13_extras.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to DataFrames\n",
    "**[Bogumił Kamiński](http://bogumilkaminski.pl/about/), April 27, 2019**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "using DataFrames"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extras - selected functionalities of selected packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4×3 Named Array{Int64,2}\n",
       "a ╲ b │  x   y   z\n",
       "──────┼───────────\n",
       "'a'   │ 71  87  98\n",
       "'b'   │ 96  80  83\n",
       "'c'   │ 83  75  79\n",
       "'d'   │ 75  85  88"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using FreqTables\n",
    "df = DataFrame(a=rand('a':'d', 1000), b=rand([\"x\", \"y\", \"z\"], 1000))\n",
    "ft = freqtable(df, :a, :b) # observe that dimensions are sorted if possible"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(71, 83)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ft[1,1], ft['b', \"z\"] # you can index the result using numbers or names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4×3 Named Array{Float64,2}\n",
       "a ╲ b │        x         y         z\n",
       "──────┼─────────────────────────────\n",
       "'a'   │ 0.277344  0.339844  0.382813\n",
       "'b'   │ 0.370656   0.30888  0.320463\n",
       "'c'   │ 0.350211  0.316456  0.333333\n",
       "'d'   │ 0.302419  0.342742  0.354839"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prop(ft, 1) # getting proportions - 1 means we want to calculate them in rows (first dimension)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4×3 Named Array{Float64,2}\n",
       "a ╲ b │        x         y         z\n",
       "──────┼─────────────────────────────\n",
       "'a'   │ 0.218462  0.266055  0.281609\n",
       "'b'   │ 0.295385  0.244648  0.238506\n",
       "'c'   │ 0.255385  0.229358  0.227011\n",
       "'d'   │ 0.230769  0.259939  0.252874"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prop(ft, 2) # and columns are normalized to 1.0 now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4-element Named Array{Int64,1}\n",
       "Dim1  │ \n",
       "──────┼──\n",
       "3     │ 4\n",
       "1     │ 2\n",
       "2     │ 4\n",
       "4     │ 0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = categorical(rand(1:3, 10))\n",
    "levels!(x, [3, 1, 2, 4]) # reordering levels and adding an extra level\n",
    "freqtable(x) # order is preserved and not-used level is shown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4-element Named Array{Int64,1}\n",
       "Dim1    │ \n",
       "────────┼──\n",
       "1       │ 2\n",
       "2       │ 1\n",
       "3       │ 1\n",
       "missing │ 1"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freqtable([1,1,2,3,missing]) # by default missings are listed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3-element Named Array{Int64,1}\n",
       "Dim1  │ \n",
       "──────┼──\n",
       "1     │ 2\n",
       "2     │ 1\n",
       "3     │ 1"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freqtable([1,1,2,3,missing], skipmissing=true) # but we can skip them"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2×2 Named Array{Int64,2}\n",
       "a ╲ b │   5    6\n",
       "──────┼─────────\n",
       "3     │ 258  258\n",
       "4     │ 254  230"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = DataFrame(a=rand(3:4, 1000), b=rand(5:6, 1000))\n",
    "ft = freqtable(df, :a, :b) # now dimensions are numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "ename": "BoundsError",
     "evalue": "BoundsError: attempt to access 2×2 Array{Int64,2} at index [3, 5]",
     "output_type": "error",
     "traceback": [
      "BoundsError: attempt to access 2×2 Array{Int64,2} at index [3, 5]",
      "",
      "Stacktrace:",
      " [1] getindex at .\\array.jl:730 [inlined]",
      " [2] getindex(::NamedArrays.NamedArray{Int64,2,Array{Int64,2},Tuple{OrderedCollections.OrderedDict{Int64,Int64},OrderedCollections.OrderedDict{Int64,Int64}}}, ::Int64, ::Int64) at C:\\Users\\bogum\\.julia\\packages\\NamedArrays\\uMdmN\\src\\index.jl:17",
      " [3] top-level scope at In[10]:1"
     ]
    }
   ],
   "source": [
    "ft[3,5] # this is an error - standard array indexing takes precedence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "258"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using NamedArrays\n",
    "ft[Name(3), Name(5)] # you have to use Name() from NamedArrays"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DataFramesMeta - working on `DataFrame`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>x</th><th>y</th><th>z</th></tr><tr><th></th><th>Int64</th><th>Char</th><th>Bool</th></tr></thead><tbody><p>8 rows × 3 columns</p><tr><th>1</th><td>1</td><td>'a'</td><td>true</td></tr><tr><th>2</th><td>2</td><td>'b'</td><td>false</td></tr><tr><th>3</th><td>3</td><td>'c'</td><td>true</td></tr><tr><th>4</th><td>4</td><td>'d'</td><td>false</td></tr><tr><th>5</th><td>5</td><td>'e'</td><td>true</td></tr><tr><th>6</th><td>6</td><td>'f'</td><td>false</td></tr><tr><th>7</th><td>7</td><td>'g'</td><td>true</td></tr><tr><th>8</th><td>8</td><td>'h'</td><td>false</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& x & y & z\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char & Bool\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' & true \\\\\n",
       "\t2 & 2 & 'b' & false \\\\\n",
       "\t3 & 3 & 'c' & true \\\\\n",
       "\t4 & 4 & 'd' & false \\\\\n",
