</th>\n",

        "      <td> 3</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 66,

       "text": [

        "   A  B\n",

        "0  1  0\n",

        "1  2  2\n",

        "2  2  3\n",

        "3  1  0\n",

        "4  2  1\n",

        "5  3  2\n",

        "6  2  3\n",

        "7  1  0\n",

        "8  3  4"

       ]

      }

     ],

     "prompt_number": 66

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043d\u0438\u0435 \u043c\u0430\u0441\u043e\u043a\n",

      "d = pd.DataFrame({'A': [1,2,2,1,2,3,2,1,3], 'B': [1,2,3,4,1,2,3,3,4]})\n",

      "mask = d>1\n",

      "print mask.T\n",

      "d = d.where(mask,2)\n",

      "d['C'] = np.where(d['B']>3, 'high', 'low')\n",

      "d.T"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "       0     1     2      3      4     5     6      7     8\n",

        "A  False  True  True  False   True  True  True  False  True\n",

        "B  False  True  True   True  False  True  True   True  True\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>0</th>\n",

        "      <th>1</th>\n",

        "      <th>2</th>\n",

        "      <th>3</th>\n",

        "      <th>4</th>\n",

        "      <th>5</th>\n",

        "      <th>6</th>\n",

        "      <th>7</th>\n",

        "      <th>8</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>A</th>\n",

        "      <td>   2</td>\n",

        "      <td>   2</td>\n",

        "      <td>   2</td>\n",

        "      <td>    2</td>\n",

        "      <td>   2</td>\n",

        "      <td>   3</td>\n",

        "      <td>   2</td>\n",

        "      <td>   2</td>\n",

        "      <td>    3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>B</th>\n",

        "      <td>   2</td>\n",

        "      <td>   2</td>\n",

        "      <td>   3</td>\n",

        "      <td>    4</td>\n",

        "      <td>   2</td>\n",

        "      <td>   2</td>\n",

        "      <td>   3</td>\n",

        "      <td>   3</td>\n",

        "      <td>    4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>C</th>\n",

        "      <td> low</td>\n",

        "      <td> low</td>\n",

        "      <td> low</td>\n",

        "      <td> high</td>\n",

        "      <td> low</td>\n",

        "      <td> low</td>\n",

        "      <td> low</td>\n",

        "      <td> low</td>\n",

        "      <td> high</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 67,

       "text": [

        "     0    1    2     3    4    5    6    7     8\n",

        "A    2    2    2     2    2    3    2    2     3\n",

        "B    2    2    3     4    2    2    3    3     4\n",

        "C  low  low  low  high  low  low  low  low  high"

       ]

      }

     ],

     "prompt_number": 67

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043e\u0431\u044a\u0435\u0434\u0438\u043d\u0435\u043d\u0438\u0435 \u0434\u0430\u0442\u0430-\u0444\u0440\u0435\u0439\u043c\u043e\u0432\n",

      "left = pd.DataFrame({'key': [1,2,1], 'l': [1, 2, 3]})\n",

      "right = pd.DataFrame({'key': [1,2,3], 'r': [4, 5, 6]})\n",

      "print left\n",

      "print right\n",

      "pd.merge(left, right, on='key')"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "   key  l\n",

        "0    1  1\n",

        "1    2  2\n",

        "2    1  3\n",

        "   key  r\n",

        "0    1  4\n",

        "1    2  5\n",

        "2    3  6\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>key</th>\n",

        "      <th>l</th>\n",

        "      <th>r</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 1</td>\n",

        "      <td> 1</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 1</td>\n",

        "      <td> 3</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 2</td>\n",

        "      <td> 2</td>\n",

        "      <td> 5</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 71,

       "text": [

        "   key  l  r\n",

        "0    1  1  4\n",

        "1    1  3  4\n",

        "2    2  2  5"

       ]

      }

     ],

     "prompt_number": 71

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0434\u043e\u0431\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u043a \u0434\u0430\u0442\u0430-\u0444\u0440\u0435\u0439\u043c\u0443\n",

      "tmp = dict([('A',[1., 3., 2., 1.]), ('B',[2.2, 1.1, 3.3, 0.0]), ('C', 1)]) # \u0435\u0449\u0451 \u043e\u0434\u0438\u043d \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "df = pd.DataFrame(tmp)\n",

      "# \u0434\u043e\u0431\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u043a \u0434\u0430\u0442\u0430-\u0444\u0440\u0435\u0439\u043c\u0443\n",

      "df = df.append(df.iloc[1:3])\n",

      "df"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "      <th>C</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 1</td>\n",

        "      <td> 2.2</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 3</td>\n",

        "      <td> 1.1</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 2</td>\n",

        "      <td> 3.3</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>3</th>\n",

        "      <td> 1</td>\n",

        "      <td> 0.0</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 3</td>\n",

        "      <td> 1.1</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 2</td>\n",

        "      <td> 3.3</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 85,

       "text": [

        "   A    B  C\n",

        "0  1  2.2  1\n",

        "1  3  1.1  1\n",

        "2  2  3.3  1\n",

        "3  1  0.0  1\n",

        "1  3  1.1  1\n",

        "2  2  3.3  1"

       ]

      }

     ],

     "prompt_number": 85

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0438\u043d\u0444\u043e\u0440\u043c\u0430\u0446\u0438\u044f\n",

      "data.info() # \u043f\u043e \u043f\u0430\u043c\u044f\u0442\u0438\n",

      "#??? data.memory_usage() # \u0430\u043d\u0430\u043b\u043e\u0433\u0438\u0447\u043d\u043e"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "<class 'pandas.core.frame.DataFrame'>\n",

        "PeriodIndex: 4 entries, 2000-01 to 2000-04\n",

        "Data columns (total 6 columns):\n",

        "A    4 non-null float64\n",

        "B    4 non-null datetime64[ns]\n",

        "C    0 non-null float32\n",

        "D    4 non-null int32\n",

        "E    4 non-null object\n",

        "F    4 non-null object\n",

        "dtypes: datetime64[ns](1), float32(1), float64(1), int32(1), object(2)"

       ]

      }

     ],

     "prompt_number": 69

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0432\u0440\u0435\u043c\u0435\u043d\u043d\u044b\u0435 \u0440\u044f\u0434\u044b\n",

      "# \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0435\n",

      "rng = pd.date_range('1/1/2012', periods=9, freq='D')\n",

      "# \u0434\u043d\u0438\n",

      "print rng.day\n",

      "# \u0434\u043d\u0438 \u043d\u0435\u0434\u0435\u043b\u0438\n",

      "print rng.weekday\n",

      "\n",

      "rng = pd.date_range('1/1/2012', periods=200, freq='S')\n",

      "ts = pd.Series(np.random.randint(0, 500, len(rng)), index=rng)\n",

      "# \u0440\u0435\u0441\u044d\u043c\u043f\u043b\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435\n",

      "ts = ts.resample('Min', how='sum')\n",

      "print ts\n",

      "# ???\n",

      "ts.tz_localize('UTC')\n",

      "print ts"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "[1 2 3 4 5 6 7 8 9]\n",

        "[6 0 1 2 3 4 5 6 0]\n",

        "2012-01-01 00:00:00    15255\n",

        "2012-01-01 00:01:00    13164\n",

        "2012-01-01 00:02:00    14217\n",

        "2012-01-01 00:03:00     5326\n",

        "Freq: T, dtype: int32\n",

        "2012-01-01 00:00:00    15255\n",

        "2012-01-01 00:01:00    13164\n",

        "2012-01-01 00:02:00    14217\n",

        "2012-01-01 00:03:00     5326\n",

        "Freq: T, dtype: int32\n"

       ]

      }

     ],

     "prompt_number": 72

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0440\u0435\u0448\u0435\u043d\u0438\u0435 \u0437\u0430\u0434\u0430\u0447\u0438\n",

      "# \u0434\u043b\u044f \u043a\u0430\u0436\u0434\u043e\u0433\u043e \u0443\u043d\u0430\u043a\u0430\u043b\u044c\u043d\u043e\u0433\u043e \u0437\u043d\u0430\u0447\u0435\u043d\u0438\u044f A \u043d\u0430\u0439\u0442\u0438 \u043c\u0438\u043d\u0438\u043c\u0430\u043b\u044c\u043d\u044b\u0439 B\n",

      "import pandas as pd\n",

      "d = pd.DataFrame({'A': [1,2,2,1,3,3], 'B': [1,2,3,3,2,1]})\n",

      "print d\n",

      "\n",

      "# \u043f\u0435\u0440\u0432\u044b\u0439 \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "print d.loc[d.groupby('A')['B'].idxmin()]\n",

      "\n",

      "# \u0432\u0442\u043e\u0440\u043e\u0439 \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "print d.sort('B').groupby('A', as_index=False).first()"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "   A  B\n",

        "0  1  1\n",

        "1  2  2\n",

        "2  2  3\n",

        "3  1  3\n",

        "4  3  2\n",

        "5  3  1\n",

        "   A  B\n",

        "0  1  1\n",

        "1  2  2\n",

        "5  3  1\n",

        "   A  B\n",

        "0  1  1\n",

        "1  2  2\n",

        "2  3  1\n"

       ]

      }

     ],

     "prompt_number": 73

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043e\u043f\u0435\u0440\u0430\u0446\u0438\u0438 \u043f\u043e \u0433\u0440\u0443\u043f\u043f\u0430\u043c (\u0438\u043d\u0434\u0443\u0446\u0438\u0440\u0443\u044e\u0442\u0441\u044f \u0440\u0430\u0437\u0431\u0438\u0435\u043d\u0438\u0435\u043c \u043f\u043e \u043e\u043f\u0440\u0435\u0434\u0435\u043b\u0451\u043d\u043d\u043e\u043c\u0443 \u043f\u0440\u0438\u0437\u043d\u0430\u043a\u0443)\n",

      "d = pd.DataFrame({'A': [1,2,2,1,1,2,2], 'B': [1,2,np.nan,5,3,1,10]})\n",

      "d = d.sort('A')\n",

      "print d\n",

      "d['shift_B'] = d.groupby('A')['B'].shift(1) # \u0441\u0434\u0432\u0438\u0433 \u0433\u0440\u0443\u043f\u043f\n",

      "d['counts'] = d.groupby(['A'])['B'].transform(len) # \u0447\u0438\u0441\u043b\u043e \u044d\u043b\u0435\u043c\u0435\u043d\u0442\u043e\u0432 \u0432 \u0433\u0440\u0443\u043f\u043f\u0435\n",

      "d"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "   A   B\n",

        "0  1   1\n",

        "3  1   5\n",

        "4  1   3\n",

        "1  2   2\n",

        "2  2 NaN\n",

        "5  2   1\n",

        "6  2  10\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "      <th>shift_B</th>\n",

        "      <th>counts</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 1</td>\n",

        "      <td>  1</td>\n",

        "      <td>NaN</td>\n",

        "      <td> 3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>3</th>\n",

        "      <td> 1</td>\n",

        "      <td>  5</td>\n",

        "      <td>  1</td>\n",

        "      <td> 3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>4</th>\n",

        "      <td> 1</td>\n",

        "      <td>  3</td>\n",

        "      <td>  5</td>\n",

        "      <td> 3</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 2</td>\n",

        "      <td>  2</td>\n",

        "      <td>NaN</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 2</td>\n",

        "      <td>NaN</td>\n",

        "      <td>  2</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>5</th>\n",

        "      <td> 2</td>\n",

        "      <td>  1</td>\n",

        "      <td>NaN</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>6</th>\n",

        "      <td> 2</td>\n",

        "      <td> 10</td>\n",

        "      <td>  1</td>\n",

        "      <td> 4</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 74,

       "text": [

        "   A   B  shift_B  counts\n",

        "0  1   1      NaN       3\n",

        "3  1   5        1       3\n",

        "4  1   3        5       3\n",

        "1  2   2      NaN       4\n",

        "2  2 NaN        2       4\n",

        "5  2   1      NaN       4\n",

        "6  2  10        1       4"

       ]

      }

     ],

     "prompt_number": 74

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "d = pd.DataFrame({'A': [1,2,2,1,1,2,2], 'B': [1,0,np.nan,1,0,1,0]})\n",

      "d = d.sort('A')\n",

      "print d\n",

      "\n",

      "# \u0434\u043b\u0438\u043d\u0430 \u0438 \u0441\u043a\u043e\u043b\u044c\u043a\u043e 1 \u0432 \u043a\u0430\u0436\u0434\u043e\u0439 \u0433\u0440\u0443\u043f\u043f\u0435\n",

      "d.groupby('A').agg({'A': len, 'B': lambda x: sum(x == 1)})"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "   A   B\n",

        "0  1   1\n",

        "3  1   1\n",

        "4  1   0\n",

        "1  2   0\n",

        "2  2 NaN\n",

        "5  2   1\n",

        "6  2   0\n"

       ]

      },

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>A</th>\n",

        "      <th>B</th>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>A</th>\n",

        "      <th></th>\n",

        "      <th></th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>1</th>\n",

        "      <td> 3</td>\n",

        "      <td> 2</td>\n",

        "    </tr>\n",

        "    <tr>\n",

        "      <th>2</th>\n",

        "      <td> 4</td>\n",

        "      <td> 1</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 75,

       "text": [

        "   A  B\n",

        "A      \n",

        "1  3  2\n",

        "2  4  1"

       ]

      }

     ],

     "prompt_number": 75

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0421\u0442\u0440\u043e\u043a\u043e\u0432\u044b\u0435 \u043e\u043f\u0435\u0440\u0430\u0446\u0438\u0438\n",

      "s = pd.Series(['AbA', 'Sasha', 'DataMining']) # ( [] )\n",

      "s.str.lower()"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 76,

       "text": [

        "0           aba\n",

        "1         sasha\n",

        "2    datamining\n",

        "dtype: object"

       ]

      }

     ],

     "prompt_number": 76

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0438\u0435\u0440\u0430\u0440\u0445\u0438\u0447\u0435\u0441\u043a\u0430\u044f (\u043c\u043d\u043e\u0433\u043e\u0443\u0440\u043e\u0432\u043d\u0435\u0432\u0430\u044f) \u0438\u043d\u0434\u0435\u043a\u0441\u0430\u0446\u0438\u044f\n",

      "tuples = list(zip(*[['bar', 'bar', 'baz', 'baz',\n",

      "'foo', 'foo', 'qux', 'qux'],\n",

      " ['one', 'two', 'one', 'two',\n",

      "'one', 'two', 'one', 'two']]))\n",

      "print tuples\n",

      "\n",

      "index = pd.MultiIndex.from_tuples(tuples, names=['first', 'second'])\n",

      "print index\n",

      "index\n",

      "\n",

      "df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=['A', 'B'])\n",

      "print df\n",

      "\n",

      "print df.stack() # \u043e\u0431\u0440\u0430\u0442\u043d\u0430\u044f \u043e\u043f\u0435\u0440\u0430\u0446\u0438\u044f unstack()"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "[('bar', 'one'), ('bar', 'two'), ('baz', 'one'), ('baz', 'two'), ('foo', 'one'), ('foo', 'two'), ('qux', 'one'), ('qux', 'two')]\n",

