Loading and preprocessing the data
- Load the libraries
- Carregamento das bibliotecas
library(ggplot2)
library(dplyr)
library(data.table)
- Code for reading in the dataset and/or processing the data
- Codigo para ler os dados e/ou processar as dados
data_na <- read.csv("activity.csv")
head(data_na)
## steps date interval
## 1 NA 2012-10-01 0
## 2 NA 2012-10-01 5
## 3 NA 2012-10-01 10
## 4 NA 2012-10-01 15
## 5 NA 2012-10-01 20
## 6 NA 2012-10-01 25
- Remocao dos valores nulos
- Remove NA values
data <- na.omit(data_na)
head(data)
## steps date interval
## 289 0 2012-10-02 0
## 290 0 2012-10-02 5
## 291 0 2012-10-02 10
## 292 0 2012-10-02 15
## 293 0 2012-10-02 20
## 294 0 2012-10-02 25
What is mean total number of steps taken per day?
- Filtro os dados para passos diarios
- Apply the filter of steps taken by day
passos <- data %>%
group_by(date) %>%
summarise(steps = sum(steps))
head(passos)
## # A tibble: 6 x 2
## date steps
## <fct> <int>
## 1 2012-10-02 126
## 2 2012-10-03 11352
## 3 2012-10-04 12116
## 4 2012-10-05 13294
## 5 2012-10-06 15420
## 6 2012-10-07 11015
- Obtemos a media de passos por dia
- obtain the mean of steps by day
media <- mean(passos$steps)
paste("A media foi de: ", media, sep = " " )
## [1] "A media foi de: 10766.1886792453"
paste("The mean is: ", media, sep = " " )
## [1] "The mean is: 10766.1886792453"
- Obtemos a mediana de passos por dia
- obtain the mean of steps by day
mediana <- median(passos$steps)
paste("A mediana foi de: ", mediana, sep = " " )
## [1] "A mediana foi de: 10765"
paste("The median is: ", mediana, sep = " " )
## [1] "The median is: 10765"
- Monto a plotagem dos dados
- Plot the data
hist(passos$steps, xlab = "Numero de passos por dia", ylab = "Frequencia", main = "Histograma do Numero Total de Passos por Dia", col = "pink")
#abline(v = media, col="green", lwd=3, lty=3)
abline(v=c(media,mediana), col=c("green", "red"), lty=c(3,2), lwd=c(3, 3))
legend("topright", c("Mean", "Median"), fill=c("Green", "red"))
What is the average daily activity pattern?
- Fazemos um pipeline para obter a media de passadas por intervalos de data
- Make a pu=ipeline to obtain the average deily activity
media_de_passadas <- data %>%
group_by(interval) %>%
summarise(steps = mean(steps))
head(media_de_passadas)
## # A tibble: 6 x 2
## interval steps
## <int> <dbl>
## 1 0 1.72
## 2 5 0.340
## 3 10 0.132
## 4 15 0.151
## 5 20 0.0755
## 6 25 2.09
- Plot the data
- Monto a plotagem dos dados
plot(media_de_passadas$interval, media_de_passadas$steps, type="l", xlab="Intervalo de 5 minutos", ylab="Media de Passadas",main="Media Diaria de Passadas por Intervalo")
- The maximum number of steps by interval
- O maior numero de passadas por intervalo
maior_intervalo <- media_de_passadas[which.max(media_de_passadas$steps),1]
paste("O Maior numero de passadas ao longo do dia foi: ", maior_intervalo, sep = " ")
## [1] "O Maior numero de passadas ao longo do dia foi: 835"
paste("The maximum number of steps by interval was: ", maior_intervalo, sep = " ")
## [1] "The maximum number of steps by interval was: 835"
Imputing missing values
- Obtain the number of NAs
- Obtemos o numero de valores nulos
size <- length(which(is.na(data_na$steps)))
paste("Numero de NAs: ", size, sep = " " )