       "\t5 & 5 & 'e' & true \\\\\n",
       "\t6 & 6 & 'f' & false \\\\\n",
       "\t7 & 7 & 'g' & true \\\\\n",
       "\t8 & 8 & 'h' & false \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "8×3 DataFrame\n",
       "│ Row │ x     │ y    │ z     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │ \u001b[90mBool\u001b[39m  │\n",
       "├─────┼───────┼──────┼───────┤\n",
       "│ 1   │ 1     │ 'a'  │ true  │\n",
       "│ 2   │ 2     │ 'b'  │ false │\n",
       "│ 3   │ 3     │ 'c'  │ true  │\n",
       "│ 4   │ 4     │ 'd'  │ false │\n",
       "│ 5   │ 5     │ 'e'  │ true  │\n",
       "│ 6   │ 6     │ 'f'  │ false │\n",
       "│ 7   │ 7     │ 'g'  │ true  │\n",
       "│ 8   │ 8     │ 'h'  │ false │"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using DataFramesMeta\n",
    "df = DataFrame(x=1:8, y='a':'h', z=repeat([true,false], outer=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8-element Array{Int64,1}:\n",
       " 2\n",
       " 2\n",
       " 4\n",
       " 4\n",
       " 6\n",
       " 6\n",
       " 8\n",
       " 8"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@with(df, :x+:z) # expressions with columns of DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8-element Array{Char,1}:\n",
       " 'b'\n",
       " 'e'\n",
       " 'h'\n",
       " 'k'\n",
       " 'g'\n",
       " 'j'\n",
       " 'm'\n",
       " 'p'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@with df begin # you can define code blocks\n",
    "    a = :x[:z]\n",
    "    b = :x[.!:z]\n",
    "    :y + [a; b]\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "ename": "UndefVarError",
     "evalue": "UndefVarError: a not defined",
     "output_type": "error",
     "traceback": [
      "UndefVarError: a not defined",
      "",
      "Stacktrace:",
      " [1] top-level scope at In[15]:1"
     ]
    }
   ],
   "source": [
    "a # @with creates hard scope so variables do not leak out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3-element BitArray{1}:\n",
       "  true\n",
       " false\n",
       " false"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = DataFrame(a = [:a, :b, :c])\n",
    "@with(df2, :a .== ^(:a)) # sometimes we want to work on a raw Symbol, ^() escapes it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>y</th><th>z</th></tr><tr><th></th><th>Int64</th><th>Int64</th></tr></thead><tbody><p>3 rows × 2 columns</p><tr><th>1</th><td>4</td><td>7</td></tr><tr><th>2</th><td>5</td><td>8</td></tr><tr><th>3</th><td>6</td><td>9</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& y & z\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 4 & 7 \\\\\n",
       "\t2 & 5 & 8 \\\\\n",
       "\t3 & 6 & 9 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "3×2 DataFrame\n",
       "│ Row │ y     │ z     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼───────┼───────┤\n",
       "│ 1   │ 4     │ 7     │\n",
       "│ 2   │ 5     │ 8     │\n",
       "│ 3   │ 6     │ 9     │"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = DataFrame(x=1:3, y=4:6, z=7:9)\n",
    "@with(df2, cols(2:3)) # cols(expression) is translated to df2[expression]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>x</th><th>y</th><th>z</th></tr><tr><th></th><th>Int64</th><th>Char</th><th>Bool</th></tr></thead><tbody><p>2 rows × 3 columns</p><tr><th>1</th><td>1</td><td>'a'</td><td>true</td></tr><tr><th>2</th><td>3</td><td>'c'</td><td>true</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& x & y & z\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char & Bool\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' & true \\\\\n",
       "\t2 & 3 & 'c' & true \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "2×3 DataFrame\n",
       "│ Row │ x     │ y    │ z    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │ \u001b[90mBool\u001b[39m │\n",
       "├─────┼───────┼──────┼──────┤\n",
       "│ 1   │ 1     │ 'a'  │ true │\n",
       "│ 2   │ 3     │ 'c'  │ true │"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@where(df, :x .< 4, :z .== true) # a very useful macro for filtering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>x</th><th>y</th><th>z</th></tr><tr><th></th><th>Int64</th><th>Int64</th><th>Char</th></tr></thead><tbody><p>8 rows × 3 columns</p><tr><th>1</th><td>1</td><td>2</td><td>'a'</td></tr><tr><th>2</th><td>2</td><td>4</td><td>'b'</td></tr><tr><th>3</th><td>3</td><td>6</td><td>'c'</td></tr><tr><th>4</th><td>4</td><td>8</td><td>'d'</td></tr><tr><th>5</th><td>5</td><td>10</td><td>'e'</td></tr><tr><th>6</th><td>6</td><td>12</td><td>'f'</td></tr><tr><th>7</th><td>7</td><td>14</td><td>'g'</td></tr><tr><th>8</th><td>8</td><td>16</td><td>'h'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& x & y & z\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 2 & 'a' \\\\\n",
       "\t2 & 2 & 4 & 'b' \\\\\n",
       "\t3 & 3 & 6 & 'c' \\\\\n",
       "\t4 & 4 & 8 & 'd' \\\\\n",
       "\t5 & 5 & 10 & 'e' \\\\\n",
       "\t6 & 6 & 12 & 'f' \\\\\n",
       "\t7 & 7 & 14 & 'g' \\\\\n",
       "\t8 & 8 & 16 & 'h' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "8×3 DataFrame\n",
       "│ Row │ x     │ y     │ z    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼───────┼──────┤\n",
       "│ 1   │ 1     │ 2     │ 'a'  │\n",
       "│ 2   │ 2     │ 4     │ 'b'  │\n",
       "│ 3   │ 3     │ 6     │ 'c'  │\n",
       "│ 4   │ 4     │ 8     │ 'd'  │\n",
       "│ 5   │ 5     │ 10    │ 'e'  │\n",
       "│ 6   │ 6     │ 12    │ 'f'  │\n",