        "first  second\n",

        "bar    one   \n",

        "       two   \n",

        "baz    one   \n",

        "       two   \n",

        "foo    one   \n",

        "       two   \n",

        "qux    one   \n",

        "       two   \n",

        "                     A         B\n",

        "first second                    \n",

        "bar   one    -0.240469 -0.533312\n",

        "      two    -0.847305  0.845316\n",

        "baz   one     0.274592  0.473476\n",

        "      two     1.433575 -0.977992\n",

        "foo   one     0.957252 -1.246396\n",

        "      two    -2.821039 -0.625924\n",

        "qux   one     0.086683 -0.450850\n",

        "      two    -1.236494  0.706156\n",

        "first  second   \n",

        "bar    one     A   -0.240469\n",

        "               B   -0.533312\n",

        "       two     A   -0.847305\n",

        "               B    0.845316\n",

        "baz    one     A    0.274592\n",

        "               B    0.473476\n",

        "       two     A    1.433575\n",

        "               B   -0.977992\n",

        "foo    one     A    0.957252\n",

        "               B   -1.246396\n",

        "       two     A   -2.821039\n",

        "               B   -0.625924\n",

        "qux    one     A    0.086683\n",

        "               B   -0.450850\n",

        "       two     A   -1.236494\n",

        "               B    0.706156\n",

        "dtype: float64\n"

       ]

      }

     ],

     "prompt_number": 77

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u043f\u0430\u043d\u0435\u043b\u044c (3D-\u0434\u0430\u0442\u0430-\u0444\u0440\u0435\u0439\u043c)\n",

      "d = pd.DataFrame({'A': [1,2,2], 'B': [1,2,3]})\n",

      "d2, d3 = d.copy(), d.copy()\n",

      "p = pd.Panel({'df1':d, 'df2':d2, 'df3':d3})\n",

      "p = p.transpose(2,0,1)\n",

      "print p['A']\n",

      "p"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "output_type": "stream",

       "stream": "stdout",

       "text": [

        "     0  1  2\n",

        "df1  1  2  2\n",

        "df2  1  2  2\n",

        "df3  1  2  2\n"

       ]

      },

      {

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 41,

       "text": [

        "<class 'pandas.core.panel.Panel'>\n",

        "Dimensions: 2 (items) x 3 (major_axis) x 3 (minor_axis)\n",

        "Items axis: A to B\n",

        "Major_axis axis: df1 to df3\n",

        "Minor_axis axis: 0 to 2"

       ]

      }

     ],

     "prompt_number": 41

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0433\u0440\u0430\u0444\u0438\u043a\u0430\n",

      "d = pd.DataFrame({\"IQ\": [1,4,3,2]})\n",

      "d.index = ['man', 'woman', 'dog', 'cat']\n",

      "d.plot(kind='barh')"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 78,

       "text": [

        "<matplotlib.axes.AxesSubplot at 0xb28fcf8>"

       ]

      },

      {

       "metadata": {},

       "output_type": "display_data",

       "png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAADECAYAAADapYZ4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1 /AAAFCFJREFUeJzt3d9vk X/x/FXN2DY/ZTiGENnB9iIMOt3GDQ6xI /QmZI\npqKJB8YTT4xRj0w88A/4JHigMdH/QGOiKBpxAiMiRcgSF5f9EPzq1EVgzHXSDRjDdv0e F0Fe8HW\ncXX3dfd Pk5cu7v13VeuuLf3/e59hbLZbFYAAABwQpnXBQAAAOAfNGcAAAAOoTkDAABwCM0ZAACA\nQ2jOAAAAHLLE6wLm69ixYzp//rzXZQAAAMyprq5OmzdvXtBrfdOcnT9/Xq2trV6XAQAAMKeenp4F\nv5bLmgAAAA6hOfO5RCLhdQlOIhczcslHJmbkYkYuZuRiF80ZAACAQ0J 2b6pq6uLmTMAAOALPT09\nevjhhxf0Wt98IQAAAPhXMpnU9PS012VYVVFRoUgkYv19ac58LpFIqK2tzesynEMuZuSSj0zMyMWM\nXMzmyuXcuXOSpMbGxsUqaVEkk0mdO3dOVVVVVt XmTMAAFBUqVRKK1as8LoM61asWKFUKmX9fZk5\nAwAARXXq1KmSO2s262qf7XpmzjhzBgAA4BBfzZz1npr0ugTnpFIp1dbWel2Gc8jFjFzyzZwb1//E\nbvW6DOcwW2VGLmaF5nJ6Ylqj5y4VrZ76qmVaXVNRtPcvNl81Z6/t/cnrEhz1h9cFOIpczMjlcm/c\nf5PXJQCBM3ruUlH/pu9qX /r5ozLmgACjTOJZpwdMiMXM3Kxy1dnzgAAAGw5fPiwenp69PXXX vB\nBx9UfX29nn32Wf3111967733NDMzo2w2q6amJj311FOLVhfNGYBAS6VSUmO112U4h9kqM3Ix82su\nW7du1datW3Xx4kW9 uqrueffeecd7dy5U01NTZKkvXv36uDBg3rooYcWpS4uawIAAFxmamoq15hJ\nUnt7u7q7uxft309zBiDQmDkz8 NZkMVALmallMvY2JjxvmXLli1btBo8b85OnDih0dFRr8sAAABQ\nJBLRmTNn8p7/66 /Fq0Gz5uzTCajdDrtdRkAAqoYW6 UgkQi4XUJTiIXs1LKJRQKqaKiQidPnsw9\nt3//fm3evHnRaijKFwJ /PFHDQ4OqqysTGNjY3rsscf0888/K5lMKpPJKB6PKxaL6ezZs0okElq \nfLlqamq0du1a3XnnncUoCQAAOKK apl2ta8v6vvPx y3NY8dO6a33347923NV155RW //bampqZU\nWVmpNWvW6Omnny5avf9WlObsp59 UkdHhyRpeHhY4XBY27Zty/2 s7NTsVhMdXV1amtrU11dXcnu\nuQXAbbW1tbn/65 dm HxlWdBXKnHhcdtbW1O1ePS41mm30cikSv zq uqXDiJrGz39a8/JuaklRe\nXq6qqir9/PPPisViGhkZueq3NVOplIaGhvLyCIfDC67L sbnExMTOn78uLZs2XLF84lEQmNjYyov\nL9fU1JSeeeYZSdLg4OC8mrOuri693hOyWSoAaFf7esW5lQZQVGx8XhjrM2c1NTX6449/toe5ePGi\nfv/9d4XDYXV0dGjHjh2qqqrK/b68vFyZTMZ2GQAwL8ycmZXSDJFN5GJGLnYV5bJmc3OzPv74Y4VC\nf5/p2r59uxKJhIaHh5VOpzU8PKxNmzapqalJzc3N uSTT3TDDTfo0qVL s9//qNIJFKMsgAAAJxn\n/bJmsXBZE0AxcFkTKL4zZ86ourr6uuawXHThwgVNTk5q1apVeb 7nsuabN8EAACKqr6 XqOjozp7\n9qzXpVhVXl6u vp66 9LcwYg0Nhb08yveyUWG7mYzZVLKBQynl2Cmec3oQUAAMA/aM4ABBp7a5px\ndsiMXMzIxS6aMwAAAIfQnAEINO5zZsZ9q8zIxYxc7PLVFwKKuQ XX6VSKS7LGJCLGbnkmzk37nUJ\nAHAFX93nrLW11esyAAAA5uTU9k0AAABYOJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAAABzCzBkA\nAIBlzJwBAACUCJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAAABzCzBkAAIBlzJwBAACUCJozn M6\nvxm5mJFLPjIxIxczcjEjF7tozgAAABzCzBkAAIBlzJwBAACUCJozn M6vxm5mJFLPjIxIxczcjEj\nF7tozgAAABzCzBkAAIBlzJwBAACUCJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAAABziq5mz8obb\nvC4DABBg9VXLtLqmwusy4APXM3O2xHItRfXa3p 8LgEAEGC72tfTnKHouKwJAMA8pVIpr0twEjNn\ndtGcAQAAOITmDACAeaqtrfW6BCe1tbV5XUJJoTkDAABwSNGas8nJSb355pvFensAABYdM2dmzJzZ\nVbTmrLq6WvF4vFhvDwAAUJKs30rjwIEDSqVSSqfTmp6eliR9//33GhoaUigU0qpVq3TfffdJkg4e\nPKiJiQlduHBBExMT2rp1qzZu3Gi7JAAArGDmzIyZM7usNmenTp1SXV2dHnnkEUnSm2  qQsXLiiZ\nTOrJJ5 UJHV3d2tkZETpdForVqzQQw89pHQ6rXfffZfGDAAABJ7V5uzXX3/VnXfemXt8  236/Tp\n02pubs49t2HDBg0MDCibzeaOXbJkidauXWuzFAAArEulUkoM9Ur652zR7LxVkB/39fXpxRdfdKYe\nFx6Hw2EtlNXtm06ePKnTp0/r7rvvliTt2rVLL730ko4ePZrbwqC7u1tNTU1Kp9NKJpOKx NKp9N6\n77339PLLL1/1vbu6uvR6T8hWqQAAFOyN 2/SAxtu9roM5yQSCS5t/osz2zetWbNGg4OD uijjzQz\nM6Mbb7xRPT09ikQi2r17t0KhkOrr69XQ0CBJOn78uD799FNdvHhRt9xyi81SAACwjpkzMxozu6x/\nIeDRRx81Pn/XXXflPTc7myZJ /bts10KAACA73i68XkymdThw4d16dIlY/MGAIBLUqmU1FjtdRnO\n4bKmXZ42Z5FIRB0dHV6WAAAA4BS2bwIAYJ6YOTPjrJldNGcAAAAOoTkDAGCe2FvTjL017aI5AwAA\ncAjNGQAA88TMmRkzZ3Z5 m3NQu1qX 91CQCAAKuvWuZ1CQgAXzVnce4tk4d7y5iRixm55CMTM3Ix\nSyQSWk0ueVgvdnFZEwAAwCFWNz4vpq6uLrW2tnpdBgAAwJyuZ NzzpwBAAA4hObM57i3jBm5mJFL\nPjIxIxczcjEjF7tozgAAABzCzBkAAIBlzJwBAACUCJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAA\nABzCzBkAAIBlzJwBAACUCJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAAABzCzBkAAIBlzJwBAACU\nCJozn M6vxm5mJFLPjIxIxczcjEjF7tozgAAABzCzBkAAIBlzJwBAACUiCVeF1CI3lOTXpfgnFQq\npdraWq/LcA65mJFLPjIxIxczcjEjF7t81Zy9tvcnr0tw1B9eF AocjEjl3xkYkYuZuRiRi6X  91\nTGJxWRMAAMAhNGcAAAAOoTkDAABwCM0ZAACAQ2jOAAAAHEJzBgAA4JBr3kpj3759SiaTevbZZyVJ\nH3zwgSKRiOrr6zU0NKRQKKRVq1bpvvvu04cffqjR0VFVVVWpoqJC09PT2rFjh1auXKlDhw4pmUwq\nk8koHo8rFotpZGRE77//vhoaGlRZWaloNKp4PL4oHxoAAMBV12zO2tra9P3336uvr0 SdOuttyoW\ni6m3t1dPPvmkJKm7u1sjIyNqaWlRZWWlysrKND4 rnXr1qmvr08rV67Utm3bcu/Z2dmpWCymhoYG\nNTU1aefOnZKk/fv3F szAgAA MY1m7NwOKzz589rYmJC2WxWS5YsUSqVUnNzc 6YDRs2aGBgQDU1\nNVq6dKkkaenSpSorK9Pstp2JREJjY2MqLy/X1NRU7rXcTRgAAOBK8545C4VCkqTVq1frl19 yT3/\nww8/KBqNXvV1v// u8LhsDo6OrRjxw5VVVUtvFoAAIASN f2TefPn1csFlNZWZkGBwcVDocViUS0\ne/duhUIh1dfXq6GhQd98841qa2vV0tKi/v5 RaNR9ff3q7W1VYlEQsPDw0qn0xoeHtamTZuUyWTU\n29urlpYWhcNh9fb2auPGjWpsbFyMzw0AAOCkUHb22qPjurq69HpPyOsyAAAA5vTf1qwefvjhBb2W\nW2kAAAA4hOYMAADAITRnAAAADqE5AwAAcAjNGQAAgENozgAAABwy533OXLKrfb3XJTgnlUqx04IB\nuZiRSz4yMSMXM3IxI5d8mZH/XfBrfdWcxRurvS7BOYmhXsU3tHldhnPIxYxc8pGJGbmYkYsZueTr\nGVn4a311E9rW1lavywAAAJhTT08PN6EFAAAoBTRnPpdIJLwuwUnkYkYu cjEjFzMyMWMXOyiOQMA\nAHAIM2cAAACWMXMGAABQImjOfI7r/GbkYkYu cjEjFzMyMWMXOyiOQMAAHAIM2cAAACWMXMGAABQ\nImjOfI7r/GbkYkYu cjEjFzMyMWMXOyiOQMAAHAIM2cAAACWMXMGAABQImjOfI7r/GbkYkYu cjE\njFzMyMWMXOyiOQMAAHAIM2cAAACWMXMGAABQImjOfI7r/GbkYkYu cjEjFzMyMWMXOxa4nUBheg9\nNel1Cc6ZiUTJxaCy/havSwAAYEF8NXP2ek/I6zLgE7va1yveWO11GQCAgGLmDAAAoETQnKEkpVIp\nr0twEnMh cjEjFzMyMWMXOyiOQMAAHAIzRlKUm1trdclOKmtrc3rEpxDJmbkYkYuZuRiF80ZAACA\nQ2jOUJKYOTNjLiQfmZiRixm5mJGLXTRnAAAADrF6E9oPP/xQo6OjqqqqUkVFhaanp7Vjxw4NDAwo\nmUwqk8koHo8rFotpZGRE77//vhoaGlRZWaloNKp4PG6zHAQYM2dmzIXkIxMzcjEjFzNysctqc9bS\n0qLKykqVlZVpfHxc69atU19fn7Zt25Y7prOzU7FYTA0NDWpqatLOnTslSfv377dZCgAAgC9Z375p\n6dKluX WlZUpm80qkUhobGxM5eXlmpqayh3L2Q0Uy6lTpzQ59Efu/ Zm5yGC/nj2OVfqceHxv7Px\nuh5XHvf19enFF190ph5XHrNeWC/zfRwOh7VQVrdvGhwcVF1dnSTp7Nmzikaj6uzsVDQaVWtrqyRp\n7969am9vl/T32bJHH30072cTtm9CId64/yY9sOFmr8twTiKR4PLDv5CJGbmYkYsZueS7nu2brJ45\n6 /vV21trVpaWtTf369oNKrR0VFdunRJw8PDSqfTGh4e1qZNm5TJZNTb26uWlhaFw2H19vZq48aN\namxstFkSAoqzsmb8xzMfmZiRixm5mJGLXWx8jpLExucAAC x8TnwL9znzIx7EeUjEzNyMSMXM3Kx\ni YMAADAITRnKEnMnJkxF5KPTMzIxYxczMjFLpozAAAAh9CcoSQxc2bGXEg MjEjFzNyMSMXu2jO\nAAAAHOKrW2mUN9zmdRnwifqqZVpdU F1GQCAgHLmJrTFxn2rAABAqeOyps9xnd MXMzIJR ZmJGL\nGbmYkYtdNGcAAAAO8dXM2ezm6QAAAC5j yYAAIASQXPmc1znNyMXM3LJRyZm5GJGLmbkYhfNGQAA\ngEOYOQMAALCMmTMAAIASQXPmc1znNyMXM3LJRyZm5GJGLmbkYpdvLmt 9913Onv2rNdlAAAAzKmu\nrk6bN29e0Gt905wBAAAEAZc1AQAAHEJzBgAA4BCaMwAAAIfQnAEAADiE5gy NTMzo0wm43UZziEX\nAPC3JV4XMGtkZERdXV0qLy/Xtm3btHr1aivH l0hn/Xzzz/P/VFet26dWlpaFqvMRdfZ2amhoSF1\ndHSosbHxmscGab0UkkuQ1svAwIBOnDihmZkZ3Xvvvbr55puvemyQ1kshuQRpvXz77bc6c aMJKmp\nqemat0MI0nopJJegrJepqSm99dZbev755 3 Lco6Ys ePbmfP/nkE2vH l0hn3Xfvn3FLscpAwMD\n2ZMnT855XJDWSzY7/1yCtF4OHTqU /mLL7645rFBWi F5BKk9XK5r7766pq/D9J6udxcuQRlvXz5\n5ZfZ7777zvrfImfOnIXD4dzPy5Yts3as3xXyWTOZjHbv3q1sNqvm5mb2Iv1/QVovhQjSennggQfm\nfWyQ1kshuQRpvUjSr7/ qo8//lhPPPHENY8L0nqR5p9LENbLn3/ qXA4rOXLl8/r ELWijPNWfay\ne EuXbrU2rF V8hn3b59e 7nL7/8smg1 U2Q1kshgrhe9u/fry1btlzzmCCul/nkErT1Eo1G9dJL\nL2nPnj1au3btVY8L2nqZby5BWC JREKPP/64jh8/Pq/jC1krzjRnlw8wh0Iha8f63UI/axD IzFf\nQVovCxWE9XLo0CHddtttWrly5TWPC9p6mW8ulwvCepGk5cuXq7q6 prHBG29SPPL5XKlul7OnDmj\nzz77TKOjo4pGo3POnBWyVpz5tubU1JSkvzvL2Z8lqb /X4ODg/M6thQVkstvv/2W /nixYuLU6Bj\ngr5eribo6 Xw4cNatWqVotHoFc8Hfb0UkkuQ1sv4 Hju58vPdgR9vRSSSxDWywsvvKCOjg61tbVp\n06ZNV/zueteKM2fO7rnnHn3wwQfKZrN65JFHcs8fPXpUoVBId9xxx5zHlqJCcvntt9/U3d0tSbrr\nrrsWvdbFdODAAQ0NDWn58uWKRqO52Zmgr5dCcgnKehkdHdWRI0d0  236/jx45qcnNRzzz0nKdjr\npdBcgrJeJOnIkSOanp6WJN19992554O8XqTCcgnKehkfH1cikcg7c3a9a4WNzwEAABzizGVNAAAA\n0JwBAAA4heYMAADAITRnAAAADqE5AwAAcAjNGQAAgENozgAAABzyf3YVQDCDNcmqAAAAAElFTkSu\nQmCC\n",

