{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Calling R, Python and C++ from Julia\n",
    "\n",
    "\n",
    "- Douglas Bates\n",
    "- U. of Wisconsin - Madison\n",
    "- github: dmbates\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## In 2012 I started using Julia\n",
    "\n",
    "### Good things\n",
    "- What Viral said: multiple dispatch, JIT method compilation, extensible type system, interesting language constructs.\n",
    "\n",
    "### Bad things\n",
    "- Many things I knew how to do in `R` were not available or needed to be relearned.\n",
    "- Re-learning takes a while for things like data visualization systems\n",
    "- `R` packages often provide data sets for illustration/experimentation. As a rule `Julia` packages don't."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## The `RCall` package (JuliaStats/RCall.jl)\n",
    "\n",
    "- Julia has the `ccall` function. Steven Johnson had written `PyCall`. Avik created `JavaCall`\n",
    "- So I started writing `RCall`.  Most of the recent work has been done by Randy Lai and Simon Byrne.\n",
    "- Basic approach\n",
    "    - Create Julia immutables to mirror `R`'s `SEXPREC` struct (see [`types.jl`](https://github.com/JuliaStats/RCall.jl/blob/master/src/types.jl) )\n",
    "    - Locate and dlopen `libR` (see [`setup.jl`](https://github.com/JuliaStats/RCall.jl/blob/master/src/setup.jl) )\n",
    "    - Start an embedded `R` process\n",
    "    - Call `R`'s API\n",
    "- Functions of interest are `reval`, `rcall`, `rcopy`, etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                        @R_str   2163 bytes  Function\n",
      "                         @rget   1576 bytes  Function\n",
      "                      @rimport   1860 bytes  Function\n",
      "                     @rlibrary   1296 bytes  Function\n",
      "                         @rput   1542 bytes  Function\n",
      "                       @rusing    452 bytes  Function\n",
      "                      @var_str    259 bytes  Function\n",
      "                       CharSxp    136 bytes  DataType\n",
      "                       ClosSxp    148 bytes  DataType\n",
      "                       CplxSxp    136 bytes  DataType\n",
      "                        IntSxp    136 bytes  DataType\n",
      "                        LglSxp    136 bytes  DataType\n",
      "                        NilSxp    112 bytes  DataType\n",
      "                         RCall    566 KB     Module\n",
      "                       RObject    168 bytes  DataType\n",
      "                       RealSxp    136 bytes  DataType\n",
      "                        StrSxp    136 bytes  DataType\n",
      "                           Sxp     92 bytes  DataType\n",
      "                         anyNA   1377 bytes  Function\n",
      "                     getAttrib   2722 bytes  Function\n",
      "                      getNames    978 bytes  Function\n",
      "                     globalEnv      8 bytes  RCall.RObject{RCall.EnvSxp}\n",
      "                      isFactor   1066 bytes  Function\n",
      "                          isNA   4223 bytes  Function\n",
      "                     isOrdered   1067 bytes  Function\n",
      "                         rcall   3149 bytes  Function\n",
      "                         rcopy     39 KB     Function\n",
      "                         reval   6289 bytes  Function\n",
      "                         rlang   2229 bytes  Function\n",
      "                        rparse    829 bytes  Function\n",
      "                        rprint   4106 bytes  Function\n",
      "                    setAttrib!   4453 bytes  Function\n",
      "                     setNames!   1054 bytes  Function\n"
     ]
    }
   ],
   "source": [
    "using RCall  # make the package available\n",
    "whos(RCall)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "search: rcall RCall remotecall remotecall_wait remotecall_fetch UniformScaling\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/latex": [
       "Evaluate a function in the global environment. The first argument corresponds to the function to be called. It can be either a FunctionSxp type, a SymSxp or a Symbol.\n"
      ],
      "text/markdown": [
       "Evaluate a function in the global environment. The first argument corresponds to the function to be called. It can be either a FunctionSxp type, a SymSxp or a Symbol.\n"
      ],
      "text/plain": [
       "Evaluate a function in the global environment. The first argument corresponds to the function to be called. It can be either a FunctionSxp type, a SymSxp or a Symbol.\n"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "?rcall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "search: rcopy precompile __precompile__ readchomp ProcessGroup\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/latex": [
       "Evaluate and convert the result of a string as an R expression.\n",