       "│ 7   │ 7     │ 14    │ 'g'  │\n",
       "│ 8   │ 8     │ 16    │ 'h'  │"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@select(df, :x, y = 2*:x, z=:y) # create a new DataFrame based on the old one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>x</th><th>y</th><th>z</th><th>a</th></tr><tr><th></th><th>Int64</th><th>Int64</th><th>Bool</th><th>Int64</th></tr></thead><tbody><p>8 rows × 4 columns</p><tr><th>1</th><td>2</td><td>1</td><td>true</td><td>1</td></tr><tr><th>2</th><td>4</td><td>2</td><td>false</td><td>1</td></tr><tr><th>3</th><td>6</td><td>3</td><td>true</td><td>1</td></tr><tr><th>4</th><td>8</td><td>4</td><td>false</td><td>1</td></tr><tr><th>5</th><td>10</td><td>5</td><td>true</td><td>1</td></tr><tr><th>6</th><td>12</td><td>6</td><td>false</td><td>1</td></tr><tr><th>7</th><td>14</td><td>7</td><td>true</td><td>1</td></tr><tr><th>8</th><td>16</td><td>8</td><td>false</td><td>1</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cccc}\n",
       "\t& x & y & z & a\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64 & Bool & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 2 & 1 & true & 1 \\\\\n",
       "\t2 & 4 & 2 & false & 1 \\\\\n",
       "\t3 & 6 & 3 & true & 1 \\\\\n",
       "\t4 & 8 & 4 & false & 1 \\\\\n",
       "\t5 & 10 & 5 & true & 1 \\\\\n",
       "\t6 & 12 & 6 & false & 1 \\\\\n",
       "\t7 & 14 & 7 & true & 1 \\\\\n",
       "\t8 & 16 & 8 & false & 1 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "8×4 DataFrame\n",
       "│ Row │ x     │ y     │ z     │ a     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mBool\u001b[39m  │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼───────┼───────┼───────┼───────┤\n",
       "│ 1   │ 2     │ 1     │ true  │ 1     │\n",
       "│ 2   │ 4     │ 2     │ false │ 1     │\n",
       "│ 3   │ 6     │ 3     │ true  │ 1     │\n",
       "│ 4   │ 8     │ 4     │ false │ 1     │\n",
       "│ 5   │ 10    │ 5     │ true  │ 1     │\n",
       "│ 6   │ 12    │ 6     │ false │ 1     │\n",
       "│ 7   │ 14    │ 7     │ true  │ 1     │\n",
       "│ 8   │ 16    │ 8     │ false │ 1     │"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@transform(df, a=1, x = 2*:x, y=:x) # create a new DataFrame adding columns based on the old one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "ename": "ArgumentError",
     "evalue": "ArgumentError: column name :a not found in the data frame; existing most similar names are: :x, :y and :z",
     "output_type": "error",
     "traceback": [
      "ArgumentError: column name :a not found in the data frame; existing most similar names are: :x, :y and :z",
      "",
      "Stacktrace:",
      " [1] lookupname at C:\\Users\\bogum\\.julia\\packages\\DataFrames\\Wvy9E\\src\\other\\index.jl:170 [inlined]",
      " [2] getindex at C:\\Users\\bogum\\.julia\\packages\\DataFrames\\Wvy9E\\src\\other\\index.jl:176 [inlined]",
      " [3] getindex(::DataFrame, ::Symbol) at C:\\Users\\bogum\\.julia\\packages\\DataFrames\\Wvy9E\\src\\dataframe\\dataframe.jl:295",
      " [4] (::getfield(Main, Symbol(\"##16#18\")))(::DataFrame) at C:\\Users\\bogum\\.julia\\packages\\DataFramesMeta\\Ykidf\\src\\DataFramesMeta.jl:73",
      " [5] #transform#9(::Base.Iterators.Pairs{Symbol,Function,Tuple{Symbol,Symbol},NamedTuple{(:a, :b),Tuple{getfield(Main, Symbol(\"##15#17\")),getfield(Main, Symbol(\"##16#18\"))}}}, ::Function, ::DataFrame) at C:\\Users\\bogum\\.julia\\packages\\DataFramesMeta\\Ykidf\\src\\DataFramesMeta.jl:384",
      " [6] (::getfield(DataFramesMeta, Symbol(\"#kw##transform\")))(::NamedTuple{(:a, :b),Tuple{getfield(Main, Symbol(\"##15#17\")),getfield(Main, Symbol(\"##16#18\"))}}, ::typeof(DataFramesMeta.transform), ::DataFrame) at .\\none:0",
      " [7] top-level scope at C:\\Users\\bogum\\.julia\\packages\\DataFramesMeta\\Ykidf\\src\\DataFramesMeta.jl:478",
      " [8] top-level scope at In[21]:1"
     ]
    }
   ],
   "source": [
    "@transform(df, a=1, b=:a) # old DataFrame is used and :a is not present there"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>x</th><th>y</th><th>z</th></tr><tr><th></th><th>Int64</th><th>Char</th><th>Bool</th></tr></thead><tbody><p>8 rows × 3 columns</p><tr><th>1</th><td>8</td><td>'h'</td><td>false</td></tr><tr><th>2</th><td>6</td><td>'f'</td><td>false</td></tr><tr><th>3</th><td>4</td><td>'d'</td><td>false</td></tr><tr><th>4</th><td>2</td><td>'b'</td><td>false</td></tr><tr><th>5</th><td>7</td><td>'g'</td><td>true</td></tr><tr><th>6</th><td>5</td><td>'e'</td><td>true</td></tr><tr><th>7</th><td>3</td><td>'c'</td><td>true</td></tr><tr><th>8</th><td>1</td><td>'a'</td><td>true</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& x & y & z\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char & Bool\\\\\n",
       "\t\\hline\n",
       "\t1 & 8 & 'h' & false \\\\\n",
       "\t2 & 6 & 'f' & false \\\\\n",
       "\t3 & 4 & 'd' & false \\\\\n",
       "\t4 & 2 & 'b' & false \\\\\n",
       "\t5 & 7 & 'g' & true \\\\\n",
       "\t6 & 5 & 'e' & true \\\\\n",
       "\t7 & 3 & 'c' & true \\\\\n",
       "\t8 & 1 & 'a' & true \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "8×3 DataFrame\n",
       "│ Row │ x     │ y    │ z     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │ \u001b[90mBool\u001b[39m  │\n",
       "├─────┼───────┼──────┼───────┤\n",
       "│ 1   │ 8     │ 'h'  │ false │\n",
       "│ 2   │ 6     │ 'f'  │ false │\n",
       "│ 3   │ 4     │ 'd'  │ false │\n",