## [1] "Numero de NAs: 2304"
paste("Number of NAs: ", size, sep = " " )
## [1] "Number of NAs: 2304"
- Sunstitui-se os NAs pelo numero medio de passadas no dataset
- Replace NAS with the mean of steps without NAs in this dataset
data_na$steps = ifelse(is.na(data_na$steps), mean(data_na$steps, na.rm = T), data_na$steps)
head(data_na)
## steps date interval
## 1 37.3826 2012-10-01 0
## 2 37.3826 2012-10-01 5
## 3 37.3826 2012-10-01 10
## 4 37.3826 2012-10-01 15
## 5 37.3826 2012-10-01 20
## 6 37.3826 2012-10-01 25
- Obtain the number of NAs
- Obtemos o numero de valores nulos
size <- length(which(is.na(data_na$steps)))
paste("Numero de NAs: ", size, sep = " " )
## [1] "Numero de NAs: 0"
paste("Number of NAs: ", size, sep = " " )
## [1] "Number of NAs: 0"
- Utilizo a funcao tapply para obtera soma de passos por data
- Use tapply function to obtain the sum of steps by day
filtro <- tapply(data_na$steps, data_na$date, sum)
head(filtro)
## 2012-10-01 2012-10-02 2012-10-03 2012-10-04 2012-10-05 2012-10-06
## 10766.19 126.00 11352.00 12116.00 13294.00 15420.00
- Converto as datas para milesegundos
- Convert dates to mileseconds format
data_na$date <- as.POSIXct(data_na$date)
## Warning in strptime(xx, f <- "%Y-%m-%d %H:%M:%OS", tz = tz): unknown
## timezone 'zone/tz/2017c.1.0/zoneinfo/America/Sao_Paulo'
- Crio a variavel mes e extrai seu valor do atributo date
- Create the mes variable and obtain your value of date atribute
data_na$mes <- as.numeric(format(data_na$date, "%m"))
- Media de passos por dia
- Avera of steps taken each day
media_dia <- mean(filtro)
- Mediana de passos por dia
- Median of steps taken each day
mediana_dia <- median(filtro)
- Crio o histrograma com o numero de passos por dia apos inserir os valores nulos tratados
- Create the histogram of number steps taken each day after treat de minning values
hist(filtro, xlab = "Numero de Passos por dia", ylab = "Frequencia", main = "Histograma com o numero total de passos por dia com NAs tratados", col = "#2073d4")
abline(v=c(media_dia,mediana_dia), col=c("green", "red"), lty=c(3,2), lwd=c(3, 3))
legend("topright", c("Mean", "Median"), fill=c("Green", "red"))
Are there differences in activity patterns between weekdays and weekends?
- Criamos a coluna dia com o dia da semama obtido do atributo date
- Create the dia column with the weekday obtained from de date attibute
data_na$dia <- weekdays(data_na$date)
head(data_na)
## steps date interval mes dia
## 1 37.3826 2012-10-01 0 10 Monday
## 2 37.3826 2012-10-01 5 10 Monday
## 3 37.3826 2012-10-01 10 10 Monday
## 4 37.3826 2012-10-01 15 10 Monday
## 5 37.3826 2012-10-01 20 10 Monday
## 6 37.3826 2012-10-01 25 10 Monday
- Aplicamos o filtro para criar as categorias ‘weekend’ e ‘weekday’
- Apply filter to create the categories ‘weekend’ and ‘weekday’ *My computer is in portuguese PT_BR and the function weekdays can convert the weekdays to [segunda, terca, quarta, quinta…] instead [monday, tuesday, ….]. My option to solve this is include two modes of days to execurrute correct a function to obtain the classes weekend ou weekday
Sys.setlocale("LC_ALL","English")
## Warning in Sys.setlocale("LC_ALL", "English"): OS reports request to set
## locale to "English" cannot be honored
## [1] ""
data_na$tipo_dia <- ifelse(data_na$dia == "sabado" | data_na$dia == "domingo" | data_na$dia == "Saturday" | data_na$dia == "Sunday", "Weekend", "Weekday")