       "\\texttt{rcopy} copies the contents of an R object into a corresponding canonical Julia type.\n",
       "\\texttt{rcopy(T,p)} converts a pointer \\texttt{p} to a Sxp object to a native Julia object of type T.\n",
       "\\texttt{rcopy(p)} performs a default conversion.\n"
      ],
      "text/markdown": [
       "Evaluate and convert the result of a string as an R expression.\n",
       "\n",
       "`rcopy` copies the contents of an R object into a corresponding canonical Julia type.\n",
       "\n",
       "`rcopy(T,p)` converts a pointer `p` to a Sxp object to a native Julia object of type T.\n",
       "\n",
       "`rcopy(p)` performs a default conversion.\n"
      ],
      "text/plain": [
       "Evaluate and convert the result of a string as an R expression.\n",
       "\n",
       "`rcopy` copies the contents of an R object into a corresponding canonical Julia type.\n",
       "\n",
       "`rcopy(T,p)` converts a pointer `p` to a Sxp object to a native Julia object of type T.\n",
       "\n",
       "`rcopy(p)` performs a default conversion.\n"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "?rcopy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><tr><th></th><th>mpg</th><th>cyl</th><th>disp</th><th>hp</th><th>drat</th><th>wt</th><th>qsec</th><th>vs</th><th>am</th><th>gear</th><th>carb</th></tr><tr><th>1</th><td>21.0</td><td>6.0</td><td>160.0</td><td>110.0</td><td>3.9</td><td>2.62</td><td>16.46</td><td>0.0</td><td>1.0</td><td>4.0</td><td>4.0</td></tr><tr><th>2</th><td>21.0</td><td>6.0</td><td>160.0</td><td>110.0</td><td>3.9</td><td>2.875</td><td>17.02</td><td>0.0</td><td>1.0</td><td>4.0</td><td>4.0</td></tr><tr><th>3</th><td>22.8</td><td>4.0</td><td>108.0</td><td>93.0</td><td>3.85</td><td>2.32</td><td>18.61</td><td>1.0</td><td>1.0</td><td>4.0</td><td>1.0</td></tr><tr><th>4</th><td>21.4</td><td>6.0</td><td>258.0</td><td>110.0</td><td>3.08</td><td>3.215</td><td>19.44</td><td>1.0</td><td>0.0</td><td>3.0</td><td>1.0</td></tr><tr><th>5</th><td>18.7</td><td>8.0</td><td>360.0</td><td>175.0</td><td>3.15</td><td>3.44</td><td>17.02</td><td>0.0</td><td>0.0</td><td>3.0</td><td>2.0</td></tr><tr><th>6</th><td>18.1</td><td>6.0</td><td>225.0</td><td>105.0</td><td>2.76</td><td>3.46</td><td>20.22</td><td>1.0</td><td>0.0</td><td>3.0</td><td>1.0</td></tr><tr><th>7</th><td>14.3</td><td>8.0</td><td>360.0</td><td>245.0</td><td>3.21</td><td>3.57</td><td>15.84</td><td>0.0</td><td>0.0</td><td>3.0</td><td>4.0</td></tr><tr><th>8</th><td>24.4</td><td>4.0</td><td>146.7</td><td>62.0</td><td>3.69</td><td>3.19</td><td>20.0</td><td>1.0</td><td>0.0</td><td>4.0</td><td>2.0</td></tr><tr><th>9</th><td>22.8</td><td>4.0</td><td>140.8</td><td>95.0</td><td>3.92</td><td>3.15</td><td>22.9</td><td>1.0</td><td>0.0</td><td>4.0</td><td>2.0</td></tr><tr><th>10</th><td>19.2</td><td>6.0</td><td>167.6</td><td>123.0</td><td>3.92</td><td>3.44</td><td>18.3</td><td>1.0</td><td>0.0</td><td>4.0</td><td>4.0</td></tr><tr><th>11</th><td>17.8</td><td>6.0</td><td>167.6</td><td>123.0</td><td>3.92</td><td>3.44</td><td>18.9</td><td>1.0</td><td>0.0</td><td>4.0</td><td>4.0</td></tr><tr><th>12</th><td>16.4</td><td>8.0</td><td>275.8</td><td>180.0</td><td>3.07</td><td>4.07</td><td>17.4</td><td>0.0</td><td>0.0</td><td>3.0</td><td>3.0</td></tr><tr><th>13</th><td>17.3</td><td>8.0</td><td>275.8</td><td>180.0</td><td>3.07</td><td>3.73</td><td>17.6</td><td>0.0</td><td>0.0</td><td>3.0</td><td>3.0</td></tr><tr><th>14</th><td>15.2</td><td>8.0</td><td>275.8</td><td>180.0</td><td>3.07</td><td>3.78</td><td>18.0</td><td>0.0</td><td>0.0</td><td>3.0</td><td>3.0</td></tr><tr><th>15</th><td>10.4</td><td>8.0</td><td>472.0</td><td>205.0</td><td>2.93</td><td>5.25</td><td>17.98</td><td>0.0</td><td>0.0</td><td>3.0</td><td>4.0</td></tr><tr><th>16</th><td>10.4</td><td>8.0</td><td>460.0</td><td>215.0</td><td>3.0</td><td>5.424</td><td>17.82</td><td>0.0</td><td>0.0</td><td>3.0</td><td>4.0</td></tr><tr><th>17</th><td>14.7</td><td>8.0</td><td>440.0</td><td>230.0</td><td>3.23</td><td>5.345</td><td>17.42</td><td>0.0</td><td>0.0</td><td>3.0</td><td>4.0</td></tr><tr><th>18</th><td>32.4</td><td>4.0</td><td>78.7</td><td>66.0</td><td>4.08</td><td>2.2</td><td>19.47</td><td>1.0</td><td>1.0</td><td>4.0</td><td>1.0</td></tr><tr><th>19</th><td>30.4</td><td>4.0</td><td>75.7</td><td>52.0</td><td>4.93</td><td>1.615</td><td>18.52</td><td>1.0</td><td>1.0</td><td>4.0</td><td>2.0</td></tr><tr><th>20</th><td>33.9</td><td>4.0</td><td>71.1</td><td>65.0</td><td>4.22</td><td>1.835</td><td>19.9</td><td>1.0</td><td>1.0</td><td>4.0</td><td>1.0</td></tr><tr><th>21</th><td>21.5</td><td>4.0</td><td>120.1</td><td>97.0</td><td>3.7</td><td>2.465</td><td>20.01</td><td>1.0</td><td>0.0</td><td>3.0</td><td>1.0</td></tr><tr><th>22</th><td>15.5</td><td>8.0</td><td>318.0</td><td>150.0</td><td>2.76</td><td>3.52</td><td>16.87</td><td>0.0</td><td>0.0</td><td>3.0</td><td>2.0</td></tr><tr><th>23</th><td>15.2</td><td>8.0</td><td>304.0</td><td>150.0</td><td>3.15</td><td>3.435