AS2017496.Rmd

---
title: "Show My Data"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    vertical_layout: fill
runtime: shiny
---

```{r setup, include=FALSE}
library(flexdashboard)
library(tidyverse)
library(plotly)
library(summarytools)
library(visdat)
library(skimr)

#word clouds
library("tm")
library("SnowballC")
library("wordcloud")
library("RColorBrewer")

```

Data Overview
======================================================================

Column {.sidebar}
-----------------------------------------------------------------------

Upload your csv file here. 

```{r}
fileInput("file", "Choose CSV File",
                multiple = FALSE,
                accept = c("text/csv",
                         "text/comma-separated-values,text/plain",
                         ".csv"))
```


Row 
-----------------------------------------------------------------------
### Total Data Records
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox(nrow(df),icon="fa-database")
})
```

### Total Missing values (Click Me!)

```{r}
renderValueBox({
req(input$file)
df <- read_csv(input$file$datapath)
na_count <- sum(is.na(df))
valueBox(na_count, 
         href= "#section-missing-values",
         icon = "fa-exclamation",
         color = ifelse(na_count > 0, "danger", "primary"))
})
```




Row
---------------------------------------------------------
### Total Variables
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox(ncol(df),color = "info", icon="fa-list-alt")
})
```



### Quantitative Variables
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox(ncol(df %>% select_if(is.numeric)),color = "info", icon="fa-sort-numeric-up")
})
```

### Qualitative Variables
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox(ncol(df %>% select_if(is.character)),color = "info", icon="fa-sort-alpha-up")
})
```



Row
---------------------------------------------------------

### Dataset Preview

```{r}
DT::renderDataTable({
  req(input$file)
df <- read_csv(input$file$datapath)
  DT::datatable(df, rownames = TRUE,options = list(bPaginate = TRUE, searching = TRUE, info = FALSE),fillContainer = TRUE)
})
```

Summary Statistics
========================================================
Row
-----------------------------------------------------------------------

### Let's refresh our memory on variable types...
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox("Click Me!",href= "#section-help" ,color = "info", icon="ion-android-happy")
})
```

Row
-----------------------------------------------------------------------

### List of Variables

```{r}


renderTable({
  
  req(input$file)
df <- read_csv(input$file$datapath)
#df[sapply(df, is.character)] <- lapply(df[sapply(df, is.character)], as.factor)


df <- df %>% 
  mutate_if(function(col){
    if(is.character(col) & length(unique(col)) < 10){
      return(TRUE)
    }else{
      return(FALSE)
    }
  },factor)


  `Qualitative Variables` <- colnames(df %>% select_if(is.character))
  `Categorical Variables` <- colnames(df %>% select_if(is.factor))
  `Quantitative Variables` <- colnames(df %>% select_if(is.numeric))
  
  max_len <- max(length(`Qualitative Variables`), length(`Categorical Variables`), 
                 length(`Quantitative Variables`))
  
  `Qualitative Variables` <- c(`Qualitative Variables`, rep(c(""), max_len - length(`Qualitative Variables`)))
  
  `Quantitative Variables` <- c(`Quantitative Variables`, rep(c(""), max_len - length(`Quantitative Variables`)))
  
  `Categorical Variables` <- c(`Categorical Variables`, rep(c(""), max_len - length(`Categorical Variables`)))
  
  names <- data.frame(`Character Variables`=`Qualitative Variables`, 
                      `Quantitative Variables`= `Quantitative Variables`, 
                      `Categorical Variables`= `Categorical Variables`)
  names
      
})

```


Row
-----------------------------------------------------------------------

### Summary Measures for Quantitative Variables
```{r}

renderTable({
  req(input$file)
  df <- read_csv(input$file$datapath)
  m <- round(descr(df, 
          stats = c("mean", "sd", "min", "q1", "med", "q3", "max", "iqr", "cv"),
          transpose = FALSE),2) 
   class(m) <-"matrix"
   `Summary Measure` <- c("Mean", "Std. Deviation", "Minimum", "1st Quartile", "Median", "3rd Quartile", "Maximum", "Inter Quartile Range", "Coefficient of Variation")
   m <- cbind(`Summary Measure`, m)
  m
})


```




Row
----------------------------------------------------------
### Summary Measures for Qualitative Variables