       "│ 4   │ 2     │ 'b'  │ false │\n",
       "│ 5   │ 7     │ 'g'  │ true  │\n",
       "│ 6   │ 5     │ 'e'  │ true  │\n",
       "│ 7   │ 3     │ 'c'  │ true  │\n",
       "│ 8   │ 1     │ 'a'  │ true  │"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@orderby(df, :z, -:x) # sorting into a new data frame, less powerful than sort, but lightweight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>z</th><th>x</th><th>x²</th></tr><tr><th></th><th>Bool</th><th>Int64</th><th>Int64</th></tr></thead><tbody><p>4 rows × 3 columns</p><tr><th>1</th><td>false</td><td>2</td><td>4</td></tr><tr><th>2</th><td>false</td><td>4</td><td>16</td></tr><tr><th>3</th><td>true</td><td>1</td><td>1</td></tr><tr><th>4</th><td>true</td><td>3</td><td>9</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& z & x & x²\\\\\n",
       "\t\\hline\n",
       "\t& Bool & Int64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & false & 2 & 4 \\\\\n",
       "\t2 & false & 4 & 16 \\\\\n",
       "\t3 & true & 1 & 1 \\\\\n",
       "\t4 & true & 3 & 9 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "4×3 DataFrame\n",
       "│ Row │ z     │ x     │ x²    │\n",
       "│     │ \u001b[90mBool\u001b[39m  │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼───────┼───────┼───────┤\n",
       "│ 1   │ false │ 2     │ 4     │\n",
       "│ 2   │ false │ 4     │ 16    │\n",
       "│ 3   │ true  │ 1     │ 1     │\n",
       "│ 4   │ true  │ 3     │ 9     │"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@linq df |> # chaining of operations on DataFrame\n",
    "    where(:x .< 5) |>\n",
    "    orderby(:z) |>\n",
    "    transform(x²=:x.^2) |>\n",
    "    select(:z, :x, :x²)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4-element Array{Int64,1}:\n",
       " 1\n",
       " 2\n",
       " 3\n",
       " 4"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f(df, col) = df[col] # you can define your own functions and put them in the chain\n",
    "@linq df |> where(:x .<= 4) |> f(:x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DataFramesMeta - working on grouped `DataFrame`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><b>GroupedDataFrame with 4 groups based on key: b</b></p><p><i>First Group (3 rows): b = 'a'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>1</td><td>'a'</td></tr><tr><th>2</th><td>5</td><td>'a'</td></tr><tr><th>3</th><td>9</td><td>'a'</td></tr></tbody></table><p>&vellip;</p><p><i>Last Group (3 rows): b = 'd'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>4</td><td>'d'</td></tr><tr><th>2</th><td>8</td><td>'d'</td></tr><tr><th>3</th><td>12</td><td>'d'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "GroupedDataFrame with 4 groups based on key: b\n",
       "\n",
       "First Group (3 rows): b = 'a'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' \\\\\n",
       "\t2 & 5 & 'a' \\\\\n",
       "\t3 & 9 & 'a' \\\\\n",
       "\\end{tabular}\n",
       "\n",
       "$\\dots$\n",
       "\n",
       "Last Group (3 rows): b = 'd'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 4 & 'd' \\\\\n",
       "\t2 & 8 & 'd' \\\\\n",
       "\t3 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "GroupedDataFrame with 4 groups based on key: b\n",
       "First Group (3 rows): b = 'a'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 1     │ 'a'  │\n",
       "│ 2   │ 5     │ 'a'  │\n",
       "│ 3   │ 9     │ 'a'  │\n",
       "⋮\n",
       "Last Group (3 rows): b = 'd'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 4     │ 'd'  │\n",
       "│ 2   │ 8     │ 'd'  │\n",
       "│ 3   │ 12    │ 'd'  │"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = DataFrame(a = 1:12, b = repeat('a':'d', outer=3))\n",
    "g = groupby(df, :b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>b</th><th>first</th><th>last</th><th>mean</th></tr><tr><th></th><th>Char</th><th>Int64</th><th>Int64</th><th>Float64</th></tr></thead><tbody><p>4 rows × 4 columns</p><tr><th>1</th><td>'a'</td><td>1</td><td>9</td><td>5.0</td></tr><tr><th>2</th><td>'b'</td><td>2</td><td>10</td><td>6.0</td></tr><tr><th>3</th><td>'c'</td><td>3</td><td>11</td><td>7.0</td></tr><tr><th>4</th><td>'d'</td><td>4</td><td>12</td><td>8.0</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cccc}\n",
       "\t& b & first & last & mean\\\\\n",
       "\t\\hline\n",
       "\t& Char & Int64 & Int64 & Float64\\\\\n",
       "\t\\hline\n",
       "\t1 & 'a' & 1 & 9 & 5.0 \\\\\n",
       "\t2 & 'b' & 2 & 10 & 6.0 \\\\\n",
       "\t3 & 'c' & 3 & 11 & 7.0 \\\\\n",
       "\t4 & 'd' & 4 & 12 & 8.0 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "4×4 DataFrame\n",
       "│ Row │ b    │ first │ last  │ mean    │\n",
       "│     │ \u001b[90mChar\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mFloat64\u001b[39m │\n",
       "├─────┼──────┼───────┼───────┼─────────┤\n",
       "│ 1   │ 'a'  │ 1     │ 9     │ 5.0     │\n",
       "│ 2   │ 'b'  │ 2     │ 10    │ 6.0     │\n",
       "│ 3   │ 'c'  │ 3     │ 11    │ 7.0     │\n",
       "│ 4   │ 'd'  │ 4     │ 12    │ 8.0     │"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using Statistics\n",
    "@by(df, :b, first=first(:a), last=last(:a), mean=mean(:a)) # similar to DataFrames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>b</th><th>first</th><th>last</th><th>mean</th></tr><tr><th></th><th>Char</th><th>Int64</th><th>Int64</th><th>Float64</th></tr></thead><tbody><p>4 rows × 4 columns</p><tr><th>1</th><td>'a'</td><td>1</td><td>9</td><td>5.0</td></tr><tr><th>2</th><td>'b'</td><td>2</td><td>10</td><td>6.0</td></tr><tr><th>3</th><td>'c'</td><td>3</td><td>11</td><td>7.0</td></tr><tr><th>4</th><td>'d'</td><td>4</td><td>12</td><td>8.0</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cccc}\n",