       "text": [

        "<matplotlib.figure.Figure at 0xb2d2908>"

       ]

      }

     ],

     "prompt_number": 78

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "# \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0435 \u043a\u0430\u0442\u0435\u0433\u043e\u0440\u0438\u0430\u043b\u044c\u043d\u043e\u0433\u043e \u043f\u0440\u0438\u0437\u043d\u0430\u043a\u0430 = \u0438\u043d\u0442\u0435\u0440\u0432\u0430\u043b\u044b \u043f\u043e\u043f\u0430\u0434\u0430\u043d\u0438\u0439\n",

      "x = np.random.randn(10000)\n",

      "y  = pd.cut(x,10)\n",

      "z = pd.value_counts(y)\n",

      "z.plot(figsize=(20,3))\n",

      "pd.DataFrame(x).plot(kind='kde')\n",

      "pd.DataFrame(z).T"

     ],

     "language": "python",

     "metadata": {},

     "outputs": [

      {

       "html": [

        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",

        "<table border=\"1\" class=\"dataframe\">\n",

        "  <thead>\n",

        "    <tr style=\"text-align: right;\">\n",

        "      <th></th>\n",

        "      <th>(-0.211, 0.629]</th>\n",

        "      <th>(-1.0502, -0.211]</th>\n",

        "      <th>(0.629, 1.468]</th>\n",

        "      <th>(-1.89, -1.0502]</th>\n",

        "      <th>(1.468, 2.308]</th>\n",

        "      <th>(-2.729, -1.89]</th>\n",

        "      <th>(2.308, 3.147]</th>\n",

        "      <th>(-3.569, -2.729]</th>\n",

        "      <th>(3.147, 3.987]</th>\n",

        "      <th>(-4.417, -3.569]</th>\n",

        "    </tr>\n",

        "  </thead>\n",

        "  <tbody>\n",

        "    <tr>\n",

        "      <th>0</th>\n",

        "      <td> 3247</td>\n",

        "      <td> 2638</td>\n",

        "      <td> 1921</td>\n",

        "      <td> 1192</td>\n",

        "      <td> 634</td>\n",

        "      <td> 226</td>\n",

        "      <td> 102</td>\n",

        "      <td> 32</td>\n",

        "      <td> 6</td>\n",

        "      <td> 2</td>\n",

        "    </tr>\n",

        "  </tbody>\n",

        "</table>\n",

        "</div>"

       ],

       "metadata": {},

       "output_type": "pyout",

       "prompt_number": 79,

       "text": [

        "   (-0.211, 0.629]  (-1.0502, -0.211]  (0.629, 1.468]  (-1.89, -1.0502]  \\\n",

        "0             3247               2638            1921              1192   \n",

        "\n",

        "   (1.468, 2.308]  (-2.729, -1.89]  (2.308, 3.147]  (-3.569, -2.729]  \\\n",

        "0             634              226             102                32   \n",

        "\n",

        "   (3.147, 3.987]  (-4.417, -3.569]  \n",

        "0               6                 2  "

       ]