```{r}

renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
df <- df %>% 
  mutate_if(function(col){
    if(is.character(col) & length(unique(col)) < 10){
      return(TRUE)
    }else{
      return(FALSE)
    }
  },factor)
 varSelectInput("Cat. Variable", "Select Categorical Variable:", Filter(is.factor, df),
                    selected = NULL)
  })




```


```{r}

renderTable({
  req(input$file)
df <- read_csv(input$file$datapath)
df <- df %>% 
  mutate_if(function(col){
    if(is.character(col) & length(unique(col)) < 10){
      return(TRUE)
    }else{
      return(FALSE)
    }
  },factor)

Group <- levels(df[[input$`Cat. Variable`]])
Frequency <- summary(df[[input$`Cat. Variable`]])
Percentage <- round(summary(df[[input$`Cat. Variable`]])*100/length(df[[input$`Cat. Variable`]]),2)

Table <- cbind(Group, Frequency, Percentage)

})

```



Correlations
=========================================================

Row
-----------------------------------------------------------------------

### What actually is "Normal" data?
```{r}
renderValueBox({
  req(input$file)
df <- read_csv(input$file$datapath)
  valueBox("Click Me!",href= "#section-help" ,color = "info", icon="ion-android-happy")
})
```

Row
-----------------------------------------------------------------------

```{r}
get_heatmap <- function(df,cor_type = "pearson"){
cmat <- cor(df %>% select_if(is.numeric),
            use = "complete.obs",method = cor_type) %>% as.data.frame() 
cmat <- cmat %>% mutate(`Variable 1` = names(cmat))
cmat <- cmat %>% pivot_longer(-`Variable 1`,names_to = "Variable 2",values_to = "corr")

ggheatmap <- ggplot(cmat, aes(`Variable 1`, `Variable 2`, fill = corr))+
  geom_tile(color = "white")+
  geom_text(aes(`Variable 1`, `Variable 2`, label = round(corr,3)), color = "black", size = 4) +
  scale_fill_gradient2(low = "blue", high = "red", mid = "white", 
                       midpoint = 0, limit = c(-1,1), space = "Lab", 
                       name=paste0(c(cor_type,"\nCorrelation"))) +
  theme_minimal()+ # minimal theme
  theme(axis.text.x = element_text(angle = 45, vjust = 1, 
                                   size = 12, hjust = 1))+
  coord_fixed()
return(ggheatmap)
}


```


### Pearson
The strength and direction of linear relationship given data is normal
```{r}
renderPlot({
  req(input$file)
df <- read_csv(input$file$datapath)
  get_heatmap(df)
})
```



### Spearman

The strength and direction of linear relationship despite data not being normal
```{r}
renderPlot({
  req(input$file)
df <- read_csv(input$file$datapath)
  get_heatmap(df,cor_type="spearman")
})
```

Missing Values
==========================================
Row
-------------------------------------------
### Missing value matrix
The variables have been arranged from left to right in descending order of missing values. 

```{r}
renderPlot({
  req(input$file)
  df <- read_csv(input$file$datapath)
  vis_miss(df, sort_miss = TRUE)
})

```


By variable plots
==========================================================

Column {.sidebar}
----------------------------------------------------------


```{r}
renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
varSelectInput("Variable 1", "Select Quantitative Variable:", Filter(is.numeric, df),
                   selected = NULL)
  #selectInput("Variable 1", label = "Select Numeric Variable:", choices = colnames(df))
})

```

```{r}
renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
df <- df %>% 
  mutate_if(function(col){
    if(is.character(col) & length(unique(col)) < 10){
      return(TRUE)
    }else{
      return(FALSE)
    }
  },factor)
 varSelectInput("Variable 2", "Select Categorical Variable:", Filter(is.factor, df),
                    selected = NULL)
  #selectInput("Variable 2", label = "Select Categorical Variable:", choices = colnames(df))
})
```


Column 
-----------------------------------------------------------------------
### Boxplot of Quantitative variable by Categorical variable
```{r}
renderPlotly({
  req(input$file)
df <- read_csv(input$file$datapath)
   plot_ly(df,
              x = ~df[[input$`Variable 1`]],
              color = ~df[[input$`Variable 2`]],
              colors = "Paired",
              type = "box") %>%
  layout(title = "",
         xaxis = list(title = "" ,
                      zeroline = FALSE))
})
```


Column 
-----------------------------------------------------------------------

### Distribution of Categorical Variable