</td><td>17.3</td><td>0.0</td><td>0.0</td><td>3.0</td><td>2.0</td></tr><tr><th>24</th><td>13.3</td><td>8.0</td><td>350.0</td><td>245.0</td><td>3.73</td><td>3.84</td><td>15.41</td><td>0.0</td><td>0.0</td><td>3.0</td><td>4.0</td></tr><tr><th>25</th><td>19.2</td><td>8.0</td><td>400.0</td><td>175.0</td><td>3.08</td><td>3.845</td><td>17.05</td><td>0.0</td><td>0.0</td><td>3.0</td><td>2.0</td></tr><tr><th>26</th><td>27.3</td><td>4.0</td><td>79.0</td><td>66.0</td><td>4.08</td><td>1.935</td><td>18.9</td><td>1.0</td><td>1.0</td><td>4.0</td><td>1.0</td></tr><tr><th>27</th><td>26.0</td><td>4.0</td><td>120.3</td><td>91.0</td><td>4.43</td><td>2.14</td><td>16.7</td><td>0.0</td><td>1.0</td><td>5.0</td><td>2.0</td></tr><tr><th>28</th><td>30.4</td><td>4.0</td><td>95.1</td><td>113.0</td><td>3.77</td><td>1.513</td><td>16.9</td><td>1.0</td><td>1.0</td><td>5.0</td><td>2.0</td></tr><tr><th>29</th><td>15.8</td><td>8.0</td><td>351.0</td><td>264.0</td><td>4.22</td><td>3.17</td><td>14.5</td><td>0.0</td><td>1.0</td><td>5.0</td><td>4.0</td></tr><tr><th>30</th><td>19.7</td><td>6.0</td><td>145.0</td><td>175.0</td><td>3.62</td><td>2.77</td><td>15.5</td><td>0.0</td><td>1.0</td><td>5.0</td><td>6.0</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td></tr></table>"
      ],
      "text/plain": [
       "32×11 DataFrames.DataFrame\n",
       "│ Row │ mpg  │ cyl │ disp  │ hp    │ drat │ wt    │ qsec  │ vs  │ am  │ gear │\n",
       "├─────┼──────┼─────┼───────┼───────┼──────┼───────┼───────┼─────┼─────┼──────┤\n",
       "│ 1   │ 21.0 │ 6.0 │ 160.0 │ 110.0 │ 3.9  │ 2.62  │ 16.46 │ 0.0 │ 1.0 │ 4.0  │\n",
       "│ 2   │ 21.0 │ 6.0 │ 160.0 │ 110.0 │ 3.9  │ 2.875 │ 17.02 │ 0.0 │ 1.0 │ 4.0  │\n",
       "│ 3   │ 22.8 │ 4.0 │ 108.0 │ 93.0  │ 3.85 │ 2.32  │ 18.61 │ 1.0 │ 1.0 │ 4.0  │\n",
       "│ 4   │ 21.4 │ 6.0 │ 258.0 │ 110.0 │ 3.08 │ 3.215 │ 19.44 │ 1.0 │ 0.0 │ 3.0  │\n",
       "│ 5   │ 18.7 │ 8.0 │ 360.0 │ 175.0 │ 3.15 │ 3.44  │ 17.02 │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 6   │ 18.1 │ 6.0 │ 225.0 │ 105.0 │ 2.76 │ 3.46  │ 20.22 │ 1.0 │ 0.0 │ 3.0  │\n",
       "│ 7   │ 14.3 │ 8.0 │ 360.0 │ 245.0 │ 3.21 │ 3.57  │ 15.84 │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 8   │ 24.4 │ 4.0 │ 146.7 │ 62.0  │ 3.69 │ 3.19  │ 20.0  │ 1.0 │ 0.0 │ 4.0  │\n",
       "│ 9   │ 22.8 │ 4.0 │ 140.8 │ 95.0  │ 3.92 │ 3.15  │ 22.9  │ 1.0 │ 0.0 │ 4.0  │\n",
       "│ 10  │ 19.2 │ 6.0 │ 167.6 │ 123.0 │ 3.92 │ 3.44  │ 18.3  │ 1.0 │ 0.0 │ 4.0  │\n",
       "│ 11  │ 17.8 │ 6.0 │ 167.6 │ 123.0 │ 3.92 │ 3.44  │ 18.9  │ 1.0 │ 0.0 │ 4.0  │\n",
       "⋮\n",
       "│ 21  │ 21.5 │ 4.0 │ 120.1 │ 97.0  │ 3.7  │ 2.465 │ 20.01 │ 1.0 │ 0.0 │ 3.0  │\n",
       "│ 22  │ 15.5 │ 8.0 │ 318.0 │ 150.0 │ 2.76 │ 3.52  │ 16.87 │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 23  │ 15.2 │ 8.0 │ 304.0 │ 150.0 │ 3.15 │ 3.435 │ 17.3  │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 24  │ 13.3 │ 8.0 │ 350.0 │ 245.0 │ 3.73 │ 3.84  │ 15.41 │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 25  │ 19.2 │ 8.0 │ 400.0 │ 175.0 │ 3.08 │ 3.845 │ 17.05 │ 0.0 │ 0.0 │ 3.0  │\n",
       "│ 26  │ 27.3 │ 4.0 │ 79.0  │ 66.0  │ 4.08 │ 1.935 │ 18.9  │ 1.0 │ 1.0 │ 4.0  │\n",
       "│ 27  │ 26.0 │ 4.0 │ 120.3 │ 91.0  │ 4.43 │ 2.14  │ 16.7  │ 0.0 │ 1.0 │ 5.0  │\n",
       "│ 28  │ 30.4 │ 4.0 │ 95.1  │ 113.0 │ 3.77 │ 1.513 │ 16.9  │ 1.0 │ 1.0 │ 5.0  │\n",
       "│ 29  │ 15.8 │ 8.0 │ 351.0 │ 264.0 │ 4.22 │ 3.17  │ 14.5  │ 0.0 │ 1.0 │ 5.0  │\n",
       "│ 30  │ 19.7 │ 6.0 │ 145.0 │ 175.0 │ 3.62 │ 2.77  │ 15.5  │ 0.0 │ 1.0 │ 5.0  │\n",
       "│ 31  │ 15.0 │ 8.0 │ 301.0 │ 335.0 │ 3.54 │ 3.57  │ 14.6  │ 0.0 │ 1.0 │ 5.0  │\n",
       "│ 32  │ 21.4 │ 4.0 │ 121.0 │ 109.0 │ 4.11 │ 2.78  │ 18.6  │ 1.0 │ 1.0 │ 4.0  │\n",
       "\n",
       "│ Row │ carb │\n",
       "├─────┼──────┤\n",
       "│ 1   │ 4.0  │\n",
       "│ 2   │ 4.0  │\n",
       "│ 3   │ 1.0  │\n",
       "│ 4   │ 1.0  │\n",
       "│ 5   │ 2.0  │\n",
       "│ 6   │ 1.0  │\n",
       "│ 7   │ 4.0  │\n",
       "│ 8   │ 2.0  │\n",
       "│ 9   │ 2.0  │\n",
       "│ 10  │ 4.0  │\n",
       "│ 11  │ 4.0  │\n",
       "⋮\n",
       "│ 21  │ 1.0  │\n",
       "│ 22  │ 2.0  │\n",
       "│ 23  │ 2.0  │\n",
       "│ 24  │ 4.0  │\n",
       "│ 25  │ 2.0  │\n",
       "│ 26  │ 1.0  │\n",
       "│ 27  │ 2.0  │\n",
       "│ 28  │ 2.0  │\n",
       "│ 29  │ 4.0  │\n",
       "│ 30  │ 6.0  │\n",
       "│ 31  │ 8.0  │\n",
       "│ 32  │ 2.0  │"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mtcars = rcopy(\"mtcars\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A simpler interface\n",
    "- the character `R` followed by a string evaluates the string as an `R` expression.  The behavior is defined in `@R_str`. See [`rstr.jl`](https://github.com/JuliaStats/RCall.jl/blob/master/src/rstr.jl).\n",
    "- Julia allows multiline strings using triple quotes\n",