      },

      {

       "metadata": {},

       "output_type": "display_data",

       "png": "iVBORw0KGgoAAAANSUhEUgAABMQAAADJCAYAAADM4e43AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1 /AAAIABJREFUeJzt3X90VPd95//XndGMRiPNSEK/kQAJDAgDNkay8aYEkhin\nJAYCdpqf63R72tOTs949u5ueNt3ubr/pt9/dbddN2u 3m7ab7rZpfpFfBgzYJgaa2Ag7wQJsS2Bh\nA0ZIAv3AQuj3r5n7/UPSMBIzkgYkzY/P83GOD3Ov7ow vq97r Ctz31fy7ZtWwAAAAAAAIAhHPEe\nAAAAAAAAALCQKIgBAAAAAADAKBTEAAAAAAAAYBQKYgAAAAAAADAKBTEAAAAAAAAYhYIYAAAAAAAA\njJI239/gtddeU1tbmyRp6dKlqqqq0qFDhxQIBCRJK1as0Pr16yVJra2tOn78uJxOp7Zu3aqSkpJp\n1wMAAAAAAACxmveC2Ic 9KHQ65dfflmS5PF49Pjjj9 x7alTp/TFL35RknTgwAHt3r172vUAAAAA\nAABArOa9ICZJV65c0XPPPac9e/ZIkgKBgPbt2yfbtlVRUaGNGzdKkrxeb g9brc79DraegAAAAAA\nACBWC1IQKy8v1zPPPKPnn39ey5cv1/bt20Nfe mll0KvbdsOvXa5XDOuBwAAAAAAAGK1IAUxaew2\nSZ/Pd8f68ALXRF8xSbIsa8b1U7388styOp33OlQAAAAAAACMy8nJUVVVVbyHMafmvSDW2dmpRYsW\nSbo906uxsVHLli2TJA0ODoa2HRgYCG038Xq69VM5nc7Q7ZcAAAAAAAC4d2fOnIn3EObcvBfETp48\nqaGhIUlSdXW1pLGC2KlTpyRJGzZsCG27adMm7d27V7Zta9u2bTOuBwAAAAAAAGJl2eENupLc8ePH\nmSFmkJqaGm3evDnew8ACIW zkLdZyNss5G0W8jYLeZuFvM1y5swZPfbYY/EexpxyxHsAAAAAAAAA\nwEJihhgAAAAAAACiYoYYAAAAAAAAkOQoiCFp1dTUxHsIWEDkbRbyNgt5m4W8zULeZiFvs5A3kh0F\nMQAAAAAAABiFHmIAAAAAAACIih5iAAAAAAAAQJKjIIakxT3rZiFvs5C3WcjbLORtFvI2C3mbhbyR\n7NLiPYC59h8Ovat1RZlaW5yl wsz5fek3P8iAAAAAAAA7kHK9RCzilaqvq1X9a19utDRp4Ist9YV\nZWpdcZbWFWWpMMsly7LiPVQAAAAAAICkkIo9xFJu tRDpT49VOqTJAWCti51Dqi tVcnr9zSt37V\nIqfDChXI1hZlqjw3Q04HBTIAAAAAAABTpHQPMafD0qp8r55cV6g/3lahH35hnZ795EpVl/l18caA\n/us/X9Gnv1enPzpyUXvfbNXb13s1NBqM97AxS9yzbhbyNgt5m4W8zULeZiFvs5C3WcgbyS7lZohN\nx7IslWanqzQ7XR9flSdJujkwonNtfTrX2qu/P9WiKzcHtWJRhtaGzSKjDxkAAAAAAEDqSLkeYhs3\nbrynzxgYCehCR7/q2/pU39qrhvY FWS6tbY4U uKsrS2OFPFWW76kAEAAAAAACPQQ8wAGS6nNiz2\nacPi233ILo/3Ifvl1Vv636daZFlW6EmW64vpQwYAAAAAAJBMUrqH2FxwOiytzPdqz7pC/efHKrT3\nC v0jR0r9chSv97vHNB/  creuq7b uPjlzU98 26q1rPRqkD9mC4J51s5C3WcjbLORtFvI2C3mb\nhbzNQt5IdvM Q y1115TW1ubJGnp0qWqqqpSa2urjh8/LqfTqa1bt6qkpESSYl4fD5ZlqcSfrhJ/\nuh5fOdaHrGtgROfb 1Tf2qd/qL2my52Dqsj1aF1xltYVZ2ptUZay6UMGAAAAAACQEBa0h9jLL7 s\nj3/84zp48KB27dolSTpw4IB2794tSTGvn2oueojNhcHRoN7t6FNda5/OtfXqfFuf8ryuUJP dcVZ\nKvHRhwwAAAAAACQ eojdpStXrui5557Tnj17JElerzf0NbfbHXod6/pE5Ulz6IESnx4oud2H7MrN\nAdW39umN5m79Y 112bLHmvSPF8iWL6IPGQAAAAAAwEJYkB5i5eXleuaZZ/TGG29IksInpblcrtDr\nWNcnC6fD0oo8rz61tkD/6WMV sHn1 ovd67So0uz1dg1qP/xi0Y99d239YcvXdT3zlzX2Ws9GhgJ\nxHvYCY971s1C3mYhb7OQt1nI2yzkbRbyNgt5I9ktWGMrj8cjn298xlTgdrEn/LbBWNdHUlNTo82b\nN4deS0rI5RJfui699YaqJf37T29W9 CofvLzN3Txygeqbc7Vpc4B5aWNaGlGUB vWqW1RZk6d/pX\nCTP RFiuq6tLqPGwTN4skzfL5M0yebNM3iYvk7dZy Rt1nIqmvceYp2dnVq0aJEk6YUXXtATTzyh\n/fv3a8 ePbJtW4cPH9bOnTslKeb1UyVKD7G5MDQa1IWOfp1r61V9a5/Ot/cpNyMtdIvluqIsLfbT\nhwwAAAAAAMwveojdhZMnT2poaEiSVF1dLUnatGmT9u7dK9u2tW3bttC2sa5PZelpDj1QkqUHSrIk\njfUha7w5qPq2Xp1u7tY/1V5XwLa1tmjsSZbrirK0Io8 ZAAAAAAAADNZ0KdMzrdUmiE2E9u21d47\novq2Xp1r7VN9W6/ae4e1usAbmkFWWehVhssZ76HOm5qampSevonJyNss5G0W8jYLeZuFvM1C3mYh\nb7MwQwwJw7IsFfncKvIt0mP3jd2S2j04qvPtfTrX2qvvnLmuix8MaFmOR2vHZ5CtK8pUrjf5HkoA\nAAAAAAAwl5ghlsKGR4N690Z/aBbZubY  T1pWl cGbrVstSfTh8yAAAAAAAQFTPEkFTcaY6x2yeL\ns6QHpaA93oestVdnr/Xoe2eva2jU1rqiTK0tztL64kytyPMqjT5kAAAAAAAghTniPQAsHIdlqWJR\nhnbeX6D/ NFyfe9z6/TN3au1ZXmOrnUP6RuvXtVT331bf/Die/rO6es63dyt/uFAvIcd1cRjYGEG\n8jYLeZuFvM1C3mYhb7OQt1nIG8mOGWKGK8xyqzBrkT66YqwPWc/QqN5p71N9a5  /2arLt4YUFl2\n vhMs7FbLfPoQwYAAAAAAJIYPcQwreFAUO919Ku rU/1rb06394nX7pzrAfZ K2WS7LpQwYAAAAA\nQKqihxiM43Y6tLY4S2uLs/TZB4sUtG1d7RpUfWuf3m7t1Q/ebNPgaFBrizJDBbL78jLkcnI3LgAA\nAAAASExULRATh2WpPDdDO9bk66sfKdd3P7dWf7NntbYuz1Vr77D 35omffp7dfr9F97Tt2uvqba5\nW33z1IeMe9bNQt5mIW zkLdZyNss5G0W8jYLeSPZMUMM96wg062PrnDroytyJUl9wwGdb tTfVuv\n9r7Zpvdu9KssO33sNsviTK0rylJeJn3IAAAAAABAfNBDDPNuOBDUxRsDqm/r1bnWsUJZptsZusVy\nfVGWluTQhwwAAAAAgEREDzHgLridDt1flKn7izKlB6Sgbau5a0j1bb2qb 3Vj95qU/9wQGuLsrR2\nfAbZynz6kAEAAAAAgPlBxQELzmFZWprr0Scr8/UHHynXdz67Vn/3ZKU dl uOnpH9D9fa9JT363T\n7x1 T/9Ye01vNEXuQ8Y962Yhb7OQt1nI2yzkbRbyNgt5m4W8keyYIYaEkJ/p1tblbm1dfrsP2Tvt\nfapv7dWP327ThY5 Lfana11xZqgXGQAAAAAAwN2ghxiSwkggqIsfDOhca6/q2/p0rq1PvnSnqkp9\nqirz68GSLGW4nPEeJgAAAAAAKYceYkCcuJwOrSnM1JrCTH1aY33ILn8woNqWbj1X167//vMrWpXv\nVXWZX1WlPi3Py5CDJv0AAAAAACACeoghKTksS60NZ/S5B4v17BMr9cMvrNNT6wt1o29Y//Wfr hz\n36/Xn//iio6916mb/SPxHi7mAD0KzELeZiFvs5C3WcjbLORtFvJGspv3GWLnzp3ThQsXFAwG9eij\nj6qsrEyHDh1SIDDWJH3FihVav369JKm1tVXHjx X0 nU1q1bVVJSMu16YEKGy6lHl2br0aXZkqTr\nPUM63dyjk1e69M3Xm1Xsc6t6/PbKtUWZPMESAAAAAACDzXsPsVdffVVbtmyRJL344ov65Cc/qaNH\nj rxxx /Y9uDBw9q165dkqQDBw5o9 7d066fih5iiGQ0aKuhvU 1zd063dKjpq5BrS/OUlWZX9Vl\nPpX602VxeyUAAAAAABHRQ wuTBTDwgUCAe3bt0 2bauioiJUxPJ6vaFt3G536HW09cBspDksrSvO\n0rriLP2raql7cFRnWnp0uqVbP3qrTWkOS9VlY7PHHlrsU6ab5vwAAAAAAKSyBbtv7OjRo3rkkUck\nSdu3b9eTTz6pp556Sm1tbaFtwieruVyuGdfDbHd7z7rfk6aPrMjV721Zph98fq3 9NeXqzTboxfe\nuaEv7K3Xfzj0rr53tlUN7X0KBFPmIaxJjx4FZiFvs5C3WcjbLORtFvI2C3kj2S3IUyZfeeUVrVy5\nUvn5 Xd8LbzANdFXTNKkW9iirY kpqZGmzdvDr2WxHKKLtfV1c3Z55XnZqj41rv6eKbkW16h083d\n tMjDeodtfTIslxVlfkVvNYgv8tOmP9/05bnMm WE3 ZvM1aJm zlsnbrGXyNmuZvM1aJm zllPR\nvPcQO3HihAoKClRZWRla19jYqGXLlkmSDh8 rB07dkiS9u/frz179si2bR0 fFg7d 6cdv1U9BDD\nXOvoG9bp5h6dbu7WmWs9WuR1qarUp oyv9YXZyk9jeb8AAAAAIDURg xGLW3t vkyZOqrKxUQ0OD\nenp69PTTT6uxsVGnTp2SJG3YsCG0/aZNm7R3717Ztq1t27bNuB6YbwWZbm1fnaftq/MUCNp670a/\nalt69P2zrbrcOaD7CzNDzfmX5Xhozg8AAAAAQBKY9xliC4kZYmapqamJ6/TNvuGAzl4bmz1W29yj\nQNBWVZlPVaV bSz1ye Z13qzceKdNxYWeZuFvM1C3mYhb7OQt1nI2yzMEAMQkul2anN5jjaX58i2\nbbV0D6m2uUfHL3bqr2quakmOJ3R75ZrCTDkdzB4DAAAAACARMEMMmAfDgaDOt/XpdHO3Trf0qLVn\nWA WZKmqzK qMp9KfOnxHiIAAAAAALPCDDEAs J2OrRhsU8bFvv025Ju9o/odEuPTrd06zunryvT\n7VR1mU9VZX49WJKlDJcz3kMGAAAAAMAYPCIPSWviMbDJINfr0raVi/TVj5Trh19cp//8WLnyM916\nrq5dn/tBvX7/hff0w7dadfFGv4KpM2lzTiVT3rh35G0W8jYLeZuFvM1C3mYhbyQ7ZogBC8xhWVqR\n59WKPK8  2CRBkYCeut6r0439 i//fyKeocCoeb8VaU 5Xpd8R4yAAAAAAAphR5iQIJp7Rlrzn 6\nuVtvXe9Vkc t6tKx2yvvL8qU28nETgAAAADAwqGHGIB5V xL14416dqxJl BoK2G9j7VtvTo/7xx\nTU1dg1pfPNacv7rMp1J/uiyLp1cCAAAAABALppogaZlwz7rTYWltcZZ s6pEf/2p1frOZ9dq28pF\nuvRBv/7ghYv60o/O669qrqrm/S71DQfiPdx5ZULeuI28zULeZiFvs5C3WcjbLOSNZMcMMSCJ D1p\n2ro8V1uX58q2bTV2Daq2uUcvXrihZ19t1PJFGaoqG s9tirfK6eD2WMAAAAAAExFDzEgRQyNBlXX\n2qvTzd063dKjD/pHtHHxWO xqjKfCjLd8R4iAAAAACAJ0UMMQMJKT3Oousyv6jK/JOlG37BOt/So\ntrlb//tUi3IzXKoq86m6zK/1xVlKT OOaQAAAACAmfgXMZIW96xPLz/TrV9flaf/9LEK/eiL6/V7\nW5bKl56mH5xt1We X6c/fOmifvp2m97vHFAyTBQlb7OQt1nI2yzkbRbyNgt5m4W8keyYIQYYwOmw\nVFmYqcrCTH3xoWL1DQd09lqPzjT36ODRyxoN2Koq86mq1K NpT75PVwaAAAAAACpix5igOFs29a1\n7iHVNvfodEu33r7eqyU5HlWVjt1eWVmYqTSa8wMAAACAseghBiDlWJal0myPSrM9 tTaAo0Egjrf\n1qfalh79zevNut4zrAdLslQ93py/xJce7yEDAAAAAHBP6CGGpMU96/PD5XTowcU /fbDi/U3eyr1\nD59eow9X5Oh8W6/ /cF39Vs/Pq9vvtak1xtvaWAksGDjIm zkLdZyNss5G0W8jYLeZuFvJHs5n2G\n2Llz53ThwgUFg0E9 uijKisrU2trq44fPy6n06mtW7eqpKREkmJeD2D 5Xpdeuy RXrsvkUK2rbe\n7xzQ6eYe7T/Xrj/7xRWtyveG o tyMuQw L2SgAAAABAYpv3HmKvvvqqtmzZIkl68cUX9clPflIH\nDx7Url27JEkHDhzQ7t27JSnm9VPRQwxYWAMjAb19vTfUf6x3KKCN473Hqkp9yvW64j1EAAAAAMA9\noofYXZgohoXzer2h1263 67XA4ivDJdTm5Zma9PSbElSa8 QTrf06LXGLv3N680q8rlVVepTVZlf\na4sy5XZylzYAAAAAIP4W7F