N13 Pandas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# N13 Анализ данных. Pandas\n",
    "\n",
    "Автор: Шабанов Павел Александрович\n",
    "\n",
    "Email: pa.shabanov@gmail.com\n",
    "\n",
    "URL: [Заметки по программированию в науках о Земле](http://progeoru.blogspot.ru/)\n",
    "\n",
    "Дата последнего обновления: 23.03.2017\n",
    "\n",
    "### План\n",
    "\n",
    "1. Модуль pandas\n",
    "\n",
    "2. pandas DataFrame\n",
    "    + подписи осей таблицы DataFrame;\n",
    "    + индексация DataFrame.\n",
    "    \n",
    "3. Чтение данных с помощью pandas DataFrame\n",
    "    + pd.read_csv();\n",
    "    + pd.read_wfw()\n",
    "    + pd.read_excel()\n",
    "\n",
    "4. Запись данных с помощью pandas DataFrame\n",
    "    + Запись данных в текстовые файлы;\n",
    "    + Запись данных в файлы MS Excel;\n",
    "### Цель: \n",
    "\n",
    "+ получить базовые знания и навыки для работы с модулем анализа данных pandas.\n",
    "\n",
    "### Ссылки:\n",
    "\n",
    "+ [Официальный сайт проекта Pandas](http://pandas.pydata.org/)\n",
    "\n",
    "+ [Tutorials from pandas.pydata](http://pandas.pydata.org/pandas-docs/stable/tutorials.html);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Модуль Pandas\n",
    "\n",
    "[Источник](https://ru.wikipedia.org/wiki/Pandas)\n",
    "\n",
    "**Pandas** — библиотека на языке Python для обработки и анализа данных. \n",
    "\n",
    "Работа pandas с данными строится поверх библиотеки NumPy, являющейся инструментом более низкого уровня. Pandas предоставляет специальные структуры данных и операции для манипулирования числовыми таблицами и временными рядами. Название библиотеки происходит от эконометрического термина «панельные данные», используемого для описания многомерных структурированных наборов информации. Pandas распространяется под новой лицензией BSD.\n",
    "\n",
    "Несмотря на то, что Python в течение долгого времени успешно применяется для подготовки данных, его использование для анализа и моделирования оставалось затруднённым. Часто требуется выполнять часть задач при помощи инструментов, отражающих специфику предметной области, таких как язык R. **Pandas** восполняет описанный недостаток, позволяя полностью проводить все этапы анализа информации на Python.\n",
    "\n",
    "Сейчас в pandas реализована лишь малая часть возможностей моделирования, однако работа в данном направлении находится в приоритете. На данный момент разработчики рекомендуют обратить внимание на такие пакеты как **statsmodels** и **scikit-learn** (Машинное обучение на python).\n",
    "\n",
    "Де-факто стандарт импортирования pandas выглядит так:\n",
    "\n",
    "> import pandas as pd\n",
    "\n",
    "**df** очень часто используется в качестве имени переменной, хранящей ссылку на оюбъект pandas DataFrame.\n",
    "\n",
    "## DataFrame\n",
    "\n",
    "Центральным объектом библиотеки (модуля) pandas является объект **DataFrame**. Дословный перевод даёт \"блок данных\", но мне нравится термин **pandas-Таблица**.\n",
    "\n",
    "По сути DataFrame - это двумерный псевдомассив, где все строки и столбцы могут иметь нечисловую нумерацию. Т.е. вместо принципа индексирования массивов из модуля numpy по двум осям (*arr[i, j]*) pandas-Таблица предлагает дополнительный функционал в виде обращения к строкам/столбцам/элементам по ключам (как в словарях)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 20 entries, 0 to 19\n",
      "Data columns (total 14 columns):\n",
      "0     20 non-null float64\n",
      "1     20 non-null float64\n",