    "- string interpolation of Julia objects is allowed using `$`, only when it is not used as valid `R` syntax.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7525806925786059"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "R\"\"\"\n",
    "suppressMessages(library(lme4))\n",
    "fm1 <- lmer(Yield ~ 1 + (1 | Batch), Dyestuff, REML = FALSE)\n",
    "getME(fm1, \"theta\")\n",
    "\"\"\"[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "RCall.RObject{RCall.VecSxp}\n"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using DataFrames\n",
    "brownian(N) = DataFrame(x = 1:N, y = cumsum(randn(N)))\n",
    "srand(1234321)  # set random number seed\n",
    "br = brownian(10_000);\n",
    "R\"\"\"\n",
    "suppressMessages(library(ggplot2))\n",
    "ggplot($br, aes(x = x, y = y)) + geom_line()\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## An even simpler interface\n",
    "\n",
    "- Most of base Julia, including the REPL, is written in Julia\n",
    "- The REPL allows for different, user-extensible, modes: julia, shell, help, ..\n",
    "- Modes can be switched by the first character you type at the prompt (`?` for help; `;` for shell)\n",
    "- `RCall` adds an R REPL mode (magic char is `$`).  `Cxx` adds a `C++` REPL mode (magic is `<`).  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## RCall summary\n",
    "\n",
    "- There is no \"glue\" code written in a compiled language.  The whole package is written in Julia.\n",
    "- It is feasible to emulate the internal R structures in Julia.  The other direction would be very difficult.\n",
    "- `ccall`, `cglobal`, `cfunction`, `Libdl.dlopen` allow low-level access to a C API.\n",
    "- String macros and REPL modes allow for a familiar interface."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## PyCall (stevengj/PyCall.jl)\n",
    "\n",
    "- the `@pyimport` macro functions like `import` in Python."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "using PyCall\n",
    "@pyimport numpy as np\n",
    "@pyimport pandas as pd\n",
    "@pyimport feather"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   Categorical      8 bytes  PyCall.PyObject\n",
      "              CategoricalIndex      8 bytes  PyCall.PyObject\n",
      "                     DataFrame      8 bytes  PyCall.PyObject\n",
      "                    DateOffset      8 bytes  PyCall.PyObject\n",
      "                 DatetimeIndex      8 bytes  PyCall.PyObject\n",
      "                     ExcelFile      8 bytes  PyCall.PyObject\n",
      "                   ExcelWriter      8 bytes  PyCall.PyObject\n",
      "                          Expr      8 bytes  PyCall.PyObject\n",
      "                  Float64Index      8 bytes  PyCall.PyObject\n",
      "                       Grouper      8 bytes  PyCall.PyObject\n",
      "                      HDFStore      8 bytes  PyCall.PyObject\n",
      "                         Index      8 bytes  PyCall.PyObject\n",
      "                    IndexSlice      8 bytes  PyCall.PyObject\n",
      "                    Int64Index      8 bytes  PyCall.PyObject\n",
      "                    MultiIndex      8 bytes  PyCall.PyObject\n",
      "                           NaT      8 bytes  PyCall.PyObject\n",
      "                         Panel      8 bytes  PyCall.PyObject\n",
      "                       Panel4D      8 bytes  PyCall.PyObject\n",
      "                        Period      8 bytes  PyCall.PyObject\n",
      "                   PeriodIndex      8 bytes  PyCall.PyObject\n",
      "                    RangeIndex      8 bytes  PyCall.PyObject\n",
      "                        Series      8 bytes  PyCall.PyObject\n",
      "                   SparseArray      8 bytes  PyCall.PyObject\n",
      "               SparseDataFrame      8 bytes  PyCall.PyObject\n",
      "                    SparseList      8 bytes  PyCall.PyObject\n",
      "                   SparsePanel      8 bytes  PyCall.PyObject\n",
      "                  SparseSeries      8 bytes  PyCall.PyObject\n",
      "              SparseTimeSeries      8 bytes  PyCall.PyObject\n",
      "                          Term      8 bytes  PyCall.PyObject\n",
      "                   TimeGrouper      8 bytes  PyCall.PyObject\n",
      "                    TimeSeries      8 bytes  PyCall.PyObject\n",
      "                     Timedelta      8 bytes  PyCall.PyObject\n",
      "                TimedeltaIndex      8 bytes  PyCall.PyObject\n",
      "                     Timestamp      8 bytes  PyCall.PyObject\n",
      "                     WidePanel      8 bytes  PyCall.PyObject\n",
      "                      __anon__     10 KB     Module\n",
      "                         algos      8 bytes  PyCall.PyObject\n",
      "                   bdate_range      8 bytes  PyCall.PyObject\n",
      "                        compat      8 bytes  PyCall.PyObject\n",
      "                   computation      8 bytes  PyCall.PyObject\n",
      "                        concat      8 bytes  PyCall.PyObject\n",