nR48e1SOPPCJp7Kl2E1yu2zNIYl0Ps3HPeuIp9qXricp8/fG25frJ\nv1yvf/OhMrmdDv3jG9f0me/V6b/87JL217erqWtQsU5OJW zkLdZyNss5G0W8jYLeZuFvJHsFuQp\nk6 88opWrlyp/Px8SVIgcLsRtxXWbyjW9ZHU1NRo8 bNodeSWE7R5bq6uoQaD8uTl19/7aQk6Uub\nN tLVSU6 kqNLvf16/1Ol35S166RoSGtyApoR9UqPVTq09lTr0/7eeRt1jJ5m7VM3mYtk7dZy Rt\n1jJ5m7VM3mYtp6J57yF24sQJFRQUqLKyMrRu//792rNnj2zb1uHDh7Vz5867Wj8VPcSA5GDbtq52\nDYZ6j51r61NFboaqy8Zur1yV75XTQXN AAAAAEgE9BCLUXt7u06ePKnKyko1NDSop6dHTz/9tDZt\n2qS9e/fKtm1t27YttH2s6wEkJ8uytCw3Q8tyM/TU kINjQZV39qr0y09 ssTV/VB/4geWjxWHKsu\n86kgk96BAAAAAIC5M 8zxBYSM8TMUlNTk9LTN012o29YZ1p6VNvcrTMtPcrNcKnE6tGnNlVqfXGW\n3Gk05091nN9mIW zkLdZyNss5G0W8jYLM8QAYAHkZ7r18VV5 viqPAWCti5 0K/nTtbre2db9X7n\ngNYVZ6m6zK Hy3xa7E fsbcgAAAAAADhmCEGIKn0DI3q7LUe1TaNzSBzOS1Vl/lVXebXhsVZynA5\n4z1EAAAAAEgpzBADgDjzpadpS0WutlTkyrZtXbk5qNrmbh04164/ 8UVrS7wjs8e86s818PsMQAA\nAADAHWjEg6Q18RhYmCFS3pZlqWJRhn7jgSL9 SdX6odfWKcn1xWqvXdYXzt6WV/Ye05ff7VRr16 \nqZ6h0TiMGneL89ss5G0W8jYLeZuFvM1C3kh2zBADkDIyXE49ujRbjy7NliS13BpSbXO3jr7XqW c\nuKry3AxVLxnrPbYy3ysHs8cAAAAAwEj0EANghOHRoOrbelXb3KM3mrvVNTCqjaU VZf5VF3qV67X\nFe8hAgAAAEBCoocYACQpd5pDG0v92ljq1 9uKlV777BON3fr9cZu/e3rLSr2uUPN e8vylSag9lj\nAAAAAJCq6CGGpMU962aZ67wLs9z6RGW /nhbhX78L9frX/ LMlmW9L9 1azf F6d/uToZb3QcEPt\nvcNz n0xO5zfZiFvs5C3WcjbLORtFvJGsmOGGADjpTksrSvO0rriLP1W9WLdHBjR6eYe1TZ369u1\n15XtSRu7tbLMrweKs RO43cJAAAAAJDM6CEGANMI2rYu3hjQG83dqm3u1vudA1pblKXqMp8eXuJX\nqT9dFs35AQAAAKQweogBgGEclqVVBV6tKvDqiw8Vq2doVGev9ai2qUc/ebtdaU5L1WV PVzm14Ml\nWfK6nfEeMgAAAABgBtz3g6TFPetmSZS8felp2lKRq69sWarvf36t/uTx5SrxuXXgXLs v7dev//C\ne/rxW226/MGAUmgC7oJLlLyxMMjbLORtFvI2C3mbhbyR7JghBgB3ybIsVSzKUMWiDP3GA0UaGAno\nreu9qm3u1v99/LKGRu1Q77GHFvvk93DJBQAAAIBEQA8xAJgnLbeGVDvee6yutVfluRmhAtnKfK c\nDnqPAQAAAEh89BADAMxaaXa6SrML9Km1BRoeDaq rVe1zT36xomr6uwfUVWZX9VlPlWV rXI64r3\ncAEAAADAGPPeQywYDCoQCMz3t4GBuGfdLMmetzvNoY2lfv3uplJ966k1 tsnK7VhsU /vNqt3/np\nO/rX xv0D29c09vXezUaTJmJu3ct2fNGbMjbLORtFvI2C3mbhbyR7OZ1htiRI0d0 fJl7d69W4sX\nL5YkHTp0KFQgW7FihdavXy9Jam1t1fHjx V0OrV161aVlJRMux4AkllBplufWJ2nT6zOUyBoq6G9\nT280d tbv2pRS/eQNpRkqXqJX9WlfhX53PEeLgAAAACklHnvIXb /Hnl5OSECmJHjx7V448/fsd2\nBw8e1K5duyRJBw4c0O7du6ddHwk9xACkgpsDIzrT0jPef6xH2Z60UO xB4qz5E7jAcEAAAAAFg49\nxOZAIBDQvn37ZNu2KioqQgUsr9cb2sbtvj0bItp6AEhVuRkuPXbfIj123yIFbVsXPxhQbVO3fnC2\nVf9P54DWFmWpusynh5f4VepPl2XRnB8AAAAAYrHg0wy2b9 uJ598Uk899ZTa2tpC68MnqrlcrhnX\nA9yzbhZT83ZYllble/WFh4r1jZ2r9L3PrdX21Xm6cnNQf/DiRf3mj8/r/zvZpNcbb6l/OHX6NZqa\nt6nI2yzkbRbyNgt5m4W8kezi pTJ8AJXeOP98NkO0dZHU1NTo82bN4deS2I5RZfr6uoSajwsk/dC\nLX 4IkdWS72qy6Ql66pU29Stb598Ty2DDq0p8unhMr cHRdVmG7rwx O/3jvZpm8zVomb7OWydus\nZfI2a5m8zVomb7OWU9GC9xBrbGzUsmXLJEmHDx/Wjh07JEn79 /Xnj17ZNu2Dh8 rJ07d067PhJ6\niAEw2cBIQG9f71Vtc7feaO7R4GhAD5f5VV3m10OLffJ70uI9RAAAAABJiB5iMTp27JguX74sj8ej\n8vJybdmyRY2NjTp16pQkacOGDaFtN23apL1798q2bW3btm3G9QCAyTJcTm1amq1NS7MlSS23hnS6\npVvH3uvUX564qvLcjFBz/pX5Xjkd9B4DAAAAYKZ5nyG2kJghZpaampqUnr6Jycj73gwHgjrX2qc3\nmrtV29ytzv4RVZX5VV3mU1WpX4u8idWjkbzNQt5mIW zkLdZyNss5G0WZogBAJKS2 nQQ6U PVTq\n0 9uKlVH37Bqm3v0y6vd tvXW1Tkc6u6zK Hy3y6vyhLacweAwAAAJDCmCEGAIYLBG01tE/MHutR\nS/eQHizJGi Q VXkc8d7iAAAAADiiBliAICU43RYWlucpbXFWfpX1dLNgRGdaelRbXO3/un0dfnT\nnapeMlYcW1 cpfQ0R7yHDAAAAAD3hH/VIGlNPAYWZiDvhZOb4dJj9y3SVz9Srh99cZ2  tFyZaen\n6QdnW/XZ79fpj45c1P76djV1DWq JhmTt1nI2yzkbRbyNgt5m4W8keyYIQYAiMphWVqV79WqfK  \n8FCxeodGdfZar2qbu/WTunY5LUsPl/lVvcSnDSU ed3OeA8ZAAAAAGZEDzEAwF2xbVuNXYOqberW\nG809aujo06p8r6rHn165fFGGLIvm/AAAAECyo4cYAADjLMtSeW6GynMz9OkHijQwEtDb18dmj/3p\n8SsaHA2outSv6jK/Npb65PfwIwcAAABAYqCHGJIW96ybhbwTX4bLqU1Ls/XMh5bo25 5X19/YpVW\nFXh1/GKnvvSjc/p3By/ou2eu6532PgWC009OJm zkLdZyNss5G0W8jYLeSPZ8et6AMC8KM1OV2l2\ngXbdX6DhQFDn2vpU29StvzpxVR/0j2hjqW/89kq/Fnld8R4uAAAAAIPQQwwAsOA6 oZV29yj2uZu\nnW3pUZHPreoyvx4u8 n oiylOeg9BgAAACQKeogBADAHCjLd sTqPH1idZ4CQVsN7X2qbenR35 6\npuZbQ1pd4FVlgVdrCjO1usCrnAxmkAEAAACYO/QQQ9LinnWzkHfqcjosrS3O0m9WleivP7Va3/7M\n/VpldciWtK  Q7/1k3f0mz86p// 8yvaX9 ud9r7NBwIxnvYmEOc32Yhb7OQt1nI2yzkjWTHDDEA\nQELJ9qRpVVZAm6sXS5KCtq3mriE1dPSpob1fR9/rVNOtIZXnelRZ4NXqgkytKfRqsT9dlsWtlgAA\nAABmRg8xAEDSGRwN6uKNfr3T3qeGjn41tPdpcDSoyoJMVRZ6VVkwdqul38PvfQAAAIB7RQ8xAAAS\ngCfNoXXFWVpXnBVa90H/iBrGC2Q/frtN793oV26GK1Qgqyz0avmiDLmcdAsAAAAATDfv/yoIBoMK\nBALz/W1gIO5ZNwt5m Vu8s7zuvRr5Tn67YcX69knVuq5px/QH2 r0AMlPl3uHNA3Xr2qJ79bp393\n8IL 9pfN vmlm7reM6QUmiidtDi/zULeZiFvs5C3WcgbyW5eZ4gdOXJEly9f1u7du7V48VgvmNbW\nVh0/flxOp1Nbt25VSUnJXa0HAGA6ToelikUZqliUoU szpMkDYwE9G5Hvxo6 vXq5Zv61q9aNBq0\nVVngVWVh5nhPMq y0plADQAAAKSyee8hdv78eeXk5IQKYgcPHtSuXbskSQcOHNDu3bvvan0k9BAD\nAMSqo29YDe1jfcje6ejTxRsDKsh0hQpkawozVbEoQ04HDfsBAABgJnqIzQGv1xt67Xa773o9AABz\noSDTrYJvnhywAAAadklEQVQKtz5ckSNJCgRtXbk5oHfa 3Who08Hz99QW  w7svLCBXJKgszVZDp\n4qmWAAAAQJJa8M7C4RPSXC7XXa8HuGfdLORtlnjm7XRYWpHn1Y41 fq9Lcv0959eo71fWKenN5bI\nl 7U8Ys39W8OXNDnf1Cv/ voZe19s1Vnr/Wof5h mXeL89ss5G0W8jYLeZuFvJHsFnyGWHiD/fDf\nrMe6Ppqamhpt3rw59FoSyym6XFdXl1DjYZm8WU7dvM eel2S9Pnx5RMnanRrxFLWsjI1dPTpr39 \nQW2DDpXmZKiyIFNp3ddU6glo98c JKfDivv4E3050fJmmbxZJm WyZtl8mZ58nIqWvAeYvv379ee\nPXtk27YOHz6snTt33tX6SOghBgCIl9GgrcudA2po71NDx1hPsg/6R7Qyz6vKQq8qCzJVWehVfia3\n/wMAACC50EMsRseOHdPly5fl8XhUXl6uLVu2aNOmTdq7d69s29a2bdtC28a6HgCARJLmsLQq36tV\n V7tGl/XMzSqC PFsSPvfqC/qrkqt9MxViArzFRlQaZW5mcow WM69gBAAAA08z7DLGFxAwxs9TU\n1KT09E1MRt5mSdW8bdvW9Z7hSbPI3r85qFJ/ qRZZEtzPHIY1LA/VfNGZORtFvI2C3mbhbzNwgwx\nAABw1yzL0mJ/uhb70/Wx xZJkoYDQV3 YEDvtPfp7LUe/fCtVnUNjGpVwViBbM34ky1zvTxYBgAA\nAJgrzBADACDB3Boc1YWOPjW096uho08XOvrldTlVWeDV6sJMrSnw6r58r9LTFvxh0QAAADAQM8QA\nAMC8y/ak6ZEl2XpkSbaksVstW7qH9E77WJHs55c6dfXmoJbkeMZ7kXm1pjBTpdnpRt1qCQAAANwt\nfrWMpDXxGFiYgbzNQt6TWZalsmyPHl Zp3/7a0v0zd2V unTD iZD5Wp1J uN5q79UdHLunT363T\nf3zpov7p9HX96uot3RocjffQZ4W8zULeZiFvs5C3WcgbyY4ZYgAAJKH0NIfWFmVpbVFWaN3N/pGx\nZv0dfdpX364LHf3K9qSFZpFVFmZqRV6G3E5 HwYAAACz0UMMAIAUFbRtNXUNhp5o2dDRr ZbQyrP\n9YSeaFlZkKnFfrcsbrUEAABAFPQQAwAAScNhWVqWm6FluRn69VV5kqSBkYAufjCghvY vdZ4S//w\nxjUNjQa1OqxAtrrAK7 HvyIAAAAgdXHPBJIW96ybhbzNQt7zJ8Pl1PriLP3GA0X6L49V6PufX6f/\n9eQafaIyT8OjQf347TZ96Ufn9Fs/Pq8//8UVPX uQ 929GskEJy3MZG3WcjbLORtFvI2C3kj2fHr\nXwAADJeX6dLmzBxtLs RJAWCtq52DYZus3yh4Yau9wxrxaIMrR6fRVZZ6FVxFrdaAgAAIDnRQwwA\nAMyofzigd2 MNexvaB/7MxBUqFn/xJ Zbme8hwoAAIA5Rg8xAABgJK/bqQ2Lfdqw2CdJsm1bHX0j\noQLZ999s1cUbAyrMck8qklUsypDTwSwyAAAAJBZ6iCFpcc 6WcjbLOSd CzLUmGWW1sqcvW7m0r1\njR2rtO9LD gPP7JMlYWZamjv05/9olF7vvO2vnLoXX3rVy169fJNtfcOa rkdPI2C3mbhbzNQt5m\nIW8kO2aIAQCAOZHmsHRfvlf35Xu1Y02 JKlvOKAL47PIjl3s1P98rVkOS2MzyMb7kfUH4jxwAAAA\nGIceYgAAYMHYtq223mG9M96H7EJ7v67cHJDL6dCS7HQtyfFoSY5HS3PStSTbo8IsN7dcAgAAxBk9\nxAAAAO6BZVkq9qWr2Jeuj67IlTRWJOscGFVT16CaugZ1tWtItc3dauoaVPfgqEqzx4pj4cWy0myP\nPGl0fgAAAMDdoSCGpFVTU6PNmzfHexhYIORtFvI2y8mTJ7V582bleV2hpv0TBkYCaro1FCqWnXi/\nS023BnW9e0i5GS4tyUkfL5J5xotm6crxpMmymFWWqDi/zULeZiFvs5A3kl1cCmKHDh1SIDDWMGTF\nihVav369JKm1tVXHjx X0 nU1q1bVVJSMu16AACQ2jJcTq3K92pVvnfS kDQVmvPsK52Darp1qAa\n2vt09N1OXe0alGUpVBwLL5YV 7j9EgAAAGPi0kPs6NGjevzxx 9Yf/DgQe3atUuSdODAAe3evXva\n9VPRQwwAALPZtq2uwVE1dQ2FimVjs8uGdHNgRIv9433Ksm8Xy8qy05XhcsZ76AAAAAmLHmJzJBAI\naN  fbJtWxUVFaEiltd7 7e/brc79DraegAAgHCWZSk3w6XcDJceKMma9LXB0aBabo31KGvqGtQv\nG2/pJ2 