cm012.Rmd

---
title: "cm012"
author: "Cody"
date: "October 12, 2017"
output: pdf_document
---

```{r}
typeof(5)
class(5)

typeof("hi")
class("hi")

a=c(1,2,3);
typeof(a)
class(a)

A=matrix(1:10, nrow = 5, ncol = 2, byrow = T)
typeof(A)
class(A)

library(tidyverse)
library(singer)
data("singer_locations")

my_plot <- singer_locations %>% 
            ggplot(aes(x = year, y = artist_hotttnesss))
typeof(my_plot)
class(my_plot)

x <- list(number_of_cats = 1, cat_color = "red", plot = my_plot)
typeof(x)
class(x)
```
Try to change one entry of singer_locations from one type to another and see what happens.
```{r}
data("singer_locations")
typeof(singer_locations$year)
singer_locations$year[5]="hi"
typeof(singer_locations$year)
head(singer_locations)
```
# Learning how to read and write to file
```{r}
read_csv() #Better than read.csv, in the tidyverse.

readxl::read_xlsx()
write_csv()
```
# Factors
```{r}
library(forcats)
library(singer)
library(tidyverse)
data("singer_locations")

sl <- singer_locations %>% 
  mutate(artist_name_factor = as_factor(artist_name))
glimpse(sl)

sl <- sl %>% 
  mutate(artist_name_factor_base= factor(artist_name))
glimpse(sl)

x <- c("a", "c", "b")
as_factor(x)
factor(x) #base factor will move the factors around which can be risky

sl %>% 
  count(artist_name) %>% 
  arrange(-n)

sl %>% 
  mutate(top_an = fct_lump(artist_name_factor, n = 7)) %>% 
  count(top_an) %>% 
  arrange(-n)

#Homework
summarize(mean_duration = mean(duration)) %>% 
  ggplot(aes(x = artist_name,....)) #finish this

sl <- sl %>% 
  mutate(city_factor = ifelse(is.na(city),
                              "missing information",
                              city),
         city_factor=as_factor(city_factor))
glimpse(sl)

sl %>% 
  mutate(top_city=fct_lump(city_factor, n = 7)) %>% 
  count(top_city) %>% 
  arrange(-n)
# Can use forcats to clean up.
```