      "2     20 non-null float64\n",
      "3     20 non-null float64\n",
      "4     20 non-null float64\n",
      "5     20 non-null float64\n",
      "6     20 non-null float64\n",
      "7     20 non-null float64\n",
      "8     20 non-null float64\n",
      "9     20 non-null float64\n",
      "10    20 non-null float64\n",
      "11    20 non-null float64\n",
      "12    20 non-null float64\n",
      "13    20 non-null float64\n",
      "dtypes: float64(14)\n",
      "memory usage: 2.3 KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "shape1 = (20, 14)\n",
    "arr = np.random.random(shape1)\n",
    "\n",
    "# Создание dataFrame на основе существующего массива\n",
    "df = pd.DataFrame(arr)   # преобразование массива arr к dataFrame\n",
    "\n",
    "print df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Подписи осей таблицы DataFrame\n",
    "\n",
    "**DataFrame** - это всегда ДВУМЕРНЫЙ псевдомассив. Почему псевдо? Потому что массив содержит только данные одного типа, а DataFrame может содержать данные разных типов по строкам/столбцам.\n",
    "\n",
    "У объекта DataFrame (df) есть атрибуты, хранящие инфорамцию об именовании строк и столбцов:\n",
    "\n",
    "+ **df.index** - это последовательность уникальных значений для строк;\n",
    "\n",
    "+ **df.columns** -  это последовательность уникальных значений для столбцов.\n",
    "\n",
    "Именовать или переименовать подписи строк/столбцов (row labels/column labels) можно либо при создании DataFrame, либо после, используя методы."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         10        20        30\n",
      "a  0.193243  0.449483  0.527647\n",
      "b  0.598115  0.485275  0.436184\n",
      "c  0.910407  0.004737  0.277489\n",
      "d  0.065928  0.212776  0.950568\n",
      "e  0.033037  0.480585  0.280643\n",
      "***********************************\n",
      "Index([u'a', u'b', u'c', u'd', u'e'], dtype='object')\n",
      "Int64Index([10, 20, 30], dtype='int64')\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "shape1 = (5, 3)\n",
    "arr = np.random.random(shape1)\n",
    "xlabels = ['a', 'b', 'c', 'd', 'e']   # \"строковые\" подписи строк DataFrame\n",
    "\n",
    "df = pd.DataFrame(arr, \n",
    "                  index=xlabels, columns=range(10, 40, 10))   # x числовые подписи столбцов DataFrame\n",
    "\n",
    "print df\n",
    "print '*'*35\n",
    "print df.index\n",
    "print df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Заменить** подписи строк/стобцов DataFrame можно простым присваиванием атрибутам index и columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before\n",
      "Index([u'a', u'b', u'c', u'd', u'e'], dtype='object')\n",
      "Int64Index([10, 20, 30], dtype='int64')\n",
      "After\n",
      "Int64Index([-5, -4, -3, -2, -1], dtype='int64')\n",
      "Index([u'A', u'B', u'C'], dtype='object')\n",
      "           A         B         C\n",
      "-5  0.193243  0.449483  0.527647\n",
      "-4  0.598115  0.485275  0.436184\n",
      "-3  0.910407  0.004737  0.277489\n",
      "-2  0.065928  0.212776  0.950568\n",
      "-1  0.033037  0.480585  0.280643\n"
     ]
    }
   ],
   "source": [
    "print 'Before'\n",
    "print df.index\n",
    "print df.columns\n",
    "\n",
    "df.index = range(-5, 0, 1)\n",
    "df.columns = ['A', 'B', 'C']\n",
    "\n",
    "print 'After'\n",
    "print df.index\n",
    "print df.columns\n",
    "print df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Индексация DataFrame\n",
    "\n",