       "\t& b & first & last & mean\\\\\n",
       "\t\\hline\n",
       "\t& Char & Int64 & Int64 & Float64\\\\\n",
       "\t\\hline\n",
       "\t1 & 'a' & 1 & 9 & 5.0 \\\\\n",
       "\t2 & 'b' & 2 & 10 & 6.0 \\\\\n",
       "\t3 & 'c' & 3 & 11 & 7.0 \\\\\n",
       "\t4 & 'd' & 4 & 12 & 8.0 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "4×4 DataFrame\n",
       "│ Row │ b    │ first │ last  │ mean    │\n",
       "│     │ \u001b[90mChar\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mFloat64\u001b[39m │\n",
       "├─────┼──────┼───────┼───────┼─────────┤\n",
       "│ 1   │ 'a'  │ 1     │ 9     │ 5.0     │\n",
       "│ 2   │ 'b'  │ 2     │ 10    │ 6.0     │\n",
       "│ 3   │ 'c'  │ 3     │ 11    │ 7.0     │\n",
       "│ 4   │ 'd'  │ 4     │ 12    │ 8.0     │"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@based_on(g, first=first(:a), last=last(:a), mean=mean(:a)) # the same as by but on grouped DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><b>GroupedDataFrame with 2 groups based on key: b</b></p><p><i>First Group (3 rows): b = 'c'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>3</td><td>'c'</td></tr><tr><th>2</th><td>7</td><td>'c'</td></tr><tr><th>3</th><td>11</td><td>'c'</td></tr></tbody></table><p>&vellip;</p><p><i>Last Group (3 rows): b = 'd'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>4</td><td>'d'</td></tr><tr><th>2</th><td>8</td><td>'d'</td></tr><tr><th>3</th><td>12</td><td>'d'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "GroupedDataFrame with 2 groups based on key: b\n",
       "\n",
       "First Group (3 rows): b = 'c'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 3 & 'c' \\\\\n",
       "\t2 & 7 & 'c' \\\\\n",
       "\t3 & 11 & 'c' \\\\\n",
       "\\end{tabular}\n",
       "\n",
       "$\\dots$\n",
       "\n",
       "Last Group (3 rows): b = 'd'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 4 & 'd' \\\\\n",
       "\t2 & 8 & 'd' \\\\\n",
       "\t3 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "GroupedDataFrame with 2 groups based on key: b\n",
       "First Group (3 rows): b = 'c'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 3     │ 'c'  │\n",
       "│ 2   │ 7     │ 'c'  │\n",
       "│ 3   │ 11    │ 'c'  │\n",
       "⋮\n",
       "Last Group (3 rows): b = 'd'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 4     │ 'd'  │\n",
       "│ 2   │ 8     │ 'd'  │\n",
       "│ 3   │ 12    │ 'd'  │"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@where(g, mean(:a) > 6.5) # filter gropus on aggregate conditions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><b>GroupedDataFrame with 4 groups based on key: b</b></p><p><i>First Group (3 rows): b = 'd'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>4</td><td>'d'</td></tr><tr><th>2</th><td>8</td><td>'d'</td></tr><tr><th>3</th><td>12</td><td>'d'</td></tr></tbody></table><p>&vellip;</p><p><i>Last Group (3 rows): b = 'a'</i></p><table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><tr><th>1</th><td>1</td><td>'a'</td></tr><tr><th>2</th><td>5</td><td>'a'</td></tr><tr><th>3</th><td>9</td><td>'a'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "GroupedDataFrame with 4 groups based on key: b\n",
       "\n",
       "First Group (3 rows): b = 'd'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 4 & 'd' \\\\\n",
       "\t2 & 8 & 'd' \\\\\n",
       "\t3 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n",
       "\n",
       "$\\dots$\n",
       "\n",
       "Last Group (3 rows): b = 'a'\n",
       "\n",
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' \\\\\n",
       "\t2 & 5 & 'a' \\\\\n",
       "\t3 & 9 & 'a' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "GroupedDataFrame with 4 groups based on key: b\n",
       "First Group (3 rows): b = 'd'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 4     │ 'd'  │\n",
       "│ 2   │ 8     │ 'd'  │\n",
       "│ 3   │ 12    │ 'd'  │\n",
       "⋮\n",
       "Last Group (3 rows): b = 'a'\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 1     │ 'a'  │\n",
       "│ 2   │ 5     │ 'a'  │\n",
       "│ 3   │ 9     │ 'a'  │"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@orderby(g, -sum(:a)) # order groups on aggregate conditions\n",
    "# this fails but should be fixed in the future"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th><th>center</th><th>centered</th></tr><tr><th></th><th>Int64</th><th>Char</th><th>Float64</th><th>Float64</th></tr></thead><tbody><p>12 rows × 4 columns</p><tr><th>1</th><td>1</td><td>'a'</td><td>5.0</td><td>-4.0</td></tr><tr><th>2</th><td>5</td><td>'a'</td><td>5.0</td><td>0.0</td></tr><tr><th>3</th><td>9</td><td>'a'</td><td>5.0</td><td>4.0</td></tr><tr><th>4</th><td>2</td><td>'b'</td><td>6.0</td><td>-4.0</td></tr><tr><th>5</th><td>6</td><td>'b'</td><td>6.0</td><td>0.0</td></tr><tr><th>6</th><td>10</td><td>'b'</td><td>6.0</td><td>4.0</td></tr><tr><th>7</th><td>3</td><td>'c'</td><td>7.0</td><td>-4.0</td></tr><tr><th>8</th><td>7</td><td>'c'</td><td>7.0</td><td>0.0</td></tr><tr><th>9</th><td>11</td><td>'c'</td><td>7.0</td><td>4.0</td></tr><tr><th>10</th><td>4</td><td>'d'</td><td>8.0</td><td>-4.0</td></tr><tr><th>11</th><td>8</td><td>'d'</td><td>8.0</td><td>0.0</td></tr><tr><th>12</th><td>12</td><td>'d'</td><td>8.0</td><td>4.0</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cccc}\n",
