Review

class: center, middle, inverse, title-slide

# Review
### Daniel Anderson
### Week 9, Class 2

---

layout: true

---
# Agenda

* Quick looping review

* Review functions

* A shiny challenge
  + But first, a brief discussion on publishing shiny apps

---
class: inverse-red middle
# Quick looping review

---
# Scenario

* Let's say you want to `scale()` (standardize) every numeric column in a data frame.

--
Use `palmerpenguins::penguins()` for our example.

```r
library(palmerpenguins)
penguins
```

```
## # A tibble: 344 x 8
##    species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g sex     year
##    <fct>   <fct>              <dbl>         <dbl>             <int>       <int> <fct>  <int>
##  1 Adelie  Torgersen           39.1          18.7               181        3750 male    2007
##  2 Adelie  Torgersen           39.5          17.4               186        3800 female  2007
##  3 Adelie  Torgersen           40.3          18                 195        3250 female  2007
##  4 Adelie  Torgersen           NA            NA                  NA          NA <NA>    2007
##  5 Adelie  Torgersen           36.7          19.3               193        3450 female  2007
##  6 Adelie  Torgersen           39.3          20.6               190        3650 male    2007
##  7 Adelie  Torgersen           38.9          17.8               181        3625 female  2007
##  8 Adelie  Torgersen           39.2          19.6               195        4675 male    2007
##  9 Adelie  Torgersen           34.1          18.1               193        3475 <NA>    2007
## 10 Adelie  Torgersen           42            20.2               190        4250 <NA>    2007
## # … with 334 more rows
```

---
# For loop method

You try first. Can you write a `for` loop that loops through each column and applies `scale()` if it's numeric?

---
One approach

```r
penguins2 <- penguins
for(i in seq_along(penguins2)) {
  if (is.numeric(penguins2[ ,i, drop = TRUE])) {
    penguins2[ ,i] <- scale(penguins2[ ,i])
  }
}
```

---

```r
penguins2
```

```
## # A tibble: 344 x 8
##    species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g sex         year
##    <fct>   <fct>              <dbl>         <dbl>             <dbl>       <dbl> <fct>      <dbl>
##  1 Adelie  Torgersen     -0.8832047     0.7843001        -1.416272  -0.5633167  male   -1.257484
##  2 Adelie  Torgersen     -0.8099390     0.1260033        -1.060696  -0.5009690  female -1.257484
##  3 Adelie  Torgersen     -0.6634077     0.4298326        -0.4206603 -1.186793   female -1.257484
##  4 Adelie  Torgersen     NA            NA                NA         NA          <NA>   -1.257484
##  5 Adelie  Torgersen     -1.322799      1.088129         -0.5628905 -0.9374027  female -1.257484
##  6 Adelie  Torgersen     -0.8465718     1.746426         -0.7762357 -0.6880121  male   -1.257484
##  7 Adelie  Torgersen     -0.9198375     0.3285561        -1.416272  -0.7191859  female -1.257484
##  8 Adelie  Torgersen     -0.8648883     1.240044         -0.4206603  0.5901153  male   -1.257484
##  9 Adelie  Torgersen     -1.799025      0.4804708        -0.5628905 -0.9062289  <NA>   -1.257484
## 10 Adelie  Torgersen     -0.3520286     1.543873         -0.7762357  0.06016004 <NA>   -1.257484
## # … with 334 more rows
```

---
# Replicate with `lapply()`?