```{r}
renderPlotly({
  req(input$file)
df <- read_csv(input$file$datapath)
  
  df %>%
    count(var = df[[input$`Variable 2`]], name = "count") %>%
    plot_ly( x = ~var, y = ~ count, type = "bar", marker = list(color = '#008ae6',
                           line = list(color = '#008ae6', width = 2)), hoverinfo = "x+y") %>%
    add_text(text = ~paste0( " (",   scales::percent(count/sum(count)),")"), 
           textposition = "bottom", 
           textfont = list(size = 12, color = "white"), 
           showlegend = FALSE) %>%
    layout(xaxis = list(title = ""), yaxis = list(title = ""))
    
})
```

### Distribution of Quantitative Variable
```{r}
renderPlotly({
  req(input$file)
df <- read_csv(input$file$datapath)
  plot_ly(x = df[[input$`Variable 1`]], type = "histogram",  marker = list(color = "#008ae6",
                            line = list(color = "darkgray",
                                        width = 1)))
})
```


Scatter plots
==========================================================


Column {.sidebar}
----------------------------------------------------------


```{r}
renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
varSelectInput("Variable 3", "Select Quantitative Variable 1:", Filter(is.numeric, df),
                   selected = NULL)
  #selectInput("Variable 1", label = "Select Numeric Variable:", choices = colnames(df))
})

```


```{r}
renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
varSelectInput("Variable 4", "Select Quantitative Variable 2:", Filter(is.numeric, df),
                   selected = NULL)
  })
```



```{r}
renderUI({
  req(input$file)
df <- read_csv(input$file$datapath)
df <- df %>% 
  mutate_if(function(col){
    if(is.character(col) & length(unique(col)) < 10){
      return(TRUE)
    }else{
      return(FALSE)
    }
  },factor)
 varSelectInput("Variable 5", "Color by Categorical Variable:", Filter(is.factor, df),
                    selected = NULL)
  #selectInput("Variable 2", label = "Select Categorical Variable:", choices = colnames(df))
})
```


Column
-----------------------------------------------------------


### Scatter plot of Quantitative Variables
```{r}
renderPlotly({
  req(input$file)
df <- read_csv(input$file$datapath)
`Quantitative Variable 1` = df[[input$`Variable 3`]]
`Quantitative Variable 2` = df[[input$`Variable 4`]]
plot_ly(data = df, x = ~`Quantitative Variable 1`, y = ~`Quantitative Variable 2`, 
        color = ~df[[input$`Variable 5`]],colors = "Set1") %>% 
  layout(xaxis = list(title = as.character(input$`Variable 3`)),
         yaxis = list(title = as.character(input$`Variable 4`)))
  
})
```


Help!
========================================================

Row
---------------------------------------------------------

### They are variables that represent a measurable quantity as a number.
```{r}
renderValueBox({
  valueBox("Quantitative Variables",color = "info", icon="fa-sort-numeric-up")
})
```

Row
---------------------------------------------------------
### They are variables that are not quantitative.
```{r}
renderValueBox({
  valueBox("Qualitative Variables",color = "info", icon="fa-sort-alpha-up")
})
```


Row
---------------------------------------------------------
### They are variables that can take on one of a limited number of possible values.
```{r}
renderValueBox({
  valueBox("Categorical Variables",color = "info", icon="fa-list-alt")
})
```

Row
---------------------------------------------------------
### Their values do not result from measuring or counting and do not have a limited number of possible values.
```{r}
renderValueBox({
  valueBox("Character Variables",color = "info", icon="fa-sort-alpha-up")
})
```

Row
---------------------------------------------------------
### "Normal" data are data that are drawn (come from) a population that has a normal distribution. So is the population normal? Well you need to run a test...

```{r}
renderValueBox({
  valueBox("Normal Data",color = "danger", icon="fa-exclamation")
})
```