head(data_na)
## steps date interval mes dia tipo_dia
## 1 37.3826 2012-10-01 0 10 Monday Weekday
## 2 37.3826 2012-10-01 5 10 Monday Weekday
## 3 37.3826 2012-10-01 10 10 Monday Weekday
## 4 37.3826 2012-10-01 15 10 Monday Weekday
## 5 37.3826 2012-10-01 20 10 Monday Weekday
## 6 37.3826 2012-10-01 25 10 Monday Weekday
- Converter a variavel tipo_dia para um tipo categorico
- Convert tipo_dia to factor variable
data_na$tipo_dia <- as.factor(data_na$tipo_dia)
summary(data_na$tipo_dia)
## Weekday Weekend
## 12960 4608
- Crio o dataframe medias_dias com as medias de passadas aos dias uteis e fins de semana
- Create a dataframe medias_dias with the average of the weekdays and weekend
medias_dias <- data_na %>%
group_by(tipo_dia,interval) %>%
summarise(steps = mean(steps))
head(medias_dias)
## # A tibble: 6 x 3
## # Groups: tipo_dia [1]
## tipo_dia interval steps
## <fct> <int> <dbl>
## 1 Weekday 0 7.01
## 2 Weekday 5 5.38
## 3 Weekday 10 5.14
## 4 Weekday 15 5.16
## 5 Weekday 20 5.07
## 6 Weekday 25 6.30
- Plotagem dos dados medios para dias uteis e fins de semana
- Plot data average of weekdays and weekend
ggplot(data = medias_dias, aes(x = interval, y = steps)) +
facet_grid(tipo_dia ~ .) +
geom_line(stat = "identity", aes(colour = tipo_dia)) +
ggtitle("No de passadas por intervalo por tipo de Dia") +
labs(x="Intervalo", y=expression("Numero de passos"))
Alternative Mode
- Primeiro criamos dataFrames para dias uteis e fins de semana
- First create a weekdays and weekend dataFrames
weekday_db <- medias_dias[medias_dias$tipo_dia == "Weekday", ]
weekday_db
## # A tibble: 288 x 3
## # Groups: tipo_dia [1]
## tipo_dia interval steps
## <fct> <int> <dbl>
## 1 Weekday 0 7.01
## 2 Weekday 5 5.38
## 3 Weekday 10 5.14
## 4 Weekday 15 5.16
## 5 Weekday 20 5.07
## 6 Weekday 25 6.30
## 7 Weekday 30 5.61
## 8 Weekday 35 6.01
## 9 Weekday 40 4.98
## 10 Weekday 45 6.58
## # ... with 278 more rows
weekend_db <- medias_dias[medias_dias$tipo_dia == "Weekend", ]
weekend_db
## # A tibble: 288 x 3
## # Groups: tipo_dia [1]
## tipo_dia interval steps
## <fct> <int> <dbl>
## 1 Weekend 0 4.67
## 2 Weekend 5 4.67
## 3 Weekend 10 4.67
## 4 Weekend 15 4.67
## 5 Weekend 20 4.67
## 6 Weekend 25 7.92
## 7 Weekend 30 4.67
## 8 Weekend 35 4.67
## 9 Weekend 40 4.67
## 10 Weekend 45 5.05
## # ... with 278 more rows
- Criamos as plotagens dos dois dataFrames
- Create the plots of dataFrames
data_weekdays<- ggplot(weekday_db) +
geom_line(aes(x = interval, y = steps), colour = "red", linetype = "dashed") +
xlab("Intervalos de Dias Uteis") +
ylab("Passsos")
data_weekends <- ggplot(weekend_db) +
geom_line(aes(x = interval, y = steps), colour = "blue", linetype = "dotted") +
xlab("Intervalos de Fins de Semana") +
ylab("Passsos")
- Instalamos e Carregamos a biblioteca para criar grid de plotagens
- Install and load the library to create a plot grid
#install.packages('gridExtra')
library(gridExtra)
##
## Attaching package: 'gridExtra'
## The following object is masked from 'package:dplyr':
##
## combine
- Exibimos o grid criado
- Displays a created grid
grid.arrange(data_weekdays,data_weekends, ncol=1, top = "Dias Uteis x Fim de Semana (Media de Passadas)", left = "Y", bottom = "X")