---

```r
penguins3 <- penguins
data.frame(
  lapply(penguins3, function(x) {
    if(is.numeric(x)) {
      x <- scale(x)
    }
    x
  })
)
```

```
##     species    island bill_length_mm bill_depth_mm flipper_length_mm  body_mass_g    sex        year
## 1    Adelie Torgersen    -0.88320467    0.78430007       -1.41627153 -0.563316704   male -1.25748435
## 2    Adelie Torgersen    -0.80993901    0.12600328       -1.06069609 -0.500969030 female -1.25748435
## 3    Adelie Torgersen    -0.66340769    0.42983257       -0.42066030 -1.186793445 female -1.25748435
## 4    Adelie Torgersen             NA            NA                NA           NA   <NA> -1.25748435
## 5    Adelie Torgersen    -1.32279862    1.08812936       -0.56289047 -0.937402749 female -1.25748435
## 6    Adelie Torgersen    -0.84657184    1.74642615       -0.77623574 -0.688012052   male -1.25748435
## 7    Adelie Torgersen    -0.91983750    0.32855614       -1.41627153 -0.719185889 female -1.25748435
## 8    Adelie Torgersen    -0.86488825    1.24004400       -0.42066030  0.590115266   male -1.25748435
## 9    Adelie Torgersen    -1.79902541    0.48047078       -0.56289047 -0.906228912   <NA> -1.25748435
## 10   Adelie Torgersen    -0.35202864    1.54387329       -0.77623574  0.060160036   <NA> -1.25748435
## 11   Adelie Torgersen    -1.12131806   -0.02591137       -1.06069609 -1.124445771   <NA> -1.25748435
## 12   Adelie Torgersen    -1.12131806    0.07536506       -1.48738661 -0.625664378   <NA> -1.25748435
## 13   Adelie Torgersen    -0.51687637    0.22727971       -1.34515644 -1.249141119 female -1.25748435
## 14   Adelie Torgersen    -0.97478674    2.05025544       -0.70512065 -0.500969030   male -1.25748435
## 15   Adelie Torgersen    -1.70744334    1.99961722       -0.20731504  0.247203059   male -1.25748435
## 16   Adelie Torgersen    -1.34111504    0.32855614       -1.13181117 -0.625664378 female -1.25748435
## 17   Adelie Torgersen    -0.95647033    0.93621471       -0.42066030 -0.937402749 female -1.25748435
## 18   Adelie Torgersen    -0.26044656    1.79706436       -0.27843012  0.371898407   male -1.25748435
## 19   Adelie Torgersen    -1.74407616    0.63238542       -1.20292626 -1.093271934 female -1.25748435
## 20   Adelie Torgersen     0.38062795    2.20217008       -0.49177539 -0.002187638   male -1.25748435
## 21   Adelie    Biscoe    -1.12131806    0.58174721       -1.91407714 -0.999750423 female -1.25748435
## 22   Adelie    Biscoe    -1.13963448    0.78430007       -1.48738661 -0.750359726   male -1.25748435
## 23   Adelie    Biscoe    -1.46932994    1.03749114       -0.84735082 -0.500969030 female -1.25748435
## 24   Adelie    Biscoe    -1.04805240    0.48047078       -1.13181117 -0.313926008   male -1.25748435
## 25   Adelie    Biscoe    -0.93815391    0.02472685       -1.48738661 -0.500969030   male -1.25748435
## 26   Adelie    Biscoe    -1.57922843    0.88557650       -0.98958100 -0.500969030 female -1.25748435
## 27   Adelie    Biscoe    -0.60845845    0.73366185       -1.27404135 -0.812707400   male -1.25748435
## 28   Adelie    Biscoe    -0.62677486    0.37919435       -0.98958100 -1.249141119 female -1.25748435
## 29   Adelie    Biscoe    -1.10300165    0.73366185       -2.05630731 -1.311488793 female -1.25748435
## 30   Adelie    Biscoe    -0.62677486    0.88557650       -1.48738661 -0.313926008   male -1.25748435
## 31   Adelie     Dream    -0.80993901   -0.22846423       -1.62961679 -1.186793445 female -1.25748435
## 32   Adelie     Dream    -1.23121655    0.48047078       -1.62961679 -0.376273682   male -1.25748435
## 33   Adelie     Dream    -0.80993901    0.32855614       -0.91846591 -1.124445771 female -1.25748435
## 34   Adelie     Dream    -0.55350920    0.88557650       -1.20292626 -0.376273682   male -1.25748435
## 35   Adelie     Dream    -1.37774787   -0.07654958       -0.42066030 -1.093271934 female -1.25748435
## 36   Adelie     Dream    -0.86488825    1.99961722       -0.34954521 -0.064535312   male -1.25748435
## 37   Adelie     Dream    -0.93815391    1.44259686       -0.77623574 -0.313926008   male -1.25748435
## 38   Adelie     Dream    -0.31539581    0.68302364       -1.48738661 -0.812707400 female -1.25748435
## 39   Adelie     Dream    -1.15795089    1.08812936       -1.41627153 -1.124445771 female -1.25748435
## 40   Adelie     Dream    -0.75498976    0.98685293       -1.20292626  0.558941429   male -1.25748435
## 41   Adelie     Dream    -1.35943145    0.42983257       -1.34515644 -1.311488793 female -1.25748435
## 42   Adelie     Dream    -0.57182562    0.63238542       -0.42066030 -0.376273682   male -1.25748435
## 43   Adelie     Dream    -1.45101353    0.68302364       -1.06069609 -1.373836467 female -1.25748435
## 44   Adelie     Dream     0.03261607    1.29068222       -0.34954521  0.247203059   male -1.25748435
## 45   Adelie     Dream    -1.26784938   -0.12718780       -1.13181117 -1.498531815 female -1.25748435
## 46   Adelie     Dream    -0.79162259    0.83493828       -0.77623574  0.496593755   male -1.25748435
## 47   Adelie     Dream    -0.51687637    0.93621471       -1.34515644 -0.968576586   male -1.25748435
## 48   Adelie     Dream    -1.17626731    0.88557650       -1.55850170 -1.529705652   <NA> -1.25748435
## 49   Adelie     Dream    -1.45101353    0.37919435       -0.77623574 -0.937402749 female -1.25748435
## 50   Adelie     Dream    -0.29707939    2.05025544       -0.70512065 -0.064535312   male -1.25748435
## 51   Adelie    Biscoe    -0.79162259    0.27791792       -1.06069609 -0.875055074 female -0.03552216
## 52   Adelie    Biscoe    -0.70004052    0.88557650       -0.91846591  0.122507710   male -0.03552216
## 53   Adelie    Biscoe    -1.63417768    0.37919435       -0.77623574 -0.937402749 female -0.03552216
## 54   Adelie    Biscoe    -0.35202864    1.18940579       -0.06508486 -0.189230660   male -0.03552216
## 55   Adelie    Biscoe    -1.72575975    0.48047078       -0.98958100 -1.623227163 female -0.03552216
## 56   Adelie    Biscoe    -0.46192713    0.73366185       -0.70512065 -0.625664378   male -0.03552216
## 57   Adelie    Biscoe    -0.90152108    0.17664149       -1.06069609 -0.812707400 female -0.03552216
## 58   Adelie    Biscoe    -0.60845845    0.83493828       -0.56289047 -0.500969030   male -0.03552216
## 59   Adelie    Biscoe    -1.35943145   -0.27910244       -1.41627153 -1.685574837 female -0.03552216
## 60   Adelie    Biscoe    -1.15795089    0.98685293       -0.49177539 -0.563316704   male -0.03552216
## 61   Adelie    Biscoe    -1.50596277   -0.12718780       -1.13181117 -1.311488793 female -0.03552216
## 62   Adelie    Biscoe    -0.48024354    1.99961722       -0.42066030  0.247203059   male -0.03552216
## 63   Adelie    Biscoe    -1.15795089   -0.07654958       -1.13181117 -0.750359726 female -0.03552216
## 64   Adelie    Biscoe    -0.51687637    0.53110900       -0.63400556 -0.189230660   male -0.03552216
## 65   Adelie    Biscoe    -1.37774787   -0.02591137       -1.20292626 -1.685574837 female -0.03552216
## 66   Adelie    Biscoe    -0.42529430    0.42983257       -0.63400556 -0.313926008   male -0.03552216
## 67   Adelie    Biscoe    -1.54259560   -0.48165530       -0.42066030 -1.062098097 female -0.03552216
## 68   Adelie    Biscoe    -0.51687637    0.98685293       -0.91846591 -0.126882986   male -0.03552216
## 69   Adelie Torgersen    -1.46932994   -0.27910244       -0.77623574 -1.436184141 female -0.03552216
## 70   Adelie Torgersen    -0.38866147    1.13876757       -0.20731504  0.309550733   male -0.03552216
## 71   Adelie Torgersen    -1.90892390    0.93621471       -0.77623574 -0.750359726 female -0.03552216
## 72   Adelie Torgersen    -0.77330618    0.63238542       -0.77623574 -0.376273682   male -0.03552216
## 73   Adelie Torgersen    -0.79162259    0.02472685       -0.34954521 -0.812707400 female -0.03552216
## 74   Adelie Torgersen     0.34399512    0.88557650       -0.27843012 -0.064535312   male -0.03552216
## 75   Adelie Torgersen    -1.54259560    0.17664149       -0.77623574 -0.625664378 female -0.03552216
## 76   Adelie Torgersen    -0.20549732    0.68302364       -0.42066030  0.060160036   male -0.03552216
## 77   Adelie Torgersen    -0.55350920   -0.17782601       -0.70512065 -0.625664378 female -0.03552216
## 78   Adelie Torgersen    -1.23121655    1.13876757       -1.20292626 -0.376273682   male -0.03552216
## 79   Adelie Torgersen    -1.41438070   -0.53229351       -0.98958100 -0.812707400 female -0.03552216
## 80   Adelie Torgersen    -0.33371222    0.98685293       -0.42066030 -0.251578334   male -0.03552216
## 81   Adelie Torgersen    -1.70744334    0.02472685       -0.84735082 -1.249141119 female -0.03552216
## 82   Adelie Torgersen    -0.18718091    0.22727971       -0.34954521  0.621289103   male -0.03552216
## 83   Adelie Torgersen    -1.32279862    0.83493828       -0.98958100 -0.500969030 female -0.03552216