    "Обращаться к элементам и осям pandas-таблицы можно по-разному. Существует аж 3 способа это сделать:\n",
    "\n",
    "1. Через индексы как в numpy-массивах. Синтаксис здесь отличается - необходимо использовать метод **iloc**\n",
    "\n",
    "2. Через подписи как в словарях. Для каждой оси - свой ключ. Для вызова этого метода необходимо использовать **loc**\n",
    "\n",
    "3. И через индексы и через подписи (смешанный тип). Обращение к одной из осей идёт по ключу, а к другой - через индекс. Для вызова этого метода необходимо использовать **ix**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=============iloc=============\n",
      "\n",
      "I      38.516166\n",
      "II     56.773774\n",
      "III    48.646303\n",
      "Name: b, dtype: float64\n",
      "140.215212856\n",
      "\n",
      "=============loc==============\n",
      "\n",
      "a    62.339347\n",
      "b    56.773774\n",
      "c    70.821103\n",
      "d    62.456930\n",
      "e    16.031861\n",
      "Name: II, dtype: float64\n",
      "62.4569303208\n",
      "\n",
      "==============ix==============\n",
      "\n",
      "a    62.339347\n",
      "b    56.773774\n",
      "c    70.821103\n",
      "d    62.456930\n",
      "e    16.031861\n",
      "Name: II, dtype: float64\n",
      "112.307090541\n",
      "88.2702966966\n"
     ]
    }
   ],
   "source": [
    "# Типы индексаций в pandas\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "shape1 = (5, 3)\n",
    "arr = np.random.random(shape1)*150.\n",
    "xlabels = ['a', 'b', 'c', 'd', 'e']   # \"строковые\" подписи строк DataFrame\n",
    "ylabels = ['I', 'II', 'III']\n",
    "df = pd.DataFrame(arr, index=xlabels, columns=ylabels)   # x числовые подписи столбцов DataFrame\n",
    "\n",
    "print '\\n{:=^30}\\n'.format('iloc')\n",
    "\n",
    "# iloc - Через индексацию\n",
    "print df.iloc[1, :]\n",
    "print df.iloc[2, 2]\n",
    "\n",
    "print '\\n{:=^30}\\n'.format('loc')\n",
    "\n",
    "# loc - Через подписи\n",
    "cols = df.columns[:]\n",
    "print df.loc[:, cols[1]]\n",
    "print df.loc['d', 'II']\n",
    "\n",
    "print '\\n{:=^30}\\n'.format('ix')\n",
    "\n",
    "# ix - смешанный тип\n",
    "cols = df.columns[:]\n",
    "inx = df.index[:]\n",
    "print df.ix[:, 1]\n",
    "print df.ix['a', 2]\n",
    "print df.ix[3, 'I']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Несмотря на наличие подписей, таблица в любом случае продолжает поддерживать двумерную индексацию элементов. Просто она скрыта от пользователя. Индексация начинается от 0. Поддерживается обратная индексация через отрицательные индексы.\n",
    "\n",
    "Векторные срезы поддерживаются, они должны быть однотипными. Смешивать индексы и ключи при \"векторном срезании\" не разрешается."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            I          II        III\n",
      "b  147.118574   49.070428  56.800492\n",
      "c  114.554625   88.076476  55.433144\n",
      "d    8.456217  114.226673  76.609781\n",
      "            I          II        III\n",
      "b  147.118574   49.070428  56.800492\n",
      "c  114.554625   88.076476  55.433144\n",
      "d    8.456217  114.226673  76.609781\n",
      "            I          II        III\n",
      "b  147.118574   49.070428  56.800492\n",
      "c  114.554625   88.076476  55.433144\n",
      "d    8.456217  114.226673  76.609781\n",
      "\n",
      "Векоторные срезы\n",
      "\n",
      "            I         II        III\n",
      "a   15.619269   6.499394  53.007220\n",
      "c  114.554625  88.076476  55.433144\n",