      "                          core      8 bytes  PyCall.PyObject\n",
      "                      crosstab      8 bytes  PyCall.PyObject\n",
      "                           cut      8 bytes  PyCall.PyObject\n",
      "                    date_range      8 bytes  PyCall.PyObject\n",
      "                      datetime      8 bytes  PyCall.PyObject\n",
      "                     datetools      8 bytes  PyCall.PyObject\n",
      "                    dependency     16 bytes  ASCIIString\n",
      "               describe_option      8 bytes  PyCall.PyObject\n",
      "                          eval      8 bytes  PyCall.PyObject\n",
      "                          ewma      8 bytes  PyCall.PyObject\n",
      "                       ewmcorr      8 bytes  PyCall.PyObject\n",
      "                        ewmcov      8 bytes  PyCall.PyObject\n",
      "                        ewmstd      8 bytes  PyCall.PyObject\n",
      "                        ewmvar      8 bytes  PyCall.PyObject\n",
      "                        ewmvol      8 bytes  PyCall.PyObject\n",
      "               expanding_apply      8 bytes  PyCall.PyObject\n",
      "                expanding_corr      8 bytes  PyCall.PyObject\n",
      "               expanding_count      8 bytes  PyCall.PyObject\n",
      "                 expanding_cov      8 bytes  PyCall.PyObject\n",
      "                expanding_kurt      8 bytes  PyCall.PyObject\n",
      "                 expanding_max      8 bytes  PyCall.PyObject\n",
      "                expanding_mean      8 bytes  PyCall.PyObject\n",
      "              expanding_median      8 bytes  PyCall.PyObject\n",
      "                 expanding_min      8 bytes  PyCall.PyObject\n",
      "            expanding_quantile      8 bytes  PyCall.PyObject\n",
      "                expanding_skew      8 bytes  PyCall.PyObject\n",
      "                 expanding_std      8 bytes  PyCall.PyObject\n",
      "                 expanding_sum      8 bytes  PyCall.PyObject\n",
      "                 expanding_var      8 bytes  PyCall.PyObject\n",
      "                     factorize      8 bytes  PyCall.PyObject\n",
      "                  fama_macbeth      8 bytes  PyCall.PyObject\n",
      "                       formats      8 bytes  PyCall.PyObject\n",
      "                   get_dummies      8 bytes  PyCall.PyObject\n",
      "                    get_option      8 bytes  PyCall.PyObject\n",
      "                     get_store      8 bytes  PyCall.PyObject\n",
      "                       groupby      8 bytes  PyCall.PyObject\n",
      "             hard_dependencies     65 bytes  Tuple{ASCIIString,ASCIIString,ASCI…\n",
      "                     hashtable      8 bytes  PyCall.PyObject\n",
      "                         index      8 bytes  PyCall.PyObject\n",
      "                       indexes      8 bytes  PyCall.PyObject\n",
      "                    infer_freq      8 bytes  PyCall.PyObject\n",
      "                          info      8 bytes  PyCall.PyObject\n",
      "                            io      8 bytes  PyCall.PyObject\n",
      "                        isnull      8 bytes  PyCall.PyObject\n",
      "                          json      8 bytes  PyCall.PyObject\n",
      "                           lib      8 bytes  PyCall.PyObject\n",
      "                      lreshape      8 bytes  PyCall.PyObject\n",
      "                         match      8 bytes  PyCall.PyObject\n",
      "                          melt      8 bytes  PyCall.PyObject\n",
      "                         merge      8 bytes  PyCall.PyObject\n",
      "          missing_dependencies      0 bytes  0-element Array{Any,1}\n",
      "                       msgpack      8 bytes  PyCall.PyObject\n",
      "                       notnull      8 bytes  PyCall.PyObject\n",
      "                            np      8 bytes  PyCall.PyObject\n",
      "                       offsets      8 bytes  PyCall.PyObject\n",
      "                           ols      8 bytes  PyCall.PyObject\n",
      "                option_context      8 bytes  PyCall.PyObject\n",
      "                       options      8 bytes  PyCall.PyObject\n",
      "                 ordered_merge      8 bytes  PyCall.PyObject\n",
      "                        pandas      8 bytes  PyCall.PyObject\n",
      "                        parser      8 bytes  PyCall.PyObject\n",
      "                  period_range      8 bytes  PyCall.PyObject\n",
      "                         pivot      8 bytes  PyCall.PyObject\n",
      "                   pivot_table      8 bytes  PyCall.PyObject\n",
      "                   plot_params    357 bytes  Dict{Any,Any} with 1 entry\n",
      "                          pnow      8 bytes  PyCall.PyObject\n",
      "                          qcut      8 bytes  PyCall.PyObject\n",
      "                read_clipboard      8 bytes  PyCall.PyObject\n",
      "                      read_csv      8 bytes  PyCall.PyObject\n",