## 84   Adelie Torgersen    -1.61586126    1.13876757       -0.56289047 -0.002187638   male -0.03552216
## 85   Adelie     Dream    -1.21290014    0.32855614       -0.70512065 -1.062098097 female -0.03552216
## 86   Adelie     Dream    -0.48024354    1.59451151       -0.49177539 -0.812707400   male -0.03552216
## 87   Adelie     Dream    -1.39606428    1.18940579       -0.77623574 -0.500969030   male -0.03552216
## 88   Adelie     Dream    -1.28616579    0.73366185       -0.84735082 -0.875055074 female -0.03552216
## 89   Adelie     Dream    -1.02973599    1.03749114       -0.84735082 -0.313926008   male -0.03552216
## 90   Adelie     Dream    -0.91983750    0.83493828       -0.77623574 -0.750359726 female -0.03552216
## 91   Adelie     Dream    -1.50596277    0.42983257        0.07714531 -0.812707400 female -0.03552216
## 92   Adelie     Dream    -0.51687637    0.48047078        0.29049058  0.122507710   male -0.03552216
## 93   Adelie     Dream    -1.81734182   -0.02591137       -1.13181117 -0.999750423 female -0.03552216
## 94   Adelie     Dream    -0.79162259    0.48047078       -1.06069609  0.309550733   male -0.03552216
## 95   Adelie     Dream    -1.41438070    0.07536506       -0.98958100 -1.124445771 female -0.03552216
## 96   Adelie     Dream    -0.57182562    0.88557650        0.50383584  0.122507710   male -0.03552216
## 97   Adelie     Dream    -1.06636882    0.73366185       -0.77623574 -0.625664378 female -0.03552216
## 98   Adelie     Dream    -0.66340769    0.68302364       -0.34954521  0.184855384   male -0.03552216
## 99   Adelie     Dream    -1.98218956   -0.53229351       -1.62961679 -1.623227163 female -0.03552216
## 100  Adelie     Dream    -0.13223166    0.68302364       -0.63400556 -0.126882986   male -0.03552216
## 101  Adelie    Biscoe    -1.63417768    0.37919435       -0.63400556 -0.594490541 female  1.18644003
## 102  Adelie    Biscoe    -0.53519279    1.44259686        0.14826040  0.652462940   male  1.18644003
## 103  Adelie    Biscoe    -1.13963448   -0.58293173       -1.27404135 -1.405010304 female  1.18644003
## 104  Adelie    Biscoe    -1.12131806    1.44259686       -0.77623574  0.060160036   male  1.18644003
## 105  Adelie    Biscoe    -1.10300165    0.73366185       -0.56289047 -1.592053326 female  1.18644003
## 106  Adelie    Biscoe    -0.77330618    0.88557650       -1.20292626 -0.812707400   male  1.18644003
## 107  Adelie    Biscoe    -0.97478674    0.02472685       -0.13619995 -0.563316704 female  1.18644003
## 108  Adelie    Biscoe    -1.04805240    1.44259686       -0.77623574 -0.376273682   male  1.18644003
## 109  Adelie    Biscoe    -1.06636882   -0.07654958       -1.41627153 -1.280314956 female  1.18644003
## 110  Adelie    Biscoe    -0.13223166    0.93621471       -0.27843012  0.714810614   male  1.18644003
## 111  Adelie    Biscoe    -1.06636882   -0.32974066       -0.20731504 -0.469795193 female  1.18644003
## 112  Adelie    Biscoe     0.30736229    1.59451151       -0.70512065  0.496593755   male  1.18644003
## 113  Adelie    Biscoe    -0.77330618    0.27791792       -0.56289047 -1.249141119 female  1.18644003
## 114  Adelie    Biscoe    -0.31539581    1.18940579       -0.27843012  0.091333873   male  1.18644003
## 115  Adelie    Biscoe    -0.79162259    1.79706436       -0.70512065 -0.376273682 female  1.18644003
## 116  Adelie    Biscoe    -0.22381374    0.58174721       -0.34954521 -0.158056823   male  1.18644003
## 117  Adelie Torgersen    -0.97478674   -0.07654958       -0.91846591 -1.623227163 female  1.18644003
## 118  Adelie Torgersen    -1.21290014    1.69578793       -0.13619995 -0.532142867   male  1.18644003
## 119  Adelie Torgersen    -1.50596277   -0.07654958       -0.84735082 -1.062098097 female  1.18644003
## 120  Adelie Torgersen    -0.51687637    0.73366185       -0.84735082 -1.093271934   male  1.18644003
## 121  Adelie Torgersen    -1.41438070    0.02472685       -0.98958100 -1.311488793 female  1.18644003
## 122  Adelie Torgersen    -1.13963448    1.34132043       -0.20731504 -0.875055074   male  1.18644003
## 123  Adelie Torgersen    -0.68172411   -0.07654958       -1.77184696 -0.937402749 female  1.18644003
## 124  Adelie Torgersen    -0.46192713    0.68302364        0.07714531 -0.407447519   male  1.18644003
## 125  Adelie Torgersen    -1.59754485   -0.63356994       -1.06069609 -1.436184141 female  1.18644003
##  [ reached 'max' / getOption("max.print") -- omitted 219 rows ]
```