       "\t& a & b & center & centered\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char & Float64 & Float64\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' & 5.0 & -4.0 \\\\\n",
       "\t2 & 5 & 'a' & 5.0 & 0.0 \\\\\n",
       "\t3 & 9 & 'a' & 5.0 & 4.0 \\\\\n",
       "\t4 & 2 & 'b' & 6.0 & -4.0 \\\\\n",
       "\t5 & 6 & 'b' & 6.0 & 0.0 \\\\\n",
       "\t6 & 10 & 'b' & 6.0 & 4.0 \\\\\n",
       "\t7 & 3 & 'c' & 7.0 & -4.0 \\\\\n",
       "\t8 & 7 & 'c' & 7.0 & 0.0 \\\\\n",
       "\t9 & 11 & 'c' & 7.0 & 4.0 \\\\\n",
       "\t10 & 4 & 'd' & 8.0 & -4.0 \\\\\n",
       "\t11 & 8 & 'd' & 8.0 & 0.0 \\\\\n",
       "\t12 & 12 & 'd' & 8.0 & 4.0 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×4 DataFrame\n",
       "│ Row │ a     │ b    │ center  │ centered │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │ \u001b[90mFloat64\u001b[39m │ \u001b[90mFloat64\u001b[39m  │\n",
       "├─────┼───────┼──────┼─────────┼──────────┤\n",
       "│ 1   │ 1     │ 'a'  │ 5.0     │ -4.0     │\n",
       "│ 2   │ 5     │ 'a'  │ 5.0     │ 0.0      │\n",
       "│ 3   │ 9     │ 'a'  │ 5.0     │ 4.0      │\n",
       "│ 4   │ 2     │ 'b'  │ 6.0     │ -4.0     │\n",
       "│ 5   │ 6     │ 'b'  │ 6.0     │ 0.0      │\n",
       "│ 6   │ 10    │ 'b'  │ 6.0     │ 4.0      │\n",
       "│ 7   │ 3     │ 'c'  │ 7.0     │ -4.0     │\n",
       "│ 8   │ 7     │ 'c'  │ 7.0     │ 0.0      │\n",
       "│ 9   │ 11    │ 'c'  │ 7.0     │ 4.0      │\n",
       "│ 10  │ 4     │ 'd'  │ 8.0     │ -4.0     │\n",
       "│ 11  │ 8     │ 'd'  │ 8.0     │ 0.0      │\n",
       "│ 12  │ 12    │ 'd'  │ 8.0     │ 4.0      │"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@transform(g, center = mean(:a), centered = :a .- mean(:a)) # perform operations within a group and return ungroped DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><p>12 rows × 2 columns</p><tr><th>1</th><td>1</td><td>'a'</td></tr><tr><th>2</th><td>5</td><td>'a'</td></tr><tr><th>3</th><td>9</td><td>'a'</td></tr><tr><th>4</th><td>2</td><td>'b'</td></tr><tr><th>5</th><td>6</td><td>'b'</td></tr><tr><th>6</th><td>10</td><td>'b'</td></tr><tr><th>7</th><td>3</td><td>'c'</td></tr><tr><th>8</th><td>7</td><td>'c'</td></tr><tr><th>9</th><td>11</td><td>'c'</td></tr><tr><th>10</th><td>4</td><td>'d'</td></tr><tr><th>11</th><td>8</td><td>'d'</td></tr><tr><th>12</th><td>12</td><td>'d'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' \\\\\n",
       "\t2 & 5 & 'a' \\\\\n",
       "\t3 & 9 & 'a' \\\\\n",
       "\t4 & 2 & 'b' \\\\\n",
       "\t5 & 6 & 'b' \\\\\n",
       "\t6 & 10 & 'b' \\\\\n",
       "\t7 & 3 & 'c' \\\\\n",
       "\t8 & 7 & 'c' \\\\\n",
       "\t9 & 11 & 'c' \\\\\n",
       "\t10 & 4 & 'd' \\\\\n",
       "\t11 & 8 & 'd' \\\\\n",
       "\t12 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×2 DataFrame\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 1     │ 'a'  │\n",
       "│ 2   │ 5     │ 'a'  │\n",
       "│ 3   │ 9     │ 'a'  │\n",
       "│ 4   │ 2     │ 'b'  │\n",
       "│ 5   │ 6     │ 'b'  │\n",
       "│ 6   │ 10    │ 'b'  │\n",
       "│ 7   │ 3     │ 'c'  │\n",
       "│ 8   │ 7     │ 'c'  │\n",
       "│ 9   │ 11    │ 'c'  │\n",
       "│ 10  │ 4     │ 'd'  │\n",
       "│ 11  │ 8     │ 'd'  │\n",
       "│ 12  │ 12    │ 'd'  │"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DataFrame(g) # convinience equivalent to collect(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><p>12 rows × 2 columns</p><tr><th>1</th><td>1</td><td>'a'</td></tr><tr><th>2</th><td>5</td><td>'a'</td></tr><tr><th>3</th><td>9</td><td>'a'</td></tr><tr><th>4</th><td>2</td><td>'b'</td></tr><tr><th>5</th><td>6</td><td>'b'</td></tr><tr><th>6</th><td>10</td><td>'b'</td></tr><tr><th>7</th><td>3</td><td>'c'</td></tr><tr><th>8</th><td>7</td><td>'c'</td></tr><tr><th>9</th><td>11</td><td>'c'</td></tr><tr><th>10</th><td>4</td><td>'d'</td></tr><tr><th>11</th><td>8</td><td>'d'</td></tr><tr><th>12</th><td>12</td><td>'d'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 'a' \\\\\n",
       "\t2 & 5 & 'a' \\\\\n",
       "\t3 & 9 & 'a' \\\\\n",
       "\t4 & 2 & 'b' \\\\\n",
       "\t5 & 6 & 'b' \\\\\n",
       "\t6 & 10 & 'b' \\\\\n",
       "\t7 & 3 & 'c' \\\\\n",
       "\t8 & 7 & 'c' \\\\\n",
       "\t9 & 11 & 'c' \\\\\n",
       "\t10 & 4 & 'd' \\\\\n",
       "\t11 & 8 & 'd' \\\\\n",
       "\t12 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×2 DataFrame\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 1     │ 'a'  │\n",
       "│ 2   │ 5     │ 'a'  │\n",
       "│ 3   │ 9     │ 'a'  │\n",
       "│ 4   │ 2     │ 'b'  │\n",
       "│ 5   │ 6     │ 'b'  │\n",
       "│ 6   │ 10    │ 'b'  │\n",
       "│ 7   │ 3     │ 'c'  │\n",
       "│ 8   │ 7     │ 'c'  │\n",
       "│ 9   │ 11    │ 'c'  │\n",
       "│ 10  │ 4     │ 'd'  │\n",
       "│ 11  │ 8     │ 'd'  │\n",
       "│ 12  │ 12    │ 'd'  │"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@transform(g) # actually this is the same as DataFrame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Char</th></tr></thead><tbody><p>6 rows × 2 columns</p><tr><th>1</th><td>3</td><td>'c'</td></tr><tr><th>2</th><td>7</td><td>'c'</td></tr><tr><th>3</th><td>11</td><td>'c'</td></tr><tr><th>4</th><td>4</td><td>'d'</td></tr><tr><th>5</th><td>8</td><td>'d'</td></tr><tr><th>6</th><td>12</td><td>'d'</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Char\\\\\n",