3qeXWkPyetLGZZFOKZbkZ3H4JAACQiuJSENu fXvo9UsvvRR6HT5ZzeVyzbgeZuOedbOQ\nt1nI2ywLlbcnzaEVeV6tyLvz9sv23mE13RrU1ZuDuvjBgH5 6aaabg1pNGhrSXb67WLZ NMvF/vT\nuf3yLnF m4W8zULeZiFvJLu4N9UPL3BN9BWTNOm3sdHWRxJ UtbU1EgSyym6XFdXl1DjYZm8WSZv\nlpMz79dfOxlafmRJtmpqavRwrrR552bdGhzVC6 e0o0eh255lqiuoVfvtXape9RSaXaGlmSny rt\nUL7b1mOPrFdZtkdnT72eUPs30ZbjnTfL5M0yebNM3izHvpyK4tJDrLGxUcuWLZMkHT58WDt27JAk\n7d /X3v27JFt2zp8 LB27tw57fqp6CEGAAAWwtBoUC23hkI9ysb6lQ2p daQfG5nxKdf5nld3H4J\nAACSEj3E5khjY6NOnTolSdqwYUNo/aZNm7R3717Ztq1t27bNuB4AACAe0tMcWp6XoeV5GZPWB21b\nHb0joYb 73cO6JXLXWrqGtRwIDipR9mSHI WZntU4nfL5XTE6f8EAADATHGZITZfmCFmlpoa7lk3\nCXmbhbzNYkre3YOjar41/vTL8YLZ1a4hdfQNqyjLPV4gu10sW5Kdrqz0uPzucl6ZkjfGkLdZyNss\n5G0WZogBAADgrvg9abrfk6b7izInrR8OBHWte0hNXWPFsjMtPXr fIeauobkdTkmFcgmbsMsyOT2\nSwAAgHvBDDEAAIAEZNu2OvpGxmeTTZ5Z1j8cVFmEp1 WZqfLze2XAABgjjFDDAAAAAvCsiwVZrlV\nmOVWVdnkr/UOjarp1lCoWHb84k01dQ2qrXdYBZluLYlQLPN7 GsfAADABP5mhKTFPetmIW zkLdZ\nyDt2WelpWlOYpjWFk2 /HAkEdb17WFfHn375dmuvXmi4oaauQbmdjlCBLPzpl4VZbjkW8PZL8jYL\neZuFvM1C3kh2FMQAAABShMvp0NJcj5bmeiatt21bnf2joadfNnUN6ldXu9XUNaie4YDKstMn9Sib\nuP0yPY3bLwEAQGqihxgAAIDB ocDdzz9sqlrSNd6hpTndY0XyMKKZTkeZXP7JQAARqGHGAAAAFKK\n1 3UqgKvVhV4J60fDdpq7bn99MvzbX362bsf6GrXkJyWwmaT3S6WFWa55XTw9EsAAJD4KIghaXHP\nulnI2yzkbRbyTkxpDktl2R6VZXv0L5Zlh9bbtq2bA6OTnn55uqVHTbcGdWtgVIv9dzb0L8vxyDN \n yV5m4W8zULeZiFvJDsKYgAAAJg1y7K0yOvSIq9LDy72TfrawMjY7ZcTxbKaK7fU1NWma91Dys1w\naUlOuoa73TpzskkZaQ5luBzKcDnH/3TIk aUd3ydZ3yd1 WUJ83BzDMAADCn6CEGAACAeRUI2mrr\nHVZT16C6h0Y1MBLU4EhQ/SMBDYyOvR4YCWhgJDj232jY65GABkeDSnNYt4tnabeLZl6XQx6XM1Rg\n87jGi2pp4 sjbJNBkQ0AgJjQQwwAAACIkdNhabE/XYv96Xf1ftu2NRyw1T8SGC eRS6a9Y8X1doG\nR0PrB0ZvbzPxvsGRoAZHg3I5rNsFtPEZalNnrU0U3zLCimped/i2419Pc8jjcshhUWQDACAZUBBD\n0uKedbOQt1nI2yzkbZa7yduyLKWnWUpPc0gZczMO27Y1ODE7bfTOGWpjM9jGi20jQXUNjkadxTYY\nNtPNneaIUGC7XTQLL7h5xottYzPYwre5/b70tOQusnF m4W8zULeSHYUxAAAAGAcy5q4BdOp3Dn6\nzKBtayg0Iy18htqUAtv4Np39I PFtckFucHxYlv/SFDDo0Glp4Xd9jll1lp47zXPHbPbIs9i86Q5\nZCVxkQ0AgLlADzEAAAAgQQWC40W2CEWz/uHg My0QIRbRG9vH15gGxwJaCRoyzNeHMuYepvo Hpv\n2IMNIm1z zbSsdfpTosiGwCkMHqIAQAAAFgwToclr9spr9spyTUnnxkI2uOz1AKhGWqDI4HxGWy3\nbwHtHwmqdyigjt6RsULb8OSHIPSP92IbGAlodLzINnWGmjvNksvhkMtpjf3ncCht/LXbYcnlHP9a\n Gvnne J NrpGH fJbeThyQAAGJDQQxJi3vWzULeZiFvs5C3Wcg7/pwOS5lupzLdzjn7zIki28SD\nDSaeInr27XqtWrVGIwFbI8Hg2J8BWyOBoEaCY6/7R4JjyxNfC8b4evyzJCnNMblQFl5wczsjFOam\nbBetSJcW9to9yyKdy mQ05JRM c4v81C3kh2SVEQa21t1fHjx V0OrV161aVlJTEe0gAAAAAxkUr\nsvVeDujXynMWZAyBoK3h8cLaaDBy4Ww4aN9RfBsN2BoJjj3JdOJrUYt0Ed4/XZHOthWxUBZepEsL\nrY 83dQiXVqEgl20Il2a05J7SsHOtCIdAESTFD3EDh48qF27dkmSDhw4oN27d0fcjh5iAAAAABJF\nIDhDEW28SDcapaAWXqSb JxQsW98faQi33QFv6hFuvDC3DSz3iYV46asd1qWHJbkGP/TsqxQAS58\nvcOyZFmSc/xPR5TtrCnvif7 6T/bQREQuGf0EIsTr9cbeu12u M4EgAAAACYHafDktNhyZPmiPdQ\nQqYt0s1QvAsvrI2O3ybbM3T79tdA0FZQUjBoyx7/c2I5aI89iTVoS7ZsBYJjfwaDt9cH7bH3BcLf\nP937wt5/x/vC3m9LsqQ7inVTC2nhBbVI6 /1fc4oRb5IBbzw90T67In3RBpT6H0Oa/z/O/pYnQ7J\n0gzvd1hyKHIh8raxhfC6o3XHi9svrbCVkWqVkT7HivBB4W d9vuFfWCk0mjE7xdhYJO X8RxW3EZ\nF 5OUhTEwiexuVxz00wUyY971s1C3mYhb7OQt1nI2yzknXjms0iXqHnb9u3CWHBK0S4QtMcKarbG\nim Rim22FAhtM6XYZtsK2Le/x9TiXTD8fRH jPb iN/XlkaD9vi2U983 T2Rvu/Y1yJ/r/Btphtr\nMPT50uDAgDwZGRrbs1L4vWd2aN9HyEO3V0583Z68wR3r7AgrI32PSLe/TdQTIn8t0ufdueWksUT6\nfmEr7alfC1uY/DmxjWs6MxbqQl LUICLWHy88wP/5IFZDCTJJMUtk0eOHNH27dslSceOHdO2bdsi\nbnf69Gl1dXUt5NAAAAAAAABSWk5OjqqqquI9jDmVFDPEBgYGJI1VUCdeR5Jq4QAAAAAAAGDuJcUM\nsWvXrumVV16Rbdvatm2bCgsL4z0kAAAAAAAAJKmkKIgBAAAAAAAAcyVxHncCAAAAAAAALICELogF\ng0EFAoFZr8f8Yr9Pj M1/tjXcyNZ9mOyjDPRsR RrDh2zUX2QOw4byBxHEzl/NrXvva12Ww4NDSk\nY8eO6b777pt2u9bWVj3//PM6f/688vLy5PP5pt3 3Llzqqmp0blz5 T3  X3 yWNPVny6NGjWrJk\nyaTPiLY VrGOU5IuX76sY8eO6dKlS6qoqJDT6Yw6/vb2dh06dEiXLl2S2 1Wbm6uJOnSpUv66U9/\nqpKSknsa/1yZba6x7vdo /fQoUN655131NDQoJGRERUVFU27fbT9O1fidbwm2nEQbuoxEa9zJVZ3\nc20YGBjQX/zFX2jFihWT3hNp/O //76OHz u vp6 f1 ZWdnS0rsLMPN17kebb/MVa5zNc75viYl\n63EQr/MmVrH LHjttddUW1urhoYG9ff3a/HixZKi55rs53e42Z7rse7TZ599Vh0dHWpoaFBNTY02\nbtw47eck2jUg1mtVKmQfr/MmVtG bzTTZR/p hTt2E2WjGd7Tse6H6P9/JMiX8/n 5yO1zUpWY D\neJ03sYp1nNFyina9MeX8lqL//SuW7aMdB3OVdzTxuj7N6jiwZ lnP/uZ3draOuN2zz//fOj1/v37\nZ9z lVdeCb1 4YUXJn3t3LlzdktLyx3vibY FrGO8 bNm3ZNTc0d66ON/8iRI6HXJ06cmPSeuRj/\nXJltrrYd27ij7d XX345pu2nOz7mQjyP10Q6DsJNPSbiea7EKtZ9 tJLL9mnT5 e9J5o43/xxRdD\nr2ebfSKZr3M92n6Zy1yjSbRrUrIeB/E6b2JxLz8Lfvazn4VeR8s12c/vcLM912PdpwMDA6HXL730\n0oyfk2jXgLu5ViV79vE8b 5W PedTrRsIl2foh27031OIonl53f4e2YS7edftOv5fJ/T8bwmJftx\nsNDnzd2azTij5RTtemPS R0pp7vZPtJ mY 8o1no69NMx8Gsb5m8cuXKpMpcNF6vN/Ta7XbPuP2W\nLVtmO4Q5Fes433zzTWVmZmrfvn169913Q ujjT89PV0DAwMKBAJ68803NTw8fO DngezzTVW0fZv\nIBDQvn379Nxzz nMmTMzbj/fx0eyHK8Laeoxkarnys2bN X1euXxeCatjzb QCAQmmJsJ GzSObr\nXI 2XxLtGpgs16T5dq/HQbzOm7vZ71euXNHXv/71Sb91ne44SObzO9xsM451n05k3t7ePulp34l2\nbY8mWa5V92Jq9sl03kT6vrGKdn2Kduwmi1iu27Hsx2g//6Jdz f7XDHtmhSrSMdBspw3sYwzWk7R\nrjemnN/Rcpqr7RdiPybq9SltthuGf2h/f7/ 7u/ btLXv/zlL8vr9U46QF0u16wHcvToUT3yyCOz\n3v5exTrO9vZ2SdKTTz6pF154Qffdd58cjtv1xKnj37x5s44cOaLh4WGtWrVKfX19syomLLTZ5hqr\naPt3 /btodcvvfTSjNtPuJfjIxWO14U09QKSqudKTU2NnnjiCTU0NMxq/OvWrdOzzz4rSXr66afn\nfXxzbb7O9Wj7ZS5yTdVrUjzd6z8A4n3exLLfy8vL9cwzz j555/X8uXLJUXPNdnP73CxnuuxHsvn\nz5/Xo48 esf6 bi2J8u1KlFEO7 T4byJ9H1jFe36NCHasZvoYjmnY9mP0X7 RbueL9Q5ncjXpHiK\ndH4ny3kTyzij5TTT9SbVz  Zcpoq1u0n3Mt TNbr06wLYuH/oPV6vfrKV74ScbvwBm2WZc3qs195\n5RWtXLlS fn5sx3OPbubcU78xqG0tFSdnZ2h8UYaf1pamnbs2CFJOnDggDIzM dq6HNqtrnGajb7\nN/wvVdNtf6/HRyocrwsp/JiQUvdcaWtr08GDB9Xe3q7y8vJJ97JHGv9bb72lr371qwoGgzp8 LB2\n7dq1IOOcK/N1rkfbL3ORa6pek Jp6vkdq3ieN3ez3z0ez6SeEdFyTfbzO1ws5/rd7NPBwcE7fuM8\nX9f2ZLlWJYpI53cynTdTv2 sprs SZGP3WQQ68/vu9mPU3/xE l6vhDndKJfk Ip2s/vZDlvZjvO\naDnNdL1J9fN7ppzudfsJ97Ifk/X6NOdPmRwYGJA09tukidcT6uvrdf78 UnrTpw4oaKiIpWXl9/z\n937xxRd1 fLleRnnqlWrQp998 bNUCO/mcbf1dWl3t7epPoNRKwi7a9o 7exsTH0enBwcMbtp9u/\nseQdTTyP12QRz3NlLjKONs7f Z3f0e7du7V582atW7duxvFPjMvhcMQ0mzCVRNqPM 2X cw1lnEu\nxDUp1cz3eRPLcTDdfo80zs7OztDr8Nkt0XI18fyOdZ9KY0 mmvqPsnhe26OZq2tVsovneRNLxtG \nb7RxRhPt iRFPnZTTaz7MdrPv2jX8wnzdU7H85qUjOJ53izE T01p muNyac39PlFMvf16YzF3lH\nE8/r00xmPUMsJydHIyMjM/5FcdOmTdq7d69s29a2bdsmfe3111 XZVm6//77JY1NeTt58qQqKyvV\n0NCgnp6e0BTIY8eO6fLly/J4PCovLw9VAqOt7 /v1969e/XNb35zVv8/sYxTkjZs2KCf/vSnqqur\nU0FBgVwul9ra2qKOv66uTg0NDbJtW5/5zGdmNaZ4mG2u0fa7FHl/Rdu/jY2NOnXqlKSxfTrd9tMd\nH7HmHU28jtdENvWYiNe5EmvGsR6j0tjFuaamZtJvTiKNX5IqKiq0b98 SdLatWtnNaZEMl/nerT9\nMle5ztU45/ualCymHgfxOm9iOQ5m2u Rxnny5EkNDQ1Jkqqrq0Prox0HyX5 h5vNuX43 1SSrl69\nOqn3x3Sfk2jXgFivVckoPPt4njexZhzt 0Yb53TZR7o SXceu8lktj /Y92P0X7 Rbuez c5Hc9r\nUrKYehzE67yZ7/M7Wk7T/Zw24fyWoucUy9/XpjsO5iLvaOJ1fZoNy55lF8yrV6 qsbFRH/7wh2P6\nBgvlzJkzOnXqlL785S/HeygzOn/ vHJycmY9dXE JXqu0SRT3tEk0nEQLlGOiWTKOFGzDEeu84/j\nYPaS6ThIhlzDJUrG0ZD9/EmU7Ml4biVKrtEkU97RcBzMXjLlTa73Lpnyjmam42DWcwuXLl2qZcuW\nzdnA5lpLS0vovtFEdunSJZ09e1bp6enxHoqkxM81mmTJO5pEOw7CJcoxkSwZJ3KW4ch1fnEcxCZZ\njoNkyTVcomQcDdnPn0TJnoznVqLkGk2y5B0Nx0FskiVvcp0byZJ3NLM5DmY9QwwAAAAAAABIBand\nfQ4AAAAAAACYgoIYAAAAAAAAjEJBDAAAAAAAAEahIAYAAAAAAACjUBADAAAAAACAUSiIAQAAAAAA\nwCj/PxKedS0Bp vjAAAAAElFTkSuQmCC\n",