---
# {purrr}

When do we use `~`?

Remember, `purrr::map()` is *exactly* like `base::lapply()`, except for the shortcut syntax

--
`~` is a shortcut for `function(.x)`

---
# Equivalent
The following are the same

.pull-left[

```r
map_df(penguins, ~{
  if(is.numeric(.x)) {
    .x <- scale(.x)
  }
  .x
})
```

```
## # A tibble: 344 x 8
##    species island    bill_length_mm[,1] bill_depth_mm[,1] flipper_length_mm[,1] body_mass_g[,1] sex     year[,1]
##    <fct>   <fct>                  <dbl>             <dbl>                 <dbl>           <dbl> <fct>      <dbl>
##  1 Adelie  Torgersen         -0.8832047         0.7843001            -1.416272      -0.5633167  male   -1.257484
##  2 Adelie  Torgersen         -0.8099390         0.1260033            -1.060696      -0.5009690  female -1.257484
##  3 Adelie  Torgersen         -0.6634077         0.4298326            -0.4206603     -1.186793   female -1.257484
##  4 Adelie  Torgersen         NA                NA                    NA             NA          <NA>   -1.257484
##  5 Adelie  Torgersen         -1.322799          1.088129             -0.5628905     -0.9374027  female -1.257484
##  6 Adelie  Torgersen         -0.8465718         1.746426             -0.7762357     -0.6880121  male   -1.257484
##  7 Adelie  Torgersen         -0.9198375         0.3285561            -1.416272      -0.7191859  female -1.257484
##  8 Adelie  Torgersen         -0.8648883         1.240044             -0.4206603      0.5901153  male   -1.257484
##  9 Adelie  Torgersen         -1.799025          0.4804708            -0.5628905     -0.9062289  <NA>   -1.257484
## 10 Adelie  Torgersen         -0.3520286         1.543873             -0.7762357      0.06016004 <NA>   -1.257484
## # … with 334 more rows
```

]

.pull-right[

```r
map_df(penguins, function(.x) {
  if(is.numeric(.x)) {
    .x <- scale(.x)
  }
  .x
})
```

]

---
# Whe to use `~`?