      "e   52.458614  75.153235  68.268671\n",
      "            I        III\n",
      "a   15.619269  53.007220\n",
      "b  147.118574  56.800492\n",
      "e   52.458614  68.268671\n",
      "            I        III\n",
      "e   52.458614  68.268671\n",
      "c  114.554625  55.433144\n",
      "d    8.456217  76.609781\n"
     ]
    }
   ],
   "source": [
    "# Один результат разными методами\n",
    "\n",
    "print df.iloc[-4:-1, :]\n",
    "print df.ix[1:-1, :]\n",
    "print df.loc[['b', 'c','d'], :]\n",
    "\n",
    "# Векторные срезы\n",
    "print '\\n{:=^30}\\n'.format('Векоторные срезы')\n",
    "vecs1 = [0, 2, 4]\n",
    "print df.iloc[vecs1, :]\n",
    "\n",
    "vecs2 = ['a', 'b', 'e']\n",
    "vecs3 = ['I', 'III']\n",
    "print df.loc[vecs2, vecs3]\n",
    "\n",
    "vecs4 = [-1, -3, -2]\n",
    "vecs5 = ['I', 'III']\n",
    "print df.ix[vecs4, vecs5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Помимо вышеперечисленных способов, DataFrame поддерживает ещё и удобное обращение к столбцам. Это можно сделать через подпись столбца в квадратных скобках. Автоматически будут выбраны все строки из текущего DataFrame.\n",
    "\n",
    "Такой метод возвращает ещё один ключевой объект модуля pandas - **Series** или ряд (близко к Time Series). Это одномерная последоветльность, индексированная подписями строк и поддерживающая обычную числовую индексацию."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   E95  M10\n",
      "0    I    2\n",
      "1   II    7\n",
      "2    V   17\n",
      "3  III   12\n",
      "4   IV   17\n",
      "\n",
      "========pandas Series=========\n",
      "\n",
      "0     2\n",
      "1     7\n",
      "2    17\n",
      "3    12\n",
      "4    17\n",
      "Name: M10, dtype: int64\n",
      "<class 'pandas.core.series.Series'>\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "dic = {'I':2, \"II\":7, 'III':12, \"IV\":17, 'V':17}\n",
    "df = pd.DataFrame(dic.items(), columns=['E95', 'M10'])\n",
    "\n",
    "print df\n",
    "print '\\n{:=^30}\\n'.format('pandas Series')\n",
    "\n",
    "ts = df['M10']\n",
    "print ts\n",
    "print type(ts)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Чтение данных с помощью pandas DataFrame\n",
    "\n",
    "### pd.read_csv()\n",
    "\n",
    "Pandas предоставляет наиболее широкие возможности для чтения текстовых файлов с числовой информацией.\n",
    "\n",
    "Вместо **numpy.genfromtxt()** здесь используется **pandas.read_csv()**. Объект, который возвращает эта функция, - либо DataFrame, либо одномерный вариант Series."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   YYYY-MM-DD  hh:mm   Original   Calculated   Flag\n",
      "0  1931-01-01  --:--     6570.0       6570.0   -999\n",
      "1  1931-01-02  --:--     6480.0       6480.0   -999\n",
      "2  1931-01-03  --:--     6320.0       6320.0   -999\n",
      "3  1931-01-04  --:--     5930.0       5930.0   -999\n",
      "4  1931-01-05  --:--     5750.0       5750.0   -999\n",
      "5  1931-01-06  --:--     5680.0       5680.0   -999\n",
      "6  1931-01-07  --:--     5530.0       5530.0   -999\n",
      "7  1931-01-08  --:--     5500.0       5500.0   -999\n",
      "8  1931-01-09  --:--     5530.0       5530.0   -999\n",
      "9  1931-01-10  --:--     5510.0       5510.0   -999\n",
      "Columns: Index([u'YYYY-MM-DD', u'hh:mm', u' Original', u' Calculated', u' Flag'], dtype='object')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0xbc2b048>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZ4AAAEKCAYAAAAiizNaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngNcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc611588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "cwd = os.getcwd()\n",