       "\t\\hline\n",
       "\t1 & 3 & 'c' \\\\\n",
       "\t2 & 7 & 'c' \\\\\n",
       "\t3 & 11 & 'c' \\\\\n",
       "\t4 & 4 & 'd' \\\\\n",
       "\t5 & 8 & 'd' \\\\\n",
       "\t6 & 12 & 'd' \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "6×2 DataFrame\n",
       "│ Row │ a     │ b    │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mChar\u001b[39m │\n",
       "├─────┼───────┼──────┤\n",
       "│ 1   │ 3     │ 'c'  │\n",
       "│ 2   │ 7     │ 'c'  │\n",
       "│ 3   │ 11    │ 'c'  │\n",
       "│ 4   │ 4     │ 'd'  │\n",
       "│ 5   │ 8     │ 'd'  │\n",
       "│ 6   │ 12    │ 'd'  │"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@linq df |> groupby(:b) |> where(mean(:a) > 6.5) |> DataFrame # you can do chaining on grouped DataFrames as well"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DataFramesMeta - rowwise operations on `DataFrame`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th></tr><tr><th></th><th>Int64</th><th>Int64</th></tr></thead><tbody><p>12 rows × 2 columns</p><tr><th>1</th><td>1</td><td>1</td></tr><tr><th>2</th><td>2</td><td>2</td></tr><tr><th>3</th><td>3</td><td>3</td></tr><tr><th>4</th><td>4</td><td>4</td></tr><tr><th>5</th><td>5</td><td>1</td></tr><tr><th>6</th><td>6</td><td>2</td></tr><tr><th>7</th><td>7</td><td>3</td></tr><tr><th>8</th><td>8</td><td>4</td></tr><tr><th>9</th><td>9</td><td>1</td></tr><tr><th>10</th><td>10</td><td>2</td></tr><tr><th>11</th><td>11</td><td>3</td></tr><tr><th>12</th><td>12</td><td>4</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& a & b\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 1 \\\\\n",
       "\t2 & 2 & 2 \\\\\n",
       "\t3 & 3 & 3 \\\\\n",
       "\t4 & 4 & 4 \\\\\n",
       "\t5 & 5 & 1 \\\\\n",
       "\t6 & 6 & 2 \\\\\n",
       "\t7 & 7 & 3 \\\\\n",
       "\t8 & 8 & 4 \\\\\n",
       "\t9 & 9 & 1 \\\\\n",
       "\t10 & 10 & 2 \\\\\n",
       "\t11 & 11 & 3 \\\\\n",
       "\t12 & 12 & 4 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×2 DataFrame\n",
       "│ Row │ a     │ b     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼───────┼───────┤\n",
       "│ 1   │ 1     │ 1     │\n",
       "│ 2   │ 2     │ 2     │\n",
       "│ 3   │ 3     │ 3     │\n",
       "│ 4   │ 4     │ 4     │\n",
       "│ 5   │ 5     │ 1     │\n",
       "│ 6   │ 6     │ 2     │\n",
       "│ 7   │ 7     │ 3     │\n",
       "│ 8   │ 8     │ 4     │\n",
       "│ 9   │ 9     │ 1     │\n",
       "│ 10  │ 10    │ 2     │\n",
       "│ 11  │ 11    │ 3     │\n",
       "│ 12  │ 12    │ 4     │"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = DataFrame(a = 1:12, b = repeat(1:4, outer=3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th><th>x</th></tr><tr><th></th><th>Int64</th><th>Int64</th><th>String</th></tr></thead><tbody><p>12 rows × 3 columns</p><tr><th>1</th><td>1</td><td>1</td><td>no</td></tr><tr><th>2</th><td>2</td><td>2</td><td>no</td></tr><tr><th>3</th><td>3</td><td>3</td><td>no</td></tr><tr><th>4</th><td>4</td><td>4</td><td>no</td></tr><tr><th>5</th><td>5</td><td>1</td><td>no</td></tr><tr><th>6</th><td>6</td><td>2</td><td>no</td></tr><tr><th>7</th><td>7</td><td>3</td><td>no</td></tr><tr><th>8</th><td>8</td><td>4</td><td>yes</td></tr><tr><th>9</th><td>9</td><td>1</td><td>no</td></tr><tr><th>10</th><td>10</td><td>2</td><td>no</td></tr><tr><th>11</th><td>11</td><td>3</td><td>no</td></tr><tr><th>12</th><td>12</td><td>4</td><td>yes</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& a & b & x\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64 & String\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 1 & no \\\\\n",
       "\t2 & 2 & 2 & no \\\\\n",
       "\t3 & 3 & 3 & no \\\\\n",
       "\t4 & 4 & 4 & no \\\\\n",
       "\t5 & 5 & 1 & no \\\\\n",
       "\t6 & 6 & 2 & no \\\\\n",
       "\t7 & 7 & 3 & no \\\\\n",
       "\t8 & 8 & 4 & yes \\\\\n",
       "\t9 & 9 & 1 & no \\\\\n",
       "\t10 & 10 & 2 & no \\\\\n",
       "\t11 & 11 & 3 & no \\\\\n",
       "\t12 & 12 & 4 & yes \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×3 DataFrame\n",
       "│ Row │ a     │ b     │ x      │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mString\u001b[39m │\n",
       "├─────┼───────┼───────┼────────┤\n",
       "│ 1   │ 1     │ 1     │ no     │\n",
       "│ 2   │ 2     │ 2     │ no     │\n",
       "│ 3   │ 3     │ 3     │ no     │\n",
       "│ 4   │ 4     │ 4     │ no     │\n",
       "│ 5   │ 5     │ 1     │ no     │\n",
       "│ 6   │ 6     │ 2     │ no     │\n",
       "│ 7   │ 7     │ 3     │ no     │\n",
       "│ 8   │ 8     │ 4     │ yes    │\n",
       "│ 9   │ 9     │ 1     │ no     │\n",
       "│ 10  │ 10    │ 2     │ no     │\n",
       "│ 11  │ 11    │ 3     │ no     │\n",
       "│ 12  │ 12    │ 4     │ yes    │"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# such conditions are often needed but are complex to write\n",
    "@transform(df, x = ifelse.((:a .> 6) .& (:b .== 4), \"yes\", \"no\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th><th>x</th></tr><tr><th></th><th>Int64</th><th>Int64</th><th>String</th></tr></thead><tbody><p>12 rows × 3 columns</p><tr><th>1</th><td>1</td><td>1</td><td>no</td></tr><tr><th>2</th><td>2</td><td>2</td><td>no</td></tr><tr><th>3</th><td>3</td><td>3</td><td>no</td></tr><tr><th>4</th><td>4</td><td>4</td><td>no</td></tr><tr><th>5</th><td>5</td><td>1</td><td>no</td></tr><tr><th>6</th><td>6</td><td>2</td><td>no</td></tr><tr><th>7</th><td>7</td><td>3</td><td>no</td></tr><tr><th>8</th><td>8</td><td>4</td><td>yes</td></tr><tr><th>9</th><td>9</td><td>1</td><td>no</td></tr><tr><th>10</th><td>10</td><td>2</td><td>no</td></tr><tr><th>11</th><td>11</td><td>3</td><td>no</td></tr><tr><th>12</th><td>12</td><td>4</td><td>yes</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& a & b & x\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64 & String\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 1 & no \\\\\n",