You can use it whenever you feel comfortable, including always. You can also just loop a function through, and pass additional arguments to that function, e.g.,

```r
map_dbl(penguins, mean, na.rm = TRUE)
```

```
##           species            island    bill_length_mm     bill_depth_mm flipper_length_mm       body_mass_g 
##                NA                NA          43.92193          17.15117         200.91520        4201.75439 
##               sex              year 
##                NA        2008.02907
```

```r
map_dbl(penguins, ~mean(.x, na.rm = TRUE))
```

---
The `~ .x` syntax is also helpful for more complex things

```r
map_dbl(penguins, ~ifelse(
    is.numeric(.x), 
    mean(.x, na.rm = TRUE), 
    0
  )
)
```

```
##           species            island    bill_length_mm     bill_depth_mm flipper_length_mm       body_mass_g 
##           0.00000           0.00000          43.92193          17.15117         200.91520        4201.75439 
##               sex              year 
##           0.00000        2008.02907
```

---
class: inverse-blue middle
# Functions review

---
# Remember

* Everything is a function

--
The following are equivalent

```r
3 + 5
```

```
## [1] 8
```

```r
`+`(3, 5)
```

```
## [1] 8
```

---
# Using functions

* Most functions are bound to a name, e.g., `mean()`

--
* Anonymous functions are also common

+ Apply the function in a loop, and it only ever exists in the loop

--
* You can also store functions in lists

+ Helpful if you want to apply lots of operations to a single vector

---
# Binding to a name

* Let's create a function that takes two arguments: (a) a data frame, and (b) the name of a discrete/categorical variable/column in the data frame.

* The function should return the count of each "level" in the categorical variable.

* For a small added challenge, have it optionally add the proportion

--
Example:

Example output with `palmerpenguins::penguins`.

.pull-left[

```
##     species count
## 1    Adelie   152
## 2 Chinstrap    68
## 3    Gentoo   124
```

]

.pull-right[

```
##     species count proportion
## 1    Adelie   152  0.4418605
## 2 Chinstrap    68  0.1976744
## 3    Gentoo   124  0.3604651
```
]

---
# You try first

Test it out with the **palmerpenguins** dataset. Do you get the same results I did?

Note - the example I used included only base R functions. You can feel free to use **dplyr** or whatevs, just be careful with NSE.

You also don't have to return a data frame output - return it however you want

---
# Where to start?

~~write a function~~

* Solve the problem for one example, generalize it to a function.

--
Use the **palmerpenguins** dataset *for* your example!

```r
library(palmerpenguins)
penguins
```

---
# How do you want to solve it?
Lots of ways, here's a base method

--
* First, split by species

```r
splt <- split(penguins, penguins$species)
```

--
* Next, count how many rows (observations) in each split

```r
sapply(splt, nrow)
```

```
##    Adelie Chinstrap    Gentoo 
##       152        68       124
```

--
Could go on, but this is basically the output.

---
# Wrap in a function

* What will the arguments be?

--
The data frame and the column

```r
get_counts <- function(df, column) {
  
}
```

--
What will the body be?

---
# Same as before
Just swap out the code for the arguments. Notice I'm indexing the columns differently. Why?

I'm also swapping out `sapply()` for `vapply()` to be a little more safe.

```r
get_counts <- function(df, column) {
  splt <- split(df, df[[column]])
  vapply(splt, nrow, FUN.VALUE = integer(1))
}
```

---
# Test it

```r
get_counts(penguins, "species")
```

```
##    Adelie Chinstrap    Gentoo 
##       152        68       124
```

```r
get_counts(penguins, "island")
```

```
##    Biscoe     Dream Torgersen 
##       168       124        52
```

---
# Extensions
Let's say we want a data frame as the output.

Can you modify what we have now to make that so?

---
# Data frame

```r
get_counts <- function(df, column) {
  splt <- split(df, df[[column]])
  counts <- vapply(splt, nrow, FUN.VALUE = integer(1))
  
  tibble::tibble(
    var_levels = names(counts), # could use names(splt)
    count = counts
  )
}
```

---
# Test it

```r
get_counts(penguins, "species")
```

```
## # A tibble: 3 x 2
##   var_levels count
##   <chr>      <int>
## 1 Adelie       152
## 2 Chinstrap     68
## 3 Gentoo       124
```

```r
get_counts(penguins, "island")
```

```
## # A tibble: 3 x 2
##   var_levels count
##   <chr>      <int>
## 1 Biscoe       168
## 2 Dream        124
## 3 Torgersen     52
```

---
# Column name

Can we make the output from the data frame have the same column that we fed it?

```r
get_counts <- function(df, column) {
  splt <- split(df, df[[column]])
  counts <- vapply(splt, nrow, FUN.VALUE = integer(1))
  
  d <- tibble::tibble(
    var_levels = names(counts), # could use names(splt)
    count = counts
  )
* names(d)[1] <- column
  d
}
```

---
# Test it

```r
get_counts(penguins, "species")
```

```
## # A tibble: 3 x 2
##   species   count
##   <chr>     <int>
## 1 Adelie      152
## 2 Chinstrap    68
## 3 Gentoo      124
```

```r
get_counts(penguins, "island")
```

```
## # A tibble: 3 x 2
##   island    count
##   <chr>     <int>
## 1 Biscoe      168
## 2 Dream       124
## 3 Torgersen    52
```

---
# {dplyr} version

Can we replicate this function using dplyr?