    "\n",
    "fileName = '6742900.day'   # Данные по Дунаю\n",
    "iPath = './dataset/dat6/{}'.format(fileName)\n",
    "\n",
    "df = pd.read_csv(iPath, skiprows=40, sep=';')\n",
    "print df.head(10)\n",
    "\n",
    "print 'Columns:', df.columns\n",
    "\n",
    "ts = df[' Original'][:]\n",
    "f = [pd.to_datetime(v) for v in df['YYYY-MM-DD']]\n",
    "ts.index = f[:]\n",
    "adf = ts.resample('A').mean()\n",
    "ax = adf.plot(grid=True)\n",
    "ax.set_xlabel('MEAN DAILY DISCHARGE')\n",
    "ax.set_ylabel('m**3/s')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### pd.read_wfw()\n",
    "\n",
    "Для некоторых текстовых файлов, которые были созданы с т.н. фиксированной длиной записи, обычный метод сепарации данных через пробелы, табуляции или другие разделители (например, точку с запятой) может не сработать. \n",
    "\n",
    "Если данные \"слипаются\", когда между ними нет свободного места, то pandas предлагает другой метод - **pd.read_fwf()**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      0  1   2      3    4        5       6       7\n",
      "0  1948  1   1  11.64  0.0   0.8330  6.6056 -1.3722\n",
      "1  1948  1   2   5.53  0.0   1.1864  6.5333  0.5944\n",
      "2  1948  1   3   6.64  0.0   1.1864  5.1667 -1.5500\n",
      "3  1948  1   4   6.45  0.0   1.1497  4.5556 -0.7556\n",
      "4  1948  1   5  32.45  0.0   1.0487  2.8778 -3.0389\n",
      "5  1948  1   6  44.98  0.0   2.1686  4.0556 -0.5333\n",
      "6  1948  1   7   8.73  0.0  10.4184  7.4667  0.8056\n",
      "7  1948  1   8   1.90  0.0   6.8844  6.1389 -0.9056\n",
      "8  1948  1   9   4.46  0.0   4.5207  2.7778 -4.6222\n",
      "9  1948  1  10   5.14  0.0   3.5110  3.2278 -3.9778\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "widths = [4, 2, 2, 10, 10, 10, 10, 10]   #\n",
    "fileName = '14101500.dly'   # Данные MOPEX\n",
    "iPath = './dataset/dat6/{}'.format(fileName)\n",
    "wf = pd.read_fwf(iPath, widths=widths, header=None)\n",
    "print wf.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### pd.read_excel()\n",
    "\n",
    "Таблицы MS Excel - это один из самых органичных форматов для агреграции их с помощью pandas DataFrame. Чтение файлов форматов xls/xlsx становится с pandas настоящим удовольствием - **pd.read_excel()**.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Year   Coal   Oil  Natural Gas  Total\n",
      "0  1900  515.0  16.0          3.0  534.0\n",
      "1  1901  531.0  18.0          4.0  553.0\n",
      "2  1902  543.0  19.0          4.0  566.0\n",
      "3  1903  593.0  20.0          4.0  617.0\n",
      "4  1904  597.0  23.0          4.0  624.0\n",
      "5  1905  636.0  23.0          5.0  664.0\n",
      "6  1906  680.0  23.0          5.0  708.0\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "fileName = 'book_tgt_climate_2.xlsx'\n",
    "filePath = './dataset/dat6'\n",
    "iPath = '{}/{}'.format(filePath, fileName)\n",
    "\n",
    "# Определяем подписи для стобцов DataFrame\n",
    "cols = pd.read_excel(iPath, skiprows=2, usecols=[0, 1, 2, 3, 4], skip_footer=999)\n",
    "\n",
    "ex = pd.read_excel(iPath, skiprows=4, parse_cols='A:E', skip_footer=8,\n",
    "                   names=cols.columns)\n",
    "\n",
    "print ex.head(7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[u'Year' u'Coal' u'Oil' u'Natural Gas' u'Total']\n"
     ]
    }