      "                    read_excel      8 bytes  PyCall.PyObject\n",
      "                      read_fwf      8 bytes  PyCall.PyObject\n",
      "                      read_gbq      8 bytes  PyCall.PyObject\n",
      "                      read_hdf      8 bytes  PyCall.PyObject\n",
      "                     read_html      8 bytes  PyCall.PyObject\n",
      "                     read_json      8 bytes  PyCall.PyObject\n",
      "                  read_msgpack      8 bytes  PyCall.PyObject\n",
      "                   read_pickle      8 bytes  PyCall.PyObject\n",
      "                      read_sas      8 bytes  PyCall.PyObject\n",
      "                      read_sql      8 bytes  PyCall.PyObject\n",
      "                read_sql_query      8 bytes  PyCall.PyObject\n",
      "                read_sql_table      8 bytes  PyCall.PyObject\n",
      "                    read_stata      8 bytes  PyCall.PyObject\n",
      "                    read_table      8 bytes  PyCall.PyObject\n",
      "                  reset_option      8 bytes  PyCall.PyObject\n",
      "                 rolling_apply      8 bytes  PyCall.PyObject\n",
      "                  rolling_corr      8 bytes  PyCall.PyObject\n",
      "                 rolling_count      8 bytes  PyCall.PyObject\n",
      "                   rolling_cov      8 bytes  PyCall.PyObject\n",
      "                  rolling_kurt      8 bytes  PyCall.PyObject\n",
      "                   rolling_max      8 bytes  PyCall.PyObject\n",
      "                  rolling_mean      8 bytes  PyCall.PyObject\n",
      "                rolling_median      8 bytes  PyCall.PyObject\n",
      "                   rolling_min      8 bytes  PyCall.PyObject\n",
      "              rolling_quantile      8 bytes  PyCall.PyObject\n",
      "                  rolling_skew      8 bytes  PyCall.PyObject\n",
      "                   rolling_std      8 bytes  PyCall.PyObject\n",
      "                   rolling_sum      8 bytes  PyCall.PyObject\n",
      "                   rolling_var      8 bytes  PyCall.PyObject\n",
      "                rolling_window      8 bytes  PyCall.PyObject\n",
      "                scatter_matrix      8 bytes  PyCall.PyObject\n",
      "          set_eng_float_format      8 bytes  PyCall.PyObject\n",
      "                    set_option      8 bytes  PyCall.PyObject\n",
      "                 show_versions      8 bytes  PyCall.PyObject\n",
      "                        sparse      8 bytes  PyCall.PyObject\n",
      "                         stats      8 bytes  PyCall.PyObject\n",
      "                          test      8 bytes  PyCall.PyObject\n",
      "               timedelta_range      8 bytes  PyCall.PyObject\n",
      "                   to_datetime      8 bytes  PyCall.PyObject\n",
      "                    to_msgpack      8 bytes  PyCall.PyObject\n",
      "                    to_numeric      8 bytes  PyCall.PyObject\n",
      "                     to_pickle      8 bytes  PyCall.PyObject\n",
      "                  to_timedelta      8 bytes  PyCall.PyObject\n",
      "                         tools      8 bytes  PyCall.PyObject\n",
      "                       tseries      8 bytes  PyCall.PyObject\n",
      "                         tslib      8 bytes  PyCall.PyObject\n",
      "                         types      8 bytes  PyCall.PyObject\n",
      "                        unique      8 bytes  PyCall.PyObject\n",
      "                          util      8 bytes  PyCall.PyObject\n",
      "                  value_counts      8 bytes  PyCall.PyObject\n",
      "                  wide_to_long      8 bytes  PyCall.PyObject\n"
     ]
    }
   ],
   "source": [
    "whos(pd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "brown (generic function with 1 method)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "function brown(N::Number)\n",
    "    v = zeros(N + 1)\n",
    "    for i in 1:N\n",
    "        v[i + 1] = v[i] + randn()\n",
    "    end\n",
    "    v\n",
    "end\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>br1</th>\n",
       "      <th>br2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.773834</td>\n",
       "      <td>-0.006075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.537581</td>\n",
       "      <td>-0.281177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.070239</td>\n",
       "      <td>0.989620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.913922</td>\n",
       "      <td>0.272461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.783253</td>\n",
       "      <td>0.943957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.847161</td>\n",
       "      <td>0.796048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.448132</td>\n",
       "      <td>1.478668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>-2.358731</td>\n",
       "      <td>0.106409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>-3.037559</td>\n",