We'll have to use non-standard evaluation

First, solve it on a use case

```r
penguins %>% 
  count(species)
```

```
## # A tibble: 3 x 2
##   species       n
##   <fct>     <int>
## 1 Adelie      152
## 2 Chinstrap    68
## 3 Gentoo      124
```

---
# Function

Will this work?

```r
get_counts <- function(df, column) {
  df %>% 
    count(column)
}
```

```r
get_counts(penguins, species)
```

```
## Error: Must group by variables found in `.data`.
## * Column `column` is not found.
```

---
# Use NSE

```r
get_counts <- function(df, column) {
  df %>% 
    count({{column}})
}
```

```r
get_counts(penguins, species)
```

```
## # A tibble: 3 x 2
##   species       n
##   <fct>     <int>
## 1 Adelie      152
## 2 Chinstrap    68
## 3 Gentoo      124
```

```r
get_counts(penguins, island)
```

```
## # A tibble: 3 x 2
##   island        n
##   <fct>     <int>
## 1 Biscoe      168
## 2 Dream       124
## 3 Torgersen    52
```

---
# Pass the dots

Alternatively, you could just pass the dots

Bonus, this will now give you the counts for multiple columns

```r
get_counts <- function(df, ...) {
  df %>% 
    count(...)
}
```

---
# Test it

```r
get_counts(penguins, species)
```

```
## # A tibble: 3 x 2
##   species       n
##   <fct>     <int>
## 1 Adelie      152
## 2 Chinstrap    68
## 3 Gentoo      124
```

```r
get_counts(penguins, species, island)
```

```
## # A tibble: 5 x 3
##   species   island        n
##   <fct>     <fct>     <int>
## 1 Adelie    Biscoe       44
## 2 Adelie    Dream        56
## 3 Adelie    Torgersen    52
## 4 Chinstrap Dream        68
## 5 Gentoo    Biscoe      124
```

---
# Conditions

* Let's add a condition that optionally reports the proportions in addition to the counts.

--
* What will be the first step?

--
* Add a new argument (and consider setting defaults for that argument)

```r
get_counts <- function(df, column, return_proportions = FALSE) {
  df %>% 
    count({{column}})
}
```

---
# Set conditional block
Create a block for operations to conduct **when the condition is TRUE**

```r
get_counts <- function(df, column, return_proportions = FALSE) {
  counts <- df %>% 
    count({{column}})
  
  if (isTRUE(return_proportions)) {
    
  }
  counts
}
```

---
# Write condition
In the block, include the code that is only evaluated when the condition is `TRUE`.

```r
get_counts <- function(df, column, return_proportions = FALSE) {
  counts <- df %>% 
    count({{column}})
  
  if (isTRUE(return_proportions)) {
    counts <- counts %>% 
      mutate(proportion = n / sum(n))
  }
  counts
}
```

---
# Test it

```r
get_counts(penguins, species)
```

```
## # A tibble: 3 x 2
##   species       n
##   <fct>     <int>
## 1 Adelie      152
## 2 Chinstrap    68
## 3 Gentoo      124
```

```r
get_counts(penguins, species, return_proportions = TRUE)
```

```
## # A tibble: 3 x 3
##   species       n proportion
##   <fct>     <int>      <dbl>
## 1 Adelie      152  0.4418605
## 2 Chinstrap    68  0.1976744
## 3 Gentoo      124  0.3604651
```

---
# Challenge
Now that we have a basic function, can you write a **new** function that *calls this function* to add the proportions and/or counts to a data frame?

Should return the original data frame, but with the counts/proportions added in as a new column.