   ],
   "source": [
    "cols = pd.read_excel(iPath, skiprows=2, usecols=[0, 1, 2, 3, 4], skip_footer=999)\n",
    "print cols.columns.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Запись данных с помощью pandas DataFrame\n",
    "\n",
    "С pandas удобно не только считывать данные, но и записывать их в файлы! Функция сохранения здесь реализованан в виде методов DataFrame **to_csv** и **to_excel**.\n",
    "\n",
    "### df.to_csv()\n",
    "\n",
    "Запись в текстовый файл похожа на чтение данных: нужны имя выходоного файла и путь к нему, указать тип разделителя, определиться с header и index (подписями столбцов/строк). Обратите внимание на параметр **float_format**, который определяет представление чисел с плавающей запятой в файле (часто - число знаков после запятой).\n",
    "\n",
    "Если таблица имела подписи (не индексы), то по умолчанию они будут также записаны в выходной файл."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         0         1         2         3         4         5         6   \\\n",
      "0  0.406673  0.849191  0.969383  0.529111  0.996786  0.546646  0.174491   \n",
      "1  0.334295  0.226263  0.646008  0.592361  0.828576  0.937463  0.409783   \n",
      "2  0.274827  0.964028  0.743230  0.706992  0.560090  0.765529  0.337996   \n",
      "3  0.401926  0.532954  0.168215  0.481798  0.099418  0.950383  0.850380   \n",
      "4  0.262429  0.509651  0.114383  0.437847  0.855988  0.803929  0.801882   \n",
      "\n",
      "         7         8         9         10        11        12        13  \n",
      "0  0.160059  0.881592  0.062729  0.652720  0.057432  0.897954  0.556557  \n",
      "1  0.163789  0.859759  0.637650  0.328381  0.949236  0.503253  0.076027  \n",
      "2  0.999088  0.861208  0.989328  0.944619  0.059847  0.337351  0.189525  \n",
      "3  0.888814  0.985114  0.722403  0.848807  0.522381  0.654874  0.372816  \n",
      "4  0.836896  0.821271  0.033005  0.265382  0.154231  0.223672  0.980796  \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x4b6d7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1965    0.450852\n",
      "1966    0.788986\n",
      "1967    0.905450\n",
      "1968    0.219924\n",
      "1969    0.014259\n",
      "Name: Ray, dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc475ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import datetime\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "now = datetime.datetime.now()\n",
    "\n",
    "fileName = 'df_random2arr_{}.txt'.format(now.strftime(\"%Y-%m-%d\"))\n",
    "filePath = './dataset/dat6'\n",
    "iPath = '{}/{}'.format(filePath, fileName)\n",
    "\n",
    "shape1 = (20, 14)\n",
    "arr = np.random.random(shape1)\n",
    "\n",
    "# Создание dataFrame на основе существующего массива\n",
    "df = pd.DataFrame(arr)   # преобразование массива arr к dataFrame\n",
    "print df.head(5) \n",
    "df.to_csv(iPath, sep=';', float_format='%.3f')\n",
    "\n",
    "plt.contourf(df)\n",
    "plt.show()\n",
    "\n",
    "ts = df[5]\n",
    "ts.name = 'Ray'\n",
    "ts.index = np.arange(len(ts)) + 1950 \n",
    "print ts.tail(5)\n",
    "fileName = 'ts_with_header_and_index.dat'\n",
    "ts.to_csv('{}/{}'.format(filePath, fileName),\n",
    "          sep=';', float_format='%.2f', header=True)\n",
    "\n",
    "ts.plot(grid=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### pd.to_excel()\n",
    "\n",
    "Аналогично и с форматами xls/xlsx. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DatetimeIndex(['2017-01-01', '2017-01-02', '2017-01-03', '2017-01-04',\n",