       "      <td>-0.322209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>-0.647172</td>\n",
       "      <td>-0.024637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>-0.324446</td>\n",
       "      <td>0.887330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>-1.087391</td>\n",
       "      <td>0.764816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>-0.092598</td>\n",
       "      <td>1.635592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.270325</td>\n",
       "      <td>1.103798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.026771</td>\n",
       "      <td>0.601902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>-1.469982</td>\n",
       "      <td>1.815822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>-2.144328</td>\n",
       "      <td>0.957945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>-3.344290</td>\n",
       "      <td>1.105669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>-2.112758</td>\n",
       "      <td>1.303400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>-3.072021</td>\n",
       "      <td>0.911289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>-2.243848</td>\n",
       "      <td>1.488212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>-1.276259</td>\n",
       "      <td>1.990742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>-2.105055</td>\n",
       "      <td>2.864032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>-1.909855</td>\n",
       "      <td>4.591759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>-2.241574</td>\n",
       "      <td>4.498321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>-2.712888</td>\n",
       "      <td>4.342676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>-2.501568</td>\n",
       "      <td>4.245411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>-2.737541</td>\n",
       "      <td>3.471929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>-1.666890</td>\n",
       "      <td>3.263433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9971</th>\n",
       "      <td>-67.201469</td>\n",
       "      <td>-37.090684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9972</th>\n",
       "      <td>-68.372847</td>\n",
       "      <td>-37.603236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9973</th>\n",
       "      <td>-69.053534</td>\n",
       "      <td>-37.842471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9974</th>\n",
       "      <td>-67.648497</td>\n",
       "      <td>-38.373551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9975</th>\n",
       "      <td>-65.773006</td>\n",
       "      <td>-39.242039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9976</th>\n",
       "      <td>-66.520155</td>\n",
       "      <td>-39.173315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9977</th>\n",
       "      <td>-66.774692</td>\n",
       "      <td>-40.773012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9978</th>\n",
       "      <td>-67.839113</td>\n",
       "      <td>-39.828624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9979</th>\n",
       "      <td>-68.676012</td>\n",
       "      <td>-40.120377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9980</th>\n",
       "      <td>-68.364612</td>\n",
       "      <td>-38.445496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9981</th>\n",
       "      <td>-66.843212</td>\n",
       "      <td>-38.913253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9982</th>\n",
       "      <td>-66.390980</td>\n",
       "      <td>-38.201632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9983</th>\n",
       "      <td>-65.733951</td>\n",
       "      <td>-37.470297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9984</th>\n",
       "      <td>-65.099306</td>\n",
       "      <td>-38.119241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9985</th>\n",
       "      <td>-66.354152</td>\n",
       "      <td>-37.442763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9986</th>\n",
       "      <td>-68.651218</td>\n",
       "      <td>-38.013503</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9987</th>\n",
       "      <td>-67.823224</td>\n",
       "      <td>-37.741410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9988</th>\n",
       "      <td>-68.053629</td>\n",
       "      <td>-38.639877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9989</th>\n",
       "      <td>-67.943860</td>\n",
       "      <td>-36.262430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9990</th>\n",
       "      <td>-66.721974</td>\n",
       "      <td>-35.792575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9991</th>\n",
       "      <td>-67.957720</td>\n",
       "      <td>-35.626207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9992</th>\n",
       "      <td>-68.531987</td>\n",
