Created on Skills Network Labs

functions_for_tidyverse.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Writing functions which include `tidyverse` grouping operations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sometimes you will have repetitive code chunks which include functionalities from the `tidyverse`. Rewriting such code as a function\n",
    "is not quite as straight forward as for \"regular\" R functions especially if the input variables of your function are variable names within the `tidyverse` parts of your function (e.g. `group_by()` statements). The below example is an extreme simplification in order to make it easier to understand."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(tidyverse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's have a look at an example which is not yet functionalized... \n",
    "Add a column with the fraction of cars in groups defined by cylinders (`cyl`) and gears (`gear`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<caption>A grouped_df: 8 × 4</caption>\n",
       "<thead>\n",
       "\t<tr><th scope=col>cyl</th><th scope=col>gear</th><th scope=col>n</th><th scope=col>frac</th></tr>\n",
       "\t<tr><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><td>4</td><td>3</td><td> 1</td><td>0.09090909</td></tr>\n",
       "\t<tr><td>4</td><td>4</td><td> 8</td><td>0.72727273</td></tr>\n",
       "\t<tr><td>4</td><td>5</td><td> 2</td><td>0.18181818</td></tr>\n",
       "\t<tr><td>6</td><td>3</td><td> 2</td><td>0.28571429</td></tr>\n",
       "\t<tr><td>6</td><td>4</td><td> 4</td><td>0.57142857</td></tr>\n",
       "\t<tr><td>6</td><td>5</td><td> 1</td><td>0.14285714</td></tr>\n",
       "\t<tr><td>8</td><td>3</td><td>12</td><td>0.85714286</td></tr>\n",
       "\t<tr><td>8</td><td>5</td><td> 2</td><td>0.14285714</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A grouped\\_df: 8 × 4\n",
       "\\begin{tabular}{llll}\n",
       " cyl & gear & n & frac\\\\\n",
       " <dbl> & <dbl> & <int> & <dbl>\\\\\n",
       "\\hline\n",
       "\t 4 & 3 &  1 & 0.09090909\\\\\n",
       "\t 4 & 4 &  8 & 0.72727273\\\\\n",
       "\t 4 & 5 &  2 & 0.18181818\\\\\n",
       "\t 6 & 3 &  2 & 0.28571429\\\\\n",
       "\t 6 & 4 &  4 & 0.57142857\\\\\n",
       "\t 6 & 5 &  1 & 0.14285714\\\\\n",
       "\t 8 & 3 & 12 & 0.85714286\\\\\n",
       "\t 8 & 5 &  2 & 0.14285714\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A grouped_df: 8 × 4\n",
       "\n",
       "| cyl &lt;dbl&gt; | gear &lt;dbl&gt; | n &lt;int&gt; | frac &lt;dbl&gt; |\n",
       "|---|---|---|---|\n",
       "| 4 | 3 |  1 | 0.09090909 |\n",
       "| 4 | 4 |  8 | 0.72727273 |\n",
       "| 4 | 5 |  2 | 0.18181818 |\n",
       "| 6 | 3 |  2 | 0.28571429 |\n",
       "| 6 | 4 |  4 | 0.57142857 |\n",
       "| 6 | 5 |  1 | 0.14285714 |\n",
       "| 8 | 3 | 12 | 0.85714286 |\n",
       "| 8 | 5 |  2 | 0.14285714 |\n",
       "\n"
      ],
      "text/plain": [
       "  cyl gear n  frac      \n",
       "1 4   3     1 0.09090909\n",
       "2 4   4     8 0.72727273\n",
       "3 4   5     2 0.18181818\n",
       "4 6   3     2 0.28571429\n",
       "5 6   4     4 0.57142857\n",
       "6 6   5     1 0.14285714\n",
       "7 8   3    12 0.85714286\n",
       "8 8   5     2 0.14285714"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "frac1_tab <- mtcars %>% \n",
    "  group_by(cyl, gear) %>%\n",
    "  tally() %>%\n",
    "  mutate(frac = n/sum(n))\n",
    "# show output\n",
    "frac1_tab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly, you can perform the same task with any other categorical variable set (e.g. `am` and `vs`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<caption>A grouped_df: 4 × 4</caption>\n",
       "<thead>\n",
       "\t<tr><th scope=col>am</th><th scope=col>vs</th><th scope=col>n</th><th scope=col>frac</th></tr>\n",
       "\t<tr><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><td>0</td><td>0</td><td>12</td><td>0.6315789</td></tr>\n",
       "\t<tr><td>0</td><td>1</td><td> 7</td><td>0.3684211</td></tr>\n",
       "\t<tr><td>1</td><td>0</td><td> 6</td><td>0.4615385</td></tr>\n",
       "\t<tr><td>1</td><td>1</td><td> 7</td><td>0.5384615</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A grouped\\_df: 4 × 4\n",
       "\\begin{tabular}{llll}\n",
       " am & vs & n & frac\\\\\n",
       " <dbl> & <dbl> & <int> & <dbl>\\\\\n",
       "\\hline\n",
       "\t 0 & 0 & 12 & 0.6315789\\\\\n",
       "\t 0 & 1 &  7 & 0.3684211\\\\\n",
       "\t 1 & 0 &  6 & 0.4615385\\\\\n",
       "\t 1 & 1 &  7 & 0.5384615\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A grouped_df: 4 × 4\n",
       "\n",
       "| am &lt;dbl&gt; | vs &lt;dbl&gt; | n &lt;int&gt; | frac &lt;dbl&gt; |\n",
       "|---|---|---|---|\n",
       "| 0 | 0 | 12 | 0.6315789 |\n",
       "| 0 | 1 |  7 | 0.3684211 |\n",
       "| 1 | 0 |  6 | 0.4615385 |\n",
       "| 1 | 1 |  7 | 0.5384615 |\n",
       "\n"
      ],
      "text/plain": [
       "  am vs n  frac     \n",
       "1 0  0  12 0.6315789\n",
       "2 0  1   7 0.3684211\n",
       "3 1  0   6 0.4615385\n",
       "4 1  1   7 0.5384615"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "frac2_tab <- mtcars %>%\n",
    "  group_by(am, vs) %>%\n",
    "  tally() %>%\n",
    "  mutate(frac = n/sum(n))\n",
    "# show output\n",
    "frac2_tab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Rewrite as function using `enquo()` and `!!` (bang bang):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "add_frac_column <- function(data, group_one, group_two) {\n",
    "  group_one <- enquo(group_one)\n",
    "  group_two <- enquo(group_two)\n",
    "    \n",
    "  data %>%\n",
    "    group_by(!! group_one, !! group_two) %>%\n",
    "    tally() %>%\n",
    "    mutate(frac = n/sum(n))\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "frac1_tab_func <- add_frac_column(mtcars, group_one = cyl, group_two = gear)\n",
    "frac2_tab_func <- add_frac_column(mtcars, group_one = am, group_two = vs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "TRUE"
      ],
      "text/latex": [
       "TRUE"
      ],
      "text/markdown": [
       "TRUE"
      ],
      "text/plain": [
       "[1] TRUE"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "TRUE"
      ],
      "text/latex": [
       "TRUE"
      ],
      "text/markdown": [
       "TRUE"
      ],
      "text/plain": [
       "[1] TRUE"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Test for equality with non functionalized couterparts\n",
    "all_equal(frac1_tab_func, frac1_tab)\n",
    "all_equal(frac2_tab_func, frac2_tab)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clearly, it may not be necessary to write a function for such a small task. However, if you only add a few more lines or call the function more often with different parameters, a functionalized form of your code is likely already better."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "conda-env-r-r"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}