---
# One solution

```r
add_counts <- function(data, column, add_proportions = FALSE) {
  counts <- get_counts(data, {{column}}, add_proportions)
  left_join(data, counts)
}
```

---
# Test it out

I'm selecting variables after just so we can see the counts

```r
add_counts(penguins, species) %>% 
  select(species, island, n)
```

```
## # A tibble: 344 x 3
##    species island        n
##    <fct>   <fct>     <int>
##  1 Adelie  Torgersen   152
##  2 Adelie  Torgersen   152
##  3 Adelie  Torgersen   152
##  4 Adelie  Torgersen   152
##  5 Adelie  Torgersen   152
##  6 Adelie  Torgersen   152
##  7 Adelie  Torgersen   152
##  8 Adelie  Torgersen   152
##  9 Adelie  Torgersen   152
## 10 Adelie  Torgersen   152
## # … with 334 more rows
```

---
# Test it again

This time let's add the proportions

```r
add_counts(penguins, species, add_proportions = TRUE) %>% 
  select(species, island, n, proportion)
```

```
## # A tibble: 344 x 4
##    species island        n proportion
##    <fct>   <fct>     <int>      <dbl>
##  1 Adelie  Torgersen   152  0.4418605
##  2 Adelie  Torgersen   152  0.4418605
##  3 Adelie  Torgersen   152  0.4418605
##  4 Adelie  Torgersen   152  0.4418605
##  5 Adelie  Torgersen   152  0.4418605
##  6 Adelie  Torgersen   152  0.4418605
##  7 Adelie  Torgersen   152  0.4418605
##  8 Adelie  Torgersen   152  0.4418605
##  9 Adelie  Torgersen   152  0.4418605
## 10 Adelie  Torgersen   152  0.4418605
## # … with 334 more rows
```

---
# Embed checks

Can you embed a warning or error (your choice) if the column fed to the function is not discrete?

Note - this is more difficult with our **dplyr** version. Try using `dplyr::pull()`.

---

```r
get_counts <- function(df, column, return_proportions = FALSE) {
  
  column_vec <- dplyr::pull(df, {{column}})
  
  if(is.numeric(column_vec)) {
    stop("Numeric column passed to function. Counts must be computed on categorical data.")
  }
  
  counts <- df %>% 
    count({{column}})
  
  if (isTRUE(return_proportions)) {
    counts <- counts %>% 
      mutate(proportion = n / sum(n))
  }
  counts
}
```

---
# Test it out
Note we can test it with either the `get_counts()` or `add_counts()` functions

```r
get_counts(penguins, bill_length_mm)
```

```
## Error in get_counts(penguins, bill_length_mm): Numeric column passed to function. Counts must be computed on categorical data.
```

```r
add_counts(penguins, bill_length_mm)
```

```
## Error in get_counts(data, {: Numeric column passed to function. Counts must be computed on categorical data.
```

---
class: inverse-blue middle
# Shiny

---
# Publishing
* We never talked about publishing shiny apps

See [here](https://statsandr.com/blog/how-to-publish-shiny-app-example-with-shinyapps-io/) for a nice step-by-step walkthrough for publishing with https://www.shinyapps.io/

--
Basically:
* Register an account with https://www.shinyapps.io/
* Add a token to your account on shinyapps
* Back locally, set your account info with the token and secret via

```r
rsconnect::setAccountInfo(
  name = "myaccount", # replace with your account name
  token = "mytokencopiedfromshinyappsio", # your token
  secret = "mysecretcopiedfromshinyappsio"
)
```

---
.center[
### Publish
]

.footnote[image from tutorial [here](https://statsandr.com/blog/how-to-publish-shiny-app-example-with-shinyapps-io/)]

![](https://statsandr.com/blog/2020-05-29-how-to-deploy-a-shiny-app-an-example-with-shinyapps-io_files/publish-shiny-app-online-shinyapps-io-4.png)

---
# Shiny app
* Create a shiny app or shiny dashboard with the `palmerpenguins` dataset

* Allow the x and y axis to be selected by the user
  + Only numeric variables should be available to be selected
  
* Allow the points to be colored by any categorical variable

+ For an added challenge, try to add in a "no color" option, which should be the default

Once you've gone this far, try to publish your app. If you're successful, continue with challenge on next slide

---
# Challenge continued

* Add a table to the app that reports descriptive data on the columns that are selected in the plot
  + e.g., `n()`, `mean()`, `sd()`
  
* Use tabs so the plot shows up in one tab, and the table shows up in a different tab

Now publish again to update it

---
class: inverse-green middle 
# Next time
## No Class Monday
## Package Development on Wednesday