       "text": [

        "<matplotlib.figure.Figure at 0xc199cf8>"

       ]

      },

      {

       "metadata": {},

       "output_type": "display_data",

       "png": "iVBORw0KGgoAAAANSUhEUgAAAmsAAADJCAYAAAB8B3UoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1 /AAAIABJREFUeJzt3X14VNd9J/DvnZFGo9ELo1f0hl4QAlkgGyRAhNqSbQlb\nsRGRiJ1CZNdPs5Bt46R90rTZbBLXSdpun/pJtm62 1Cv  wm3SSKvTGSkQwYmBrBgGBwZQxCiDeB\nEAi9C70LSTN3/yCMJaPRzEhz86379/t5Hj2eczk68xv7cPybc889RxBFUQQRERERKZJO7gCIiIiI\nyDUma0REREQKxmSNiIiISMGYrBEREREpGJM1IiIiIgVjskZERESkYEFSNNrZ2QmLxQK9Xo ioiIk\nJibOW398fBxvvvkmXnnlFSQlJQEAamtrYbfbAQCZmZnIzc2VIlQiIiIiRZMkWbPZbKisrAQA1NTU\noLy8fN769fX1ePbZZ2ddMxqN2LJlixThEREREamGJMmayWRyvjYYDPPWHRgYgMlkgtFonHXdbrdj\n7969EEURGRkZyMvLkyJUIiIiIkWTJFmbeShCcHDwvHWtViuef/55tLS0zLpeWlrqfH3gwAHfBkhE\nRESkEpIkaw/WmgGAIAjz1u3q6sK ffvQ3d2N9PR055q1mdwlfIcOHYJer19YsERERER ZDabkZ f\n73F9SZK18fFxAPdn2B68BoCmpibodDrk5OQ4r 3atQsA0NzcDLPZ7Lze1taGtLQ0AMDExMS876fX\n63mblIiIiFShsbHRq/qSJGsFBQWoqqqCKIooKSlxXm9oaIAgCLOSNQDo7  H1WqdNbPW1tYGm80G\nAFi7dq0UYRIREREpniDOXGCmUhaLhTNr5DGr1YrHH39c7jBIBdhXyBvsL SpxsZGFBcXe1yfm IS\nERERKRiTNdIcfvMlT7GvkDfYX0gqkqxZIyIiIvKFyclJ9Pb2yh2G10JCQhATE OTtpiskeZwXQl5\nin2FvMH 4nuTk5Po6upCcnIydDp13Qzs6 vDyMgIwsPDF92Wuj45ERERaUZvb68qEzUAiI6OxuDg\noE/aUt nJ1okfvMlT7GvkDfYX6ShxkQNuH8ogLuDATylzn8DRERERBrBZI00x2q1yh0CqQT7CnmD\n/YWkwmSNiIiISMH4NChpDteVkKfYV8gb7C/aIooi3njjDQQFBWFkZASvvvoqYmNjJXkvJmtERERE\nXqqrq8NTTz2FjRs3Ynh4GD// c/xgx/8QJL3kiRZ6 zshMVigV6vR1FRERITE etPz4 jjfffBOv\nvPKK8yB3b9sg8hT3QiJPsa QN9hf5PHMv37ik3YO7VrnVf2WlhaUlZUBACIiInwSgyuSJGs2mw2V\nlZUAgJqaGpSXl89bv76 Hs8   yi2iAiIiLt8TbJ8hWHwzGrbDQaJXsvSZI1k8nkfG0wGOatOzAw\nAJPJ9NCH9KYNIm/wmy 5c2foHj683IcBMRV3L/ZiS1Y0QoL4PBbNj2OLtkxPT88q 2pPtblIMvqI\nouh8HRwcPG9dV9PG3rRBROQrhy734c/2XcakXURmTCjO3BrCrt9dxKWeUblDIyIFWbNmDRoaGgAA\nw8PDmJqakuy9JJlZs9vtztfuMs2uri7s27cP3d3dSE9Pd65Z86YNYHbS92CvG5ZZnqu8Z88e5Obm\nKiYelpVT/ujaAP7XyRt4adkEygtyYbVasSUUaFmixw8/bMXrJRm4e/WsYuJlWVnlmfusKSGeQCgP\nDg468wKl dKXvoQ33ngDJ0 exNDQEL797W8/VGdwcBCtra0Pfb6Zdw89IYgzp7B8pLq6GhUVFRBF\nEXV1dc4FeE1NTdDpdMjJyXnod5qbm2E2m53/UVy1MReLxYK8vDxffwwKUFwETHO51DOK1z5sxT88\ntwIZ0aEAZveVj28N4R OtuEfvrgCy2NC5QyVFIpji 91dHQoNlnzhKv4GxsbUVxc7HE7ksysFRQU\noKqqCqIooqSkxHm9oaEBgiA8lKz19/fDarXOmllz1QbRYnEwpc btDvw0/qb NMvpDgTNWB2X1mf\nEok/2ZSMn1iu43 Wr0KYQS9HqKRgHFtIKpLMrPkbZ9aIaDH 95kOtN dwF XZLhddvHPJ9vROzqF\n1z2oS0SLw5m1 /h4E2nOzHUlRJd6RnHwUh  9QfLHkq 5uorXy9IRu/oFPZf6vNXiKQSHFtIKkzW\niEizJu0O/PTYTfzpF5IRbfLsqXODXoe/LErFLz6 g 6RSYkjJCJiskYaxHUl9MCvP lEcmQInlwe\nNeefu or6VGhqFgdhzetNxEAK0nIRzi2 F5ISAj6 vpU fdsbGwMer1v1rZK8oABEZHSXe4dw4GW\nPvzL9uwFrT37ymNLUV89gBM3BvF4hlmCCIkoJiYGIyMj6OjoUN0aUb1ej/j4eJ 0xWSNNIeP19P9\npz/b8J83zX/7c76 EqQT8CebUvCm9SY2pkbCoOeNCq3j2CKN8PBwhIeHyx2GrDi6EJHmvPNpFxIi\nDHg6c 7bn55alxyB9OhQVDf1 CgyIqKHMVkjzeE3X21rvzuB9y/04JubH3768/M86Stf35iEd891\nYWhi2lchkkpxbCGpMFkjIs0QRRE/P9GOynUJiA83 KTN5CVGPJ5uxt6mbp 0R0T0eUzWSHO4F5J2\nHbt F6OTdmzLifOovqd9Zcfapai92MvZNY3j2EJSYbJGRJow7RDxi4/vYNfGJOh1vn2qLDEihLNr\nRCQZSZ4G7ezshMVigV6vR1FRERITE13WPXnyJLq6ugAAqampyM/PBwDU1tbCbrcDADIzM5GbmytF\nqKRBXFeiTR9e7kN8eDDykiM9/h1v sqOx5biW 9fwh8 thShwTw3VIs4tpBUJEnWbDYbKisrAQA1\nNTUoLy93WXfz5s3O14cOHXK NhqN2LJlixThEZHGOEQR/ 9cN75TmCrZeyRGhuDRxAgcutyPL632\n7DYrEZEnJLkNajKZnK8NBveLeG/cuIGf/exnWLFihfOa3W7H3r178d5776GxsVGKMEmjuK5Eez6 \nNQRTsA5rloZ59Xve9pUvr4lD9YVu2B3q222dFo9jC0lFkpm1mcdCBAe7P28vPT0dr776Kt5//30s\nX74cAFBaWur88wMHDvg SCLSjPcv9OJLq Mk3wE9Z2kYIkKCcLp9EJvTeKoBEfmGJDNrD9aaAfB4\ncDQajYiIiJjzzzxJ Ig8xXUl2tI9MomWnlGX53/Ox9u IggCtq Jx3vnuUmuFnFsIalIkqyNj48D\nuD/D9uA1ADQ1NaG5uXlW3f7 fufrmTNybW1tztcTExNu33Pm9LPVamWZZZZZBgD8wtKIFcYJhATp\n/PJ uo4LaOsdwrW MUV8fpZZZlm5ZU8JogRH2Xd0dKC vh6iKKKkpMR5kOnbb78NQRCwa9cuZ93a\n2lrcu3cPALB /Xqkp6cDAI4dO Z8SnTt2rXIyspy X4WiwV5eXm /hgUoKxWnt nJa/WtGDXhmSs\nS5575n4 C 0rv2q8g/7xafzZHyzz ndJvTi2kKcaGxtRXFzscf0gKYJISkrCzp07H7q e/fuh66V\nlZXN2UZhYaHP4yIibbk9OIG 0Sk8mujfQ6CfXRWDP9nbgt0bk7iNBxEtGjfFJc3hN1/tOH7jLh7P\nMC94E9yF9pW4MAPWLA3H0da7C/p9UieOLSQVJmtEFLBs7UMoWLZElvd LjsG 1t6ZXlvIgosTNZI\ncxayuJPUZ TeNFr7xhd1C3QxfWV9SiQGxqdwtXdswW2QunBsIakwWSOigNR4exirl4Y7nwL1N71O\nQOmqWOxv6ZPl/YkocDBZI83huhJtOHNrCBuXeX4O6FwW21dKV0bjaOsAJqYdi2qH1IFjC0mFyRoR\nBRyHKOJM xA2LDJZW6zYMAOy401oaOODBkS0cEzWSHO4riTwtfaNw2TQIykyZFHt KKvlKyIxuEr\n/e4rkupxbCGpMFkjooBz5tYQNqTIO6v2wOZ0M1q6x9A3NiV3KESkUkzWSHO4riTw2Xx0C9QXfcUY\npMPmtCX46NrAotsiZePYQlJhskZEAWX43jSu94/j0QT/nlown5KsaBzhrVAiWiAma6Q5XFcS2Bpv\nDyM3IRwGH2zZ4au 8mhiOIZ/v 8bBS6OLSQVSc4G7ezshMVigV6vR1FRERITE13WPXnypPPA9tTU\nVOTn53vdBhHRA766BepLOkFA8YpoHLnaj6/HJMsdDhGpjCQzazabDZWVldixYwdOnz49b93Nmzej\noqICFRUV6Ov7bPNIb9og8gbXlQQuu0OErX3x 6s94Mu UrwiCkdbB AQRZ 1ScrCsYWkIkmyZjKZ\nnK8NBoPb jdu3MDPfvYzrFixYsFtEBFd7h1DVGgQEiIWt2WHFNKiQhEWrEdLN4 fIiLvSJKsiTO \nOQYHB7utn56ejldffRVnzpxZcBtEnuK6ksB16uYgClJ9d3C7r/vKExlmHL/Op0IDFccWkookyZrd\nbne FgTBo98xGo2IiIhYcBsz/5JYrVaWWXZZPn/ vKLiYdl35dM3BxE6cEMx8Xy /ESGGUdaunD8\nuDLiYZllluUte0oQRd8voKiurkZFRQVEUURdXR3KysoAAE1NTdDpdMjJyXHW7e/vR3R0NADggw8 \nwPPPPz9vG3OxWCzIy8vz9ccgIhXpHpnEqzWX8NuvroFe59mXRH8TRRH/6XcX8d2iNGTHh8kdDhHJ\npLGxEcXFxR7XD5IiiIKCAlRVVUEURZSUlDivNzQ0QBCEWcnaiRMncO/ePQDA vXr3bZBRDSXUzcH\nsSElQrGJGnD/LkFhhhnHr99lskZEHpNkZs3fOLNG3rBarXxqKwB9d/8VlD0ShycyzD5rU4q cq1v\nDD86fB3/9oc5Hi8TIXXg2EKe8nZmze2atc7OzkUFREQktb7RKVztHffZlh1SWh4dCr1OwDVukEtE\nHnKbrL311ls4fvw4AmACjggA90IKREdbB7A5bQlCfHBqwUxS9BVBELApNRKn2od83jbJi2MLScXt\nyPbiiy8iIyMD77zzDk6ePOmPmIiIvHK0dQBPZUbJHYbHClKX4PTNQbnDICKVcJus5eTkICUlBTt2\n7EB0dDT 5m/ BjabDUePHvVDeES t5DHpkm5WvvG0Tc6hbVJEe4re0mqvpKbEI7bg/fQPzYlSfsk\nD44tJBW3yVp3dzempqZw6NAhXLx4EX/ 53 OmJgYHDt2zB/xERHNa9/FHjz3SKyinwL9vCCdgPzk\nCNh4K5SIPOB264633noLq1atwtNPP43Y2FgAQGRkJB599FHJgyOSAteVBI6Re9M41noX//rCI5K0\nL2VfKUhdAuuNuyhdFSPZe5B/cWwhqbhN1srKyrB27dqHrq9cuVKSgIiIPLX/Uh82LItEtEl9R9Jt\nXBaJfz7ZjslpBww fjCCiAKL2xFirkQNwKyNbYnUhOtKAsPEtAN7z3fjDx9dKtl7SNlXIo1BWB4d\nik/vjEj2HuRfHFtIKl5/nTt37pwUcRAReeVASy y48OwPCZU7lAWbGNqJGztfCqUiObnNlmrr6 f\nVb5 /bpkwRD5A9eVqN/ktAPvnuvGV9clSPo UveVjSlLcOYWHzIIFBxbSCpuk7WBgYFZZR6PQkRy\nO3i5DytiQrEy1iR3KIuSEW3E5LSI24MTcodCRArm9UHuDw5dn09nZycsFgv0ej2KioqQmJjosu6F\nCxdw6dIlOBwObNq0CSkpKQCA2tpa2O12AEBmZiZyc3O9DZVoTjy/T92mHSLePdeFHzydIfl7Sd1X\nBEHA pRI2NqHULHEKNn7kH9wbCGpuEzW6urqMD4 jpaWFkxNfbZx42OPPea2UZvNhsrKSgBATU0N\nysvLXdbt6 vD9u3bAQD79 93JmtGoxFbtmzx7FMQkWb8 9V JEaE4JH4MLlD8YmNyyKx/1IvKtbE\nyx0KESmUy2Rt69atAIBjx46hsLDQq0ZNps9uTRgMhnnrumrbbrdj7969EEURGRkZyMvL8yoGIlf4\nzVe9HKKId8914xtfSPbL /mjr6xLjsBPj7VhYtoBI7fwUDWOLSQVt7dBvU3UAMw69D042LP9jw4f\nPoyNGzc6y6Wlpc7XBw4c8DoGIgo8J9sGYQzSYZ0ER0vJJcygx4oYEz7tGEZB6hK5wyEiBfL6a9zk\n5KTbOg/WmgGePZBQX1 PrKws5wkJn dJwjdzfxur1coyyy7Le/bsUVQ8LHte3nu G48a nHixAm/\nvN D11J/vjh7v/OpUCX9 2ZZmf2F5cApe0oQZ06DeWDfvn3Ytm3bvHWqq6tRUVEBURRRV1eHsrIy\nAEBTUxN0Ot2sDXWPHz OuLg4ZGdnz2qjra0NaWlpAO6vn3twW3YuFouFt0nJY1YrFwGr0fX cXz/\n4DX83x2rEeSnc0D91Veu94/jR4db8Yuv5PCJexXj2EKeamxsRHFxscf1g7x9A4fD4bZOQUEBqqqq\nIIoiSkpKnNcbGhogCIIzWevu7saJEyeQnZ2NlpYWDA8P4 WXXwZwP1mz2WwAXJ iQLQQHEzVqe5i\nL764KsZviRrgv76SHmXElF3ErcF7WGbmU6FqxbGFpOJyZu3HP/4xdu/ejaqqKqSmpjqvt7S04LXX\nXvNbgJ7gzBpRYJuYdqCyqgn/sj0bcWHzP7SkVv94/CbSoozYzqdCiQKetzNrLtesvf7660hKSkJm\nZiZefPFF5w/3OyO1W8h6AZLXqbZBrIoz T1R82df2bAsEmfaeZqBmnFsIam4fcBg06ZNs8pms1my\nYIiI5vLRtQE8lRkldxiSWpcUgebuUYxP2d1XJiJNcZusJSTMPnvvySeflCoWIr/guhJ1GZqYxqd3\nhrE5zf9fFP3ZV8IMeqyMNeHTOyN e0/yLY4tJBXuwEhEima9cRfrUyIRZtDLHYrkeCuUiObiNlnb\nv38/2tvbcePGDfzmN7/B1atX/REXkWS4rkRd5LwF6u  suH354R6uaMSKQTHFpKK22RtcnISycnJ\nOHv2LL761a isbHRH3EREWFwYhpXesewISVS7lD8Ij3KCLsoon3wntyhEJGCuE3WdDoddDodjMb7\ne/ EhIRIHhSRlLiuRD1s7YPIS46AQaYzM/3dVwRBwEbeClUtji0kFY9m1kZGRpzJGqfnichfGtqG\nsElj52U uBVKRPSA22QtKSkJNTU12LRpE5qamtDa2uqPuIgkw3Ul6jBpd CTjmFsXCbfLVA5 sq6\npAi09HALDzXi2EJScXvc1ObNm7F582YAQHZ2NjIzMyUPiojo3J0RpJmNMIcGyx2KX5kMeqyKM Fs\nxwi kKatWUUimptXZ4MGBQUhKMj9r3R2dsJisUCv16OoqAiJiYku6164cAGXLl2Cw HApk2bkJKS\n4nUbRN7guhJ1OHVzUPZkRa6 siHl/ro1uT8/eYdjC0nF7W1QURRx9epVNDc3o7m5Gb/85S/dNmqz\n2VBZWYkdO3bg9OnT89bt6 vD9u3b8cILL DcuXMLaoOIAosoimhoG8SmVG08Bfp5G5ZF4swtbuFB\nRPe5Tdaqq6sxPDzsfCpUp3P/VJbJZHK NhjmP8uvsLBw0W0QeYPrSpSvtX8cQToBqWajrHHI1VfS\nzEaIENF l1t4qAnHFpKK23uawcHBWLdunbOcnZ3tttGZ3waDgz1bb3L48GFs3LhxUW0QUWBouDmE\nTWlLIAiC3KHIQhAEbEiJxKn2QaRGyZuwEpH83E6TDQ8Pzyp3d3e7bdRu/ wpJk8G2/r6emRlZSE2\nNnbBbcz8RmO1Wllm2WX5wTWlxMPyw UjF245t yQM57HH39ctvd/IsOMY613FfHfg2Xl9xeW1Vn2\nlCC6WRTxd3/3d4iJiUFMTAwAoKWlBa 99tq8jVZXV6OiogKiKKKurg5lZWUAgKamJuh0OuTk5Djr\nHj9 