      "               '2017-01-05', '2017-01-06', '2017-01-07', '2017-01-08',\n",
      "               '2017-01-09', '2017-01-10', '2017-01-11', '2017-01-12',\n",
      "               '2017-01-13', '2017-01-14', '2017-01-15', '2017-01-16',\n",
      "               '2017-01-17', '2017-01-18', '2017-01-19', '2017-01-20',\n",
      "               '2017-01-21', '2017-01-22', '2017-01-23', '2017-01-24',\n",
      "               '2017-01-25', '2017-01-26', '2017-01-27', '2017-01-28',\n",
      "               '2017-01-29', '2017-01-30', '2017-01-31'],\n",
      "              dtype='datetime64[ns]', freq='D')\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEB9JREFUeJzt3X+sZGV9x/H3xwUjIkIremMWZKndKpsoVq5oW00vtbWs\ntkUjtqBVS2s2JGLapCassSla/8GaJtYW2W7MBvinWFO1W9gCre6grd2yoIIuFLpZEBZNETXoIg2s\nfPvHHcx4vbsz9865O3Offb+SSeac88yZ77nPvZ8595nzI1WFJKktT5t0AZKk7hnuktQgw12SGmS4\nS1KDDHdJapDhLkkNMtwlqUGGuyQ1yHCXpAYdM6k3Pvnkk2vdunWTevsV9+ijj3L88cdPugwtk/23\nerXed7fddtvDVfXcYe0mFu7r1q3j1ltvndTbr7her8fc3Nyky9Ay2X+rV+t9l+Qbo7RzWEaSGmS4\nS1KDDHdJapDhLkkNMtwlqUFDwz3JtiQPJfn6IZYnyceS7E1yR5KXd1+mJGkpRtlzvwo49zDLNwLr\n+49NwJXjlyVJGsfQcK+qLwDfPUyT84Brat4u4KQkz++qQEnS0nVxEtNa4IGB6f39ed9a2DDJJub3\n7pmZmaHX63Xw9tPpwIEDTW9f6+y/6XXOOed0sp6dO3d2sp5pdUTPUK2qrcBWgNnZ2Wr5LLLWz5Jr\nnf03varqsMvXbb6e+y5/wxGqZnp1Ee4PAqcOTJ/SnydNpSSdrGdYyEiT1MWhkNuBd/SPmnkV8EhV\n/dSQjDQtqmro47RLrxvaRppmQ/fck/w9MAecnGQ/cBlwLEBVbQF2AK8H9gI/BC5aqWIlSaMZGu5V\ndeGQ5QW8u7OKJEljm9glf1e7LsZt/dde0krx8gPL5JitpGlmuEtSgwx3SWqQ4S5JDTLcJalBhrsk\nNchwl6QGGe6S1CDDXZIaZLhLUoMMd0lqkOEuSQ0y3CWpQYa7JDXIcJekBhnuktQgw12SGmS4S1KD\nDHdJapDhLkkNMtwlqUGGuyQ1yHCXpAYZ7pLUIMNdkhpkuEtSgwx3SWqQ4S5JDTLcJalBI4V7knOT\n3J1kb5LNiyw/Mck/J7k9yZ4kF3VfqiRpVEPDPcka4ApgI7ABuDDJhgXN3g3cWVVnAnPAXyV5ese1\nSpJGNMqe+9nA3qraV1WPA9cC5y1oU8AJSQI8C/gucLDTSiVJIxsl3NcCDwxM7+/PG/S3wBnAN4Gv\nAX9cVU92UqEkacmO6Wg9vwl8Ffg14IXAvyb5YlV9f7BRkk3AJoCZmRl6vV5Hbz+dWt++1tl/q5d9\nN1q4PwicOjB9Sn/eoIuAy6uqgL1J7gVeDNwy2KiqtgJbAWZnZ2tubm6ZZa8CN1xP09vXOvtv9bLv\ngNGGZXYD65Oc3v+S9AJg+4I29wOvBUgyA7wI2NdloZKk0Q3dc6+qg0kuAW4E1gDbqmpPkov7y7cA\nHwKuSvI1IMClVfXwCtYtSTqMkcbcq2oHsGPBvC0Dz78JvK7b0ibnzA/exCOPPTH2etZtvn6s1594\n3LHcflkzP1ZJR1BXX6g25ZHHnuC+y98w1jp6vd7Y437jfjhIOnp5+QFJapDhLkkNMtwlqUGGuyQ1\nyHCXpAYZ7pLUIMNdkhpkuEtSgwx3SWqQ4S5JDTLcJalBhrskNchwl6QGGe6S1CDDXZIaZLhLUoMM\nd0lqkOEuSQ0y3CWpQYa7JDXIcJekBh0z6QKkrp35wZt45LEnxl7Pus3XL/u1Jx53LLdf9rqxa5CW\ny3BXcx557Anuu/wNY62j1+sxNze37NeP88EgdcFhGUlqkOEuSQ0y3CWpQYa7JDXIcJekBhnuktQg\nw12SGjRSuCc5N8ndSfYm2XyINnNJvppkT5Kbuy1TkrQUQ09iSrIGuAL4DWA/sDvJ9qq6c6DNScDH\ngXOr6v4kz1upgiW1axrOLoY2zjAe5QzVs4G9VbUPIMm1wHnAnQNt3gp8uqruB6iqh7ouVFL7puHs\nYmjjDONRwn0t8MDA9H7glQva/AJwbJIecALw11V1zcIVJdkEbAKYmZmh1+sto+QjY9zaDhw40Mn2\nTfPPaJpNQ//Zd8szDX3XRR2T1tW1ZY4BzgJeCxwH/GeSXVV1z2CjqtoKbAWYnZ2tcT9dV8wN14/9\nyd/F3kMXdRyVpqH/7LvlmYa+66iOSRsl3B8ETh2YPqU/b9B+4DtV9SjwaJIvAGcC9yBJOuJGOVpm\nN7A+yelJng5cAGxf0OafgFcnOSbJM5kftrmr21IlSaMauudeVQeTXALcCKwBtlXVniQX95dvqaq7\nktwA3AE8CXyiqr6+koVLkg5tpDH3qtoB7Fgwb8uC6Y8AH+muNEnScnmGqiQ1yHCXpAYZ7pLUIMNd\nkhpkuEtSgwx3SWqQ4S5JDTLcJalBXV04TJLGdsIZm3nJ1YveD2hprh63DoDxLj08aYa7pKnxg7su\n93ruHXFYRpIaZLhLUoMMd0lqkOEuSQ0y3CWpQYa7JDXIQyEX4bG2klY7w30RHmsrabVzWEaSGmS4\nS1KDDHdJapDhLkkNMtwlqUGGuyQ1yHCXpAYZ7pLUIMNdkhpkuEtSgwx3SWqQ4S5JDRop3JOcm+Tu\nJHuTHPJyiUlekeRgkvO7K1GStFRDwz3JGuAKYCOwAbgwyYZDtPswcFPXRUqSlmaUPfezgb1Vta+q\nHgeuBc5bpN17gH8EHuqwPknSMowS7muBBwam9/fn/ViStcCbgCu7K02StFxd3azjo8ClVfVkkkM2\nSrIJ2AQwMzNDr9fr6O27N25tBw4c6GT7pvlnNM2mof/su+WZhr7roo6Jq6rDPoBfAm4cmH4f8L4F\nbe4F7us/DjA/NPPGw633rLPOqml12qXXjb2OnTt3TkUdR6Np6D/7bnmmoe+6qmOlALfWkNyuqpH2\n3HcD65OcDjwIXAC8dcEHxOlPPU9yFXBdVX12jM8cSdIYhoZ7VR1McglwI7AG2FZVe5Jc3F++ZYVr\nlCQt0Uhj7lW1A9ixYN6ioV5VfzB+WdLynXDGZl5y9SFPxxjd1ePUADDeTdalcXT1hao0NX5w1+Xc\nd/l4wdrr9Zibm1v269dtvn6s95fG5eUHJKlBhrskNchwl6QGGe6S1CDDXZIaZLhLUoMMd0lqkOEu\nSQ0y3CWpQZ6hKmmqdHJ27w3jrePE444dv4YJM9wlTY1xLxsB8x8OXaxntXNYRpIaZLhLUoMMd0lq\nkOEuSQ0y3CWpQR4tcwgejiVpNTPcF+HhWJJWO4dlJKlBhrskNchwl6QGGe6S1CDDXZIaZLhLUoMM\nd0lqkOEuSQ0y3CWpQYa7JDXIyw+oSZO+NpDXBdKkGe5qjtcGkkYclklybpK7k+xNsnmR5W9LckeS\nryX5UpIzuy9VkjSqoeGeZA1wBbAR2ABcmGTDgmb3Ar9aVS8BPgRs7bpQSdLoRtlzPxvYW1X7qupx\n4FrgvMEGVfWlqvpef3IXcEq3ZUqSlmKUcF8LPDAwvb8/71D+CPiXcYqSJI2n0y9Uk5zDfLi/+hDL\nNwGbAGZmZuj1el2+/dRpfftaZ/+tXvbdaOH+IHDqwPQp/Xk/IclLgU8AG6vqO4utqKq20h+Pn52d\nrbm5uaXWu3rccD1Nb1/r7L/Vy74DRhuW2Q2sT3J6kqcDFwDbBxskeQHwaeDtVXVP92VKkpZi6J57\nVR1McglwI7AG2FZVe5Jc3F++Bfhz4DnAx5MAHKyq2ZUrW5J0OCONuVfVDmDHgnlbBp6/C3hXt6VJ\nkpbLa8tIUoMMd0lqkOEuSQ0y3CWpQYa7JDXIcJekBhnuktQgw12SGmS4S1KDDHdJapDhLkkNMtwl\nqUGGuyQ1yHCXpAYZ7pLUIMNdkhpkuEtSgwx3SWqQ4S5JDTLcJalBhrskNchwl6QGGe6S1CDDXZIa\nZLhLUoMMd0lq0DGTLkA60pKM1u7Dh19eVR1Uo6Uapf+G9R2033/uueuoU1VDHzt37hzaRpPRRd8d\nDf1nuEtSgwx3SWqQ4S5JDRop3JOcm+TuJHuTbF5keZJ8rL/8jiQv775USdKohoZ7kjXAFcBGYANw\nYZINC5ptBNb3H5uAKzuuU5K0BKPsuZ8N7K2qfVX1OHAtcN6CNucB19S8XcBJSZ7fca2SpBGNcpz7\nWuCBgen9wCtHaLMW+NZgoySbmN+zZ2Zmhl6vt8Ryp8c555wztM2wY2137tzZUTXq2oEDB1b17+fR\nzL6bd0RPYqqqrcBWgNnZ2ZqbmzuSb9+pYcfJ9no9VvP2He3sv9XLvps3yrDMg8CpA9On9OcttY0k\n6QgZJdx3A+uTnJ7k6cAFwPYFbbYD7+gfNfMq4JGq+tbCFUmSjoyhwzJVdTDJJcCNwBpgW1XtSXJx\nf/kWYAfwemAv8EPgopUrWZI0zEhj7lW1g/kAH5y3ZeB5Ae/utjRJ0nJ5hqokNchwl6QGGe6S1CDD\nXZIalEldtD7Jt4FvTOTNj4yTgYcnXYSWzf5bvVrvu9Oq6rnDGk0s3FuX5Naqmp10HVoe+2/1su/m\nOSwjSQ0y3CWpQYb7ytk66QI0Fvtv9bLvcMxdkprknrskNchwX6IkH0jy3sMsvyrJ+f3nvSRH/bf2\n0pEw+Le3YP5ckusmUdMkGe5TpH+/Wk2hJEf0xjZaXP+y4ubWCPwhjSDJ+5Pck+TfgRf1570sya4k\ndyT5TJKfGbKOK5PcmmRPkg8OzL8vyYeTfBl4y8puiZL8fpJbknw1yd8lWZPkwMDy85Nc1X9+VZIt\nSf4L+MskP5vks/0+35XkpZPajqNJknVJ7k5yDfB14EcDy37cX32/3v87uyfJby2yruOTbOv/Dnwl\nycL7QTfDvZEhkpzF/A1KXsb8z+vLwG3ANcB7qurmJH8BXAb8yWFW9f6q+m5/7/xzSV5aVXf0l32n\nql6+clshgCRnAL8H/EpVPZHk48DbhrzsFOCXq+pHSf4G+EpVvTHJrzH/O/Cyla1afeuBd1bVrsEP\n40WsA84GXgjsTPLzC5a/H/h8Vf1hkpOAW5L8W1U9uiJVT5DhPtxrgM9U1Q8BkmwHjgdOqqqb+22u\nBj41ZD2/279B+DHA84ENwFPh/snOq9ZiXgucBexOAnAc8NCQ13yqqp7aU3w18GaAqvp8kuckeXZV\nfX+lCtaPfaOqdo3Q7h+q6kngf5LsA168YPnrgN8Z+N7sGcALgLu6K3U6GO5HQJLTgfcCr6iq7/X/\njXzGQJPm9hqmVICrq+p9PzEz+dOByWf85Evsmykx2A+Dx28v7K+Fx3YvnA7w5qq6u6vCppVj7sN9\nAXhjkuOSnAD8NvO/aN9L8pp+m7cDNx9qBcCz+695JMkMsHElC9YhfQ44P8nzAPpj6KcB/5vkjP4X\ndW86zOu/SH8YJ8kc8LB77RNxuP56S5KnJXkh8HPAwhC/EXhP+v+6JfnFlS93MtxzH6Kqvpzkk8Dt\nzP8Lv7u/6J3AliTPBPZxmPvGVtXtSb4C/DfwAPAfK1u1FlNVdyb5M+CmfjA8wfztITcD1wHfBm4F\nnnWIVXwA2JbkDubvFfzOFS9aizlcf90P3ML8DtXFVfV//Rx/yoeAjwJ39H8H7gV+6ovXFniGqiQ1\nyGEZSWqQ4S5JDTLcJalBhrskNchwl6QGGe6S1CDDXZIaZLhLUoP+H/Q1d5xhk6MUAAAAAElFTkSu\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc2d4048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "shape1 = (31, 3)\n",
    "\n",
    "fileName = 'df_{}x{}.xlsx'.format(*shape1)\n",
    "filePath = './dataset/dat6'\n",
    "iPath = '{}/{}'.format(filePath, fileName)\n",
    "\n",
    "arr = np.random.random(shape1)\n",
    "\n",
    "box = ['dollar', 'euro', 'ruble']\n",
    "\n",
    "times = pd.date_range('2017-01-01', '2017-01-31', step='1d' )\n",
    "print times\n",
    "\n",
    "df = pd.DataFrame(arr, columns=box, index=times)\n",
    "\n",
    "df.to_excel(iPath, sheet_name='Money', float_format='%.2f',\n",
    "            header=True, index=True)\n",
    "\n",
    "ax = df.boxplot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}