       "\t2 & 2 & 2 & no \\\\\n",
       "\t3 & 3 & 3 & no \\\\\n",
       "\t4 & 4 & 4 & no \\\\\n",
       "\t5 & 5 & 1 & no \\\\\n",
       "\t6 & 6 & 2 & no \\\\\n",
       "\t7 & 7 & 3 & no \\\\\n",
       "\t8 & 8 & 4 & yes \\\\\n",
       "\t9 & 9 & 1 & no \\\\\n",
       "\t10 & 10 & 2 & no \\\\\n",
       "\t11 & 11 & 3 & no \\\\\n",
       "\t12 & 12 & 4 & yes \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×3 DataFrame\n",
       "│ Row │ a     │ b     │ x      │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mString\u001b[39m │\n",
       "├─────┼───────┼───────┼────────┤\n",
       "│ 1   │ 1     │ 1     │ no     │\n",
       "│ 2   │ 2     │ 2     │ no     │\n",
       "│ 3   │ 3     │ 3     │ no     │\n",
       "│ 4   │ 4     │ 4     │ no     │\n",
       "│ 5   │ 5     │ 1     │ no     │\n",
       "│ 6   │ 6     │ 2     │ no     │\n",
       "│ 7   │ 7     │ 3     │ no     │\n",
       "│ 8   │ 8     │ 4     │ yes    │\n",
       "│ 9   │ 9     │ 1     │ no     │\n",
       "│ 10  │ 10    │ 2     │ no     │\n",
       "│ 11  │ 11    │ 3     │ no     │\n",
       "│ 12  │ 12    │ 4     │ yes    │"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# one option is to use a function that works on a single observation and broadcast it\n",
    "myfun(a, b) = a > 6 && b == 4 ? \"yes\" : \"no\"\n",
    "@transform(df, x = myfun.(:a, :b))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>a</th><th>b</th><th>x</th></tr><tr><th></th><th>Int64</th><th>Int64</th><th>String</th></tr></thead><tbody><p>12 rows × 3 columns</p><tr><th>1</th><td>1</td><td>1</td><td>no</td></tr><tr><th>2</th><td>2</td><td>2</td><td>no</td></tr><tr><th>3</th><td>3</td><td>3</td><td>no</td></tr><tr><th>4</th><td>4</td><td>4</td><td>no</td></tr><tr><th>5</th><td>5</td><td>1</td><td>no</td></tr><tr><th>6</th><td>6</td><td>2</td><td>no</td></tr><tr><th>7</th><td>7</td><td>3</td><td>no</td></tr><tr><th>8</th><td>8</td><td>4</td><td>yes</td></tr><tr><th>9</th><td>9</td><td>1</td><td>no</td></tr><tr><th>10</th><td>10</td><td>2</td><td>no</td></tr><tr><th>11</th><td>11</td><td>3</td><td>no</td></tr><tr><th>12</th><td>12</td><td>4</td><td>yes</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& a & b & x\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64 & String\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 1 & no \\\\\n",
       "\t2 & 2 & 2 & no \\\\\n",
       "\t3 & 3 & 3 & no \\\\\n",
       "\t4 & 4 & 4 & no \\\\\n",
       "\t5 & 5 & 1 & no \\\\\n",
       "\t6 & 6 & 2 & no \\\\\n",
       "\t7 & 7 & 3 & no \\\\\n",
       "\t8 & 8 & 4 & yes \\\\\n",
       "\t9 & 9 & 1 & no \\\\\n",
       "\t10 & 10 & 2 & no \\\\\n",
       "\t11 & 11 & 3 & no \\\\\n",
       "\t12 & 12 & 4 & yes \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "12×3 DataFrame\n",
       "│ Row │ a     │ b     │ x      │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m │ \u001b[90mString\u001b[39m │\n",
       "├─────┼───────┼───────┼────────┤\n",
       "│ 1   │ 1     │ 1     │ no     │\n",
       "│ 2   │ 2     │ 2     │ no     │\n",
       "│ 3   │ 3     │ 3     │ no     │\n",
       "│ 4   │ 4     │ 4     │ no     │\n",
       "│ 5   │ 5     │ 1     │ no     │\n",
       "│ 6   │ 6     │ 2     │ no     │\n",
       "│ 7   │ 7     │ 3     │ no     │\n",
       "│ 8   │ 8     │ 4     │ yes    │\n",
       "│ 9   │ 9     │ 1     │ no     │\n",
       "│ 10  │ 10    │ 2     │ no     │\n",
       "│ 11  │ 11    │ 3     │ no     │\n",
       "│ 12  │ 12    │ 4     │ yes    │"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# or you can use @byrow! macro that allows you to process DataFrame rowwise\n",
    "@byrow! df begin\n",
    "    @newcol x::Vector{String}\n",
    "    :x = :a > 6 && :b == 4 ? \"yes\" : \"no\"\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualizing data with StatsPlots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "using StatsPlots # you might need to setup Plots package and some plotting backend first\n",
    "default(fmt=:png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we present only a minimal functionality of the package"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "using Random\n",
    "Random.seed!(1)\n",
    "df = DataFrame(x = sort(randn(1000)), y=randn(1000), z = [fill(\"b\", 500); fill(\"a\", 500)]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@df df plot(:x, :y, legend=:topleft, label=\"y(x)\") # a most basic plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@df df density(:x, label=\"\") # density plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@df df histogram(:y, label=\"y\") # and a histogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@df df boxplot(:z, :x, label=\"x\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@df df violin(:z, :y, label=\"y\") "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Julia 1.0.3",
   "language": "julia",
   "name": "julia-1.0"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}