HHFxcQ/N2LlqYy4WiwV5eXnzf1IiUoW 0Sl8fe9FvFOZiyCdNmfWAMDuELHzN034x7KVSF7C\nzciJAkljYyOKi4s9rh/krsIzzzyDDRs2OMtHjx5122hBQQGqqqogiiJKSkqc1xsaGiAIgjNZ6 7u\nxokTJ5CdnY2WlhYMDw/j5ZdfnrcNosWyWq18akvBTrcPIj85QhGJmpx9Ra8TULjcjKOtA6hclyBL\nDOQdji0kFbfJ2sxEDQCefPJJt40mJSVh586dD13fvXv3rHJ8fDy 973vedUGEQW2hrZBPL1CnoPb\nlebJ5VH4pxPtTNaINM6jA/du3LgBm82G6elpnmBAqsdvvso1Me3A c4RrFfIwe1y95WcpWEYm7Tj\nev 4rHGQZ TuLxS43CZrp06dQkdHB27duoWgoCDueUZEkmm8PYSsWBMiQtxO muCThDwVGYUjlzp\nlzsUIpKR22Sto6MDmzdvdu6vFhYWJnlQRFJayJM45B8NbYPYrKBd 5XQV55ZGQPL1X5MO7hBrtIp\nob9QYHKbrOn1en/EQUQaZ3eIOHWTRyx9XqrZiISIEJxpH5I7FCKSidtkbXx8HFNTUwCAqakpOBwO\nyYMikhLXlShTc/coYkzBSIhQzjYVSukrz66KwYeX  QOg9xQSn hwOPR1h2//vWv0dnZCbvdjtLS\nUn/ERUQao7RboEpSlGHG26dvY2BsClEmnuhCpDXzzqx1dnbi2LFjiIyMRHZ2NjZs2MA1a6R6XFei\nPKIo4mTboOJugSqlr5gMemxOWwLLVT5ooGRK6S8UeFzOrJ06dQpDQ0PYtm0bdDodHA4H6uvr0dPT\ng/z8fH/GSEQB7ubdCUzZHVgREyp3KIr17KoY/Nzaji/nxmv2zFQirXI5s3bjxg0888wzzqdAdTod\nnnrqKVy/ft1vwRFJgetKlOfBrJrSkhAl9ZU1S8Mw5RDR0jMmdyjkgpL6CwUWl8laaOjc33CDg7le\ngoh8q751AIUZPLVgPoIg4LnsGHxwsVfuUIjIz1wmayMjI jt7UVPT8 sn5GREZ8H4XA4YLfbfd4u\n0Vy4rkRZbg5MYHDCjjUJylsPq7S 8kxWNE62DWL43rTcodAclNZfKHC4XLPW09ODpqamOa 709nZ\nCYvFAr1ej6KiIiQmJrqse/DgQbS2tqK8vBxJSUnO67W1tc4ELjMzE7m5uW7fl4jU52jrAAqXm6FT\n2C1QJTKHBmPDskgcudKPijXxcodDRH7iMln75je/iaCgh//Yk3vyNpsNlZWVAICamhqUl5e7rFta\nWorm5uaHrhuNRmzZssXtexF5i tKlEMURRxtHcBfFaXJHcqclNhXns Owf84cQvlq MUt8ZP65TY\nXygwuEzW5krU5rs k8lkcr42GAwLCAuw2 3Yu3cvRFFERkYG8vLyFtQOESlXa/84puwisuNM7isT\nACA3IRwOUURT1yhyE8LlDoeI/MDtCQYLIYqfnWG30AcSSktLsX37dnz5y19GV1eXr0Ij4roSBTna\nehdFy82KnSFSYl8RBAHPPxLLBw0USIn9hQKDJMnazIcFfDEIe5LwzfxLYrVaWWbZZfn8 fOKiker\nZYco4ui1ASwZalNEPGoqR/Rdxun2IQxOTCsiHpZZZnlhZU8J4sxpMB prq5GRUUFRFFEXV0dysrK\nAABNTU3Q6XTIycmZVb 5uRlms3nWAwZtbW1IS7u/jqWurg5bt251 X4Wi4W3SYlU5pOOYbx16jb2\nVKxS7Myakr1R34blUUa88OhSuUMhIi81NjaiuLjY4/pBUgRRUFCAqqoqiKKIkpIS5/WGhgYIgjAr\nWTty5AhaW1thNBqRnp6OwsJCAPeTNZvNBgBYu3atFGESkYw vNSHZ1dGM1FboOezY/DT ps80YBI\nAyRJ1pKSkrBz586Hru/evfuhazOTuZkeJG1Evma1WvnUlsxGJ 043T6EP/1CityhzEvJfSUnPgzB\negFn74xgXVKE3OEQlN1fSN0kWbNGRDSfj64NYF1SBJYYJfm qAmCIGArHzQg0gQma6Q5/OYrv0OX\n 1C6KlruMNxSel8pXhGNTzqG0T0yKXcoBOX3F1IvJmtE5FfX 8fRMzqF/ORIuUNRvTCDHs9kRaO6\nqVvuUIhIQkzWSHMW8tg0 U7dxV58cVUM9DrlL4pXQ1 pWBOPQ1f6McLzQmWnhv5C6sRkjYj8ZmzS\njqOtA3guO0buUAJGfLgBG1Iisf9Sn9yhEJFEmKyR5nBdiXyOXO3HY4kRiA1b2DF0/qaWvvJCbjxq\nmnowZXfIHYqmqaW/kPowWSMivxBFEbUXe1GWEyt3KAFnRawJqVFGHL7SL3coRCQBJmukOVxXIo/z\nnaNwOESsTVTP4eNq6isv5yWg6mwXZ9dkpKb QurCZI2I/KL2Yg/KcuK4275EVi8NxzJzCD68zNk1\nokCjiGTN4XDMOvydSEpcV J/fWNT I9bw9iSpfy91WZSW195OS8RVWc7McnZNVmorb QekiyfXhn\nZycsFgv0ej2KioqQmJjosu7BgwfR2tqK8vLyWQe5e9MGESnbgUt9KFpuRphBL3coAe2R DBkRIfi\n4KU bMuJkzscIvIRSWbWbDYbKisrsWPHDpw fXreuqWlpXjyyScX1QaRN7iuxL/sDhH7L/Zi6yPq\ne7BAjX3l5bwE/PZsFyanObvmb2rsL6QOkiRrJpPJ dpgWNgj r5og4jk19A2iKURBmTGmNxXpkVb\nFReGFbGh KCFZ4YSBQpJkjVRFJ2vg4ODZWuDaC5cV JftRd7sE2l23Wota 8nJeId8514R5n1/xK\nrf2FlE SZG3mwwILffLLF20Qkbxu3p3AjYEJ/EG6We5QNCUr1oTsuDDOrhEFCEkeMBgfHwdwf3bs\nwWsAaGpqgk6nQ05OzoLbcMVqtTq/1TxYN8Ayy3OV9 zZg9zcXMXEE8jlDy72YrVpDLaGk4qIx9vy\nzDVISojHm/LLeXn4wYfXENV/GcE6 ePRQlnN/YVl/5ZnLvXyhCDOvN/oIx0dHaivr4coiigpKUF8\nfDwA4O2334YgCNi1a5ez7pEjR9Da2gqj0Yj09HQUFhbO28ZcLBYL8vLyfP0xKEBZrZ8l9iSd8Sk7\nXvrtBeyr27zjAAALsUlEQVSpyEZ8uDrXnaq9r/zkSCtylobjhVzX4yf5jtr7C/lPY2MjiouLPa4v\nSbLmb0zWiJRnf0svTrcP4cdblssdimZd7x/H9w5cxS  koPQYG6bQqQU3iZritgUl4gCiyiK2Nfc\ngzIVbtcRSDKiQ5GbEI7ai1y7RqRmTNZIc2auKyFpfHpnBNMOID85Qu5QFiUQ spLeQn43blujE/x\nlBipBUJ/IWViskZEPld9oQflq3kOqBKkR4XisaRw1DZzdo1IrZiskeZwAbC0OobuoblrFCUqOwd0\nLoHSV15al4DfnefsmtQCpb Q8jBZIyKfev9CD0pXRsMYxOFFKdKiQrE2KRzvN/fIHQoRLQBHU9Ic\nriuRzuikHUeu9qMsQA4RD6S 8tK6RLx3vgdjk5xdk0og9RdSFiZrROQzH17uQ15ShGr3VQtkqVFG\n5CVHcHaNSIWYrJHmcF2JNCbtDvzuXDdefHSp3KH4TKD1lcp1Cdjb1INRzq5JItD6CykHkzUi8onD\nV/qRHm3EyjjvjlEh/0k1G5GfHIH3L3B2jUhNmKyR5nBdie9NO0T89mwXKtcmyB2KTwViX6lcl4Dq\nCz0YmpiWO5SAE4j9hZSByRoRLdqRK/1IiDBgdUK43KGQG8vMRjyRbsavPumUOxQi8lCQFI12dnbC\nYrFAr9ejqKgIiYmJXtetra2F3X5/XUVmZiZyc3OlCJU0iOtKfGti2oF/ 487 GFxhtyh Fyg9pU/\nyk/A7vdasDU7FqlRRrnDCRiB2l9IfpIkazabDZWVlQCAmpoalJeXe13XaDRiy5YtUoRHRD703vlu\nrF4ahpylYXKHQh4yhwZjx2NLsefULfy30kyeNEGkcJLcBjWZPltgbDDM/wi/q7p2ux179 7Fe  9\nh8bGRt8HSZrFdSW c2foHqqbuvG1DUlyhyKJQO4rX1odh4HxKViuDsgdSsAI5P5C8pJkZk0URefr\n4ODgBdUtLS11vj5w4IAPoyMiXxBFEW9ab Irjy1FYmSI3OGQl4J0Ar5TmIbvH7yGdckRiDHNP1YT\nkXwkmVl7sNYMgNvpdU/qukv4gNnfaKxWK8ssuyw/uKaUeNRarrnQg9FJBxIGrygiHinKjz/ uKLi\n8XU5K9aE3LAxvL7vrPOLs5LiU1s50PsLy9L8/8gTgjhzastHqqurUVFRAVEUUVdXh7KyMgBAU1MT\ndDodcnJy3NZta2tDWloaAKCurg5bt251 X4WiwV5eXm /hhE5EJz1yheP9yKn29byVk1lZuyO/AX\ndVdQlGHGCwG0oTGRkjU2NqK4uNjj kFSBFFQUICqqiqIooiSkhLn9YaGBgiCMCtZc1W3ra0NNpsN\nALB27VopwiSNslqtfGprEW4NTuAnR1rxncLUgE/UtNBXgvU6/PDpDHzr/UtYFR GXG6/smBa6C8k\nD0mStaSkJOzcufOh67t37/a4bmFhoRShEdEitN dwPcPXsMr65OwKXWJ3OGQjyyNMOCvitLwt5br\n O9bs5C8hNt5ECkJN8UlzeE334X5tGMY36m7gq uS8AXV8XIHY5faKmvbFgWiVfyE/H9g9cwMDYl\ndziqpKX Qv7FZI2I3Dp0uQ9/  838F fStdMoqZFz2XHYktWNL574Cr6mbARKQaTNdKchTyJo1UO\nUcT/ bgDv/qkEz99fgXWJUfIHZJfabGvvJSXiKLlUfjLD66gZ3RS7nBURYv9hfyDyRoRzWly2oG/\n/ gGznYM45 2rURaVKjcIZGfvLQuAaWrYvAXtVfQ2jcudzhEmsdkjTSH60rcGxifwnf3XwVE4I3n\nshAVqs0NU7XcV77y6FL88fpE/JcDV3Hixl25w1EFLfcXkpYkT4MSkXpd6xvDjw5fx9MrovBKfiJ0\nPDdSs55eEY3kJSH48ZHrON85gq tT4IhiN/xifyNf tIc7iuxLVjrQP43oFr2LUxCX 8PknziRr7\nCrAqLgx7KrLRMzqFV2suoaV7VO6QFIv9haTCmTUiwqTdgV98fAfHrg/g70szsSLWJHdIpCBLjEH4\n4dPp OjaAH50pBXrkiLwtQ1JiAszyB0akSZIctyUv/G4KaKFu9w7hn88fhPx4QZ8 /FlMGt0fRp5\nZmzSjnfOdaG2uRdPZJjx5dx4pJq5iS6RNxRx3BQRKd/1/nG8e64Ln3QM42vrk7AlKxqCxm97knsm\ngx5/vD4JFavjsK 5F9 pu4L0KCO2ZEXjiQwzQoP1codIFHCYrJHmaPX8PlEUcb1/Ah/fGsLJtkF0\njUzi Udi8a3Ny2Ay8H wc9FqX/GEOTQYf5SfiB1rl LUzUEcvtyPPaduY0NKBL6QZsbGZZEI01i/\nYn8hqUiSrHV2dsJisUCv16OoqAiJiYle1/WmDSKaTRRFdI9M4XLv2P2fnlFc6R1HpDEI ckR2Ll2\nKfJTIhGk40waLY5Br0NhRhQKM6IwMDaFkzcHYbnajzetN5ETH4aC1CVYlxSOVLORM7dECyTJmrV9\n /Zh27ZtAICamhqUl5d7XdebNrhmjbRMFEX0jU3hSu84rvSO4VLP/QRNLwAr40xYGWvCyjgTsmJN\nmt0vjfxvbNKOj28P4Uz7EM52jGDK7sBjSRFYvTQMWbEmZESHwshtQEijFLFmzWT67Ekyg2H p4Vc\n1fWmDSK1E0UR9 wiJqbsmJh2YGLagfEpB8Ym7ff/OfXZP8emHBifsmP4nh23B /h1uAEgvU6rIgJ\nxco4E57LjsG345YhxhTMmQySjcmgd864AcCd4Xs4e3sYLT1jOHipD 13J5AQGYJlS0KQGBGCxMgQ\nJEQYsMQYhMiQIEQa9TAG6diHiSBRsjZzsi44eP5v8q7qetOGUv371X58dG1g1rW5pjFdzW2Kc9b2\n7Pc9f5 538ObOBccj6vPt6jP4/59hoaGERkZ4XGcc7Y55 8 fNFVjHaH6EzKJqYdmJhyIEgvwBik\nc/6YDHqYgnUIDZ7xz99fizEFI9ygR/nqEKQsCUFECJefSoFrkHwnMSIEidkh GL2/fKk3YGbAxO4\nPXQPd4bv4UrvGI5fv4uhe9MYmrj/4wAQotchSCcgWH//Ry8IEAQBAgBBgPOfgOAsw/lnwqw6wu//\n1Fnv99cerAb47PeFz7UN6AQB7tLG/v5 REdH  TfFz0sWK/DX5dkyB2GLCQZ4e12u/O1u29Frup6\n04bZbEZjY6O3YUrODKAiTu4o6GE6AErc2FME4HBfzQHg3v2f8WHgyi2Jw9Iwk8mkyLElkET8/mdl\nGICwuWp48HdCKeL0AAbljiKgNTYOuK kAmaz2av6kiRr4 P3D/4VRdH5GgCampqg0 mQk5Pjtq6r\n63PJz8/3WexERERESiJJslZQUICqqiqIooiSkhLn9YaGBgiCMCtZc1XX1XUiIiIiLQmIEwyIiIiI\nAhWfmyYiIiJSMNUnaw6HY9bDCERERERKsthcRdXP x88eBCtra0oLy9HUlKS8zpPP6D51NbWOv/S\nZGZmIjc3V aISGk4hpA3OKbQfObKVbwdY1SdrJWWlqK5ufmh6zabDZWVlQDcn35A2mM0GrFlyxa5\nwyAF4xhC3uCYQvOZK1fxdoxRdbLmCk8/oPnY7Xbs3bsXoigiIyODR5XRQziGkDc4ppC3vB1jAjJZ\nC4TTD0g6paWlztcHDhyQMRJSKo4h5A2OKeQtb8cYVSRrvb29ePfdd2dd27p1K1JTU es783pBxS4\nPOk3/B8xzYVjCC0UxxTyhLdjjCqStdjYWHzjG9/wuL43px9Q4HLVb9ra2pCWlgYAmJiY8HdYpAIc\nQ8gbHFPIW96OMareFPfIkSNobW2F0WhEeno6CgsLAQAdHR2or693nn4QHx8vc6SkJMeOHUNXVxcA\nYO3atcjKypI5IlIajiHkDY4pNJ 5chVvxxhVJ2tEREREgU71m ISERERBTIma0REREQKxmSNiIiI\nSMGYrBEREREpGJM1IiIiIgVjskZERESkYEzWiIiIiBSMyRoRERGRgv1/EaccxejShjkAAAAASUVO\nRK5CYII=\n",

       "text": [

        "<matplotlib.figure.Figure at 0xc199438>"

       ]

      }

     ],

     "prompt_number": 79

    },

    {

     "cell_type": "code",

     "collapsed": false,

     "input": [

      "#???\n",

      "# \u043a\u0430\u0442\u0435\u0433\u043e\u0440\u043d\u044b\u0439 \u0442\u0438\u043f (new)\n",

      "tmp = dict([('A',[1, 1, 2, 2]), ('B',[1., 2., 1., 2.])]) # \u0435\u0449\u0451 \u043e\u0434\u0438\u043d \u0441\u043f\u043e\u0441\u043e\u0431\n",

      "df = pd.DataFrame(tmp)\n",

      "df['A'] = df['A'].astype('category') # \u043f\u0440\u0435\u043e\u0431\u0440\u0430\u0437\u043e\u0432\u0430\u043d\u0438\u0435 \u0432 \u043a\u0430\u0442\u0435\u0433\u043e\u0440\u043d\u044b\u0439 \u0442\u0438\u043f\n",

      "print df['A'].cat.categories\n",

      "df"

     ],

     "language": "python",

     "metadata": {},

     "outputs": []

    }

   ],

   "metadata": {}

  }

 ]

}