       "      <td>-34.892304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9993</th>\n",
       "      <td>-68.036708</td>\n",
       "      <td>-34.475120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9994</th>\n",
       "      <td>-69.109629</td>\n",
       "      <td>-34.392904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9995</th>\n",
       "      <td>-69.369302</td>\n",
       "      <td>-34.677581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9996</th>\n",
       "      <td>-69.953834</td>\n",
       "      <td>-36.628048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9997</th>\n",
       "      <td>-69.380250</td>\n",
       "      <td>-37.220133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9998</th>\n",
       "      <td>-69.382725</td>\n",
       "      <td>-36.422765</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9999</th>\n",
       "      <td>-69.528264</td>\n",
       "      <td>-35.077929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10000</th>\n",
       "      <td>-68.787819</td>\n",
       "      <td>-35.410307</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10001 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "PyObject              br1        br2\n",
       "0       0.000000   0.000000\n",
       "1      -0.773834  -0.006075\n",
       "2      -0.537581  -0.281177\n",
       "3       0.070239   0.989620\n",
       "4       0.913922   0.272461\n",
       "5       0.783253   0.943957\n",
       "6       1.847161   0.796048\n",
       "7       0.448132   1.478668\n",
       "8      -2.358731   0.106409\n",
       "9      -3.037559  -0.322209\n",
       "10     -0.647172  -0.024637\n",
       "11     -0.324446   0.887330\n",
       "12     -1.087391   0.764816\n",
       "13     -0.092598   1.635592\n",
       "14      0.270325   1.103798\n",
       "15      0.026771   0.601902\n",
       "16     -1.469982   1.815822\n",
       "17     -2.144328   0.957945\n",
       "18     -3.344290   1.105669\n",
       "19     -2.112758   1.303400\n",
       "20     -3.072021   0.911289\n",
       "21     -2.243848   1.488212\n",
       "22     -1.276259   1.990742\n",
       "23     -2.105055   2.864032\n",
       "24     -1.909855   4.591759\n",
       "25     -2.241574   4.498321\n",
       "26     -2.712888   4.342676\n",
       "27     -2.501568   4.245411\n",
       "28     -2.737541   3.471929\n",
       "29     -1.666890   3.263433\n",
       "...          ...        ...\n",
       "9971  -67.201469 -37.090684\n",
       "9972  -68.372847 -37.603236\n",
       "9973  -69.053534 -37.842471\n",
       "9974  -67.648497 -38.373551\n",
       "9975  -65.773006 -39.242039\n",
       "9976  -66.520155 -39.173315\n",
       "9977  -66.774692 -40.773012\n",
       "9978  -67.839113 -39.828624\n",
       "9979  -68.676012 -40.120377\n",
       "9980  -68.364612 -38.445496\n",
       "9981  -66.843212 -38.913253\n",
       "9982  -66.390980 -38.201632\n",
       "9983  -65.733951 -37.470297\n",
       "9984  -65.099306 -38.119241\n",
       "9985  -66.354152 -37.442763\n",
       "9986  -68.651218 -38.013503\n",
       "9987  -67.823224 -37.741410\n",
       "9988  -68.053629 -38.639877\n",
       "9989  -67.943860 -36.262430\n",
       "9990  -66.721974 -35.792575\n",
       "9991  -67.957720 -35.626207\n",
       "9992  -68.531987 -34.892304\n",
       "9993  -68.036708 -34.475120\n",
       "9994  -69.109629 -34.392904\n",
       "9995  -69.369302 -34.677581\n",
       "9996  -69.953834 -36.628048\n",
       "9997  -69.380250 -37.220133\n",
       "9998  -69.382725 -36.422765\n",
       "9999  -69.528264 -35.077929\n",
       "10000 -68.787819 -35.410307\n",
       "\n",
       "[10001 rows x 2 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(Dict(\"br1\" => pd.Series(brown(10_000)), \"br2\" => pd.Series(brown(10_000))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "feather.write_dataframe(df, \"/tmp/br.feather\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RCall.RObject{RCall.VecSxp}\n",
       "Source: local data frame [10,001 x 2]\n",
       "\n",
       "           br1          br2\n",
       "         <dbl>        <dbl>\n",
       "1   0.00000000  0.000000000\n",
       "2  -0.77383429 -0.006074936\n",
       "3  -0.53758071 -0.281177366\n",
       "4   0.07023901  0.989619762\n",
       "5   0.91392200  0.272461480\n",
       "6   0.78325339  0.943957326\n",
       "7   1.84716072  0.796047645\n",
       "8   0.44813220  1.478668055\n",
       "9  -2.35873065  0.106408928\n",
       "10 -3.03755867 -0.322209156\n",
       "..         ...          ...\n"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "R\"\"\"\n",
    "library(feather)\n",
    "read_feather(\"/tmp/br.feather\")\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Julia 1.1.0",
   "language": "julia",
   "name": "julia-1.1"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "0.4.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}