}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(yfix), y)

}

broadtime(testdat$timestamp)

testdat <- data.frame(timestamp = c("2023-05-02 03:30:00 EDT", "2023-05-02 04:15:00 EDT", "2023-05-02 07:30:00 EDT"))

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(yfix), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(y), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(yfix), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

testdat <- data.frame(timestamp = c("2023-05-02 03:30:00 EDT", "2023-05-02 04:15:00 EDT", "2023-05-02 07:30:00 EDT"))

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

hour(yfix)

yfix <- ymd_hms(testdat$timestamp, tz = "America/New_York")

hour(yfix)

hour(yfix) < 6

rollbackward(ymd_hms(yfix))

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), yfix)

ymd_hms(yfix)

rollbackward(ymd_hms(yfix))

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

mutate(y = case_when(

hour(yfix < 6) ~ rollbackward(yfix)))

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

mutate(yfix = case_when(

hour(yfix < 6) ~ rollbackward(yfix)))

}

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

rollbackward(ymd_hms(yfix))

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, newy = rollbackward(ymd_hms(yfix)), newy=y)

}

broadtime(testdat$timestamp)

testdat %>% add_column(finaltime = NA) %>%

yfix <- ymd_hms(y, tz = "America/New_York")

testdat %>% add_column(finaltime = NA) %>%

yfix <- ymd_hms(timestamp, tz = "America/New_York")

testdat %>% add_column(finaltime = NA) %>%

ymd_hms(timestamp, tz = "America/New_York")

testdat %>% add_column(finaltime = NA)

testdat %>% add_column(finaltime = NA) %>%

ymd_hms(timestamp, tz = "America/New_York")

testdat %>% add_column(finaltime = NA) %>%

ymd_hms(timestamp, tz = "America/New_York")

testdat %>% add_column(finaltime = NA) %>%

str_remove("EDT")

test2 <- data.frame(timestamp = c("2023-05-02 03:30:00", "2023-05-02 04:15:00 EDT", "2023-05-02 07:30:00 EDT"))

test2 %>% add_column(finaltime = NA) %>%

ymd_hms(timestamp, tz = "America/New_York")

test2 <- data.frame(timestamp = c("2023-05-02 03:30:00", "2023-05-02 04:15:00", "2023-05-02 07:30:00"))

test2 %>% add_column(finaltime = NA) %>%

ymd_hms(timestamp, tz = "America/New_York")

test2 %>%

ymd_hms(timestamp, tz = "America/New_York")

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(y)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

as.POSIXct(strptime(y, "%Y-%m-%d %H:%M:%S"))

t <- strftime(y, format="%H:%M:%S")

if (t < as.POSIXct("06:00:00", format="%H:%M:%S"){

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(y)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

as.Date(ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y))

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

ymd_hms(ifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y))

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y))

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

fifelse(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), y)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollbackward(ymd_hms(yfix)), yfix)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollback(ymd_hms(yfix)), yfix)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, rollback(ymd_hms(yfix)), yfix)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, yfix - 1, yfix)

}

broadtime(testdat$timestamp)

if_else(hour(yfix) < 6, yfix - days(1), yfix)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, yfix - days(1), yfix)

}

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(yfix) < 6, yfix - days(1), yfix)

}

broadtime(testdat$timestamp)

broadtime <- function(y){

yfix <- ymd_hms(y, tz = "America/New_York")

print(yfix)

if_else(hour(y) < 6, y - days(1), y)

}

broadtime(testdat$timestamp)

testdat %>% mutate(broadtime = if_else(hour(yfix) < 6, yfix - days(1), yfix))

remotes::update_packages()

knitr::opts_chunk$set(echo = TRUE)

knitr::opts_knit$set(root.dir = 'C:/Users/larn_/OneDrive - Louisiana State University/Career')

df <- read.csv("C:/Users/larn_/OneDrive - Louisiana State University/Career/fitness_class_2212.csv")

df %>% select(where(is.na))

# Load necessary libraries

library(tidyverse)

install.packages("vctrs")

install.packages("vctrs")

knitr::opts_chunk$set(echo = TRUE)

knitr::opts_knit$set(root.dir = 'C:/Users/larn_/OneDrive - Louisiana State University/Career')

# Load necessary libraries

library(tidyverse)

rnorm(50)

x <- rnorm(50)

y <- rnorm(50)

dat <- as.data.frame(x, y)

dat <- data.frame(x, y)

cor(dat$x, dat$y)

cov(dat$x, dat$y)

?cor

cor.test(x, y, dat)

?cor.test

cor.test(x, y, data = dat)

install.packages("usethis")

install.packages("fs")

install.packages("here")

library(tidyverse)

library(rmarkdown)

library(usethis)

library(fs)

library(here)

use_this("rstats-wtf/wtf-explore-libraries", destdir = "C:/Users/larn_/Documents/GitHub/Workshops-Certifications/R-workflow")

use_course("rstats-wtf/wtf-explore-libraries", destdir = "C:/Users/larn_/Documents/GitHub/Workshops-Certifications/R-workflow")

# A first exploration of installed packages ------------------------------------

library(fs)

library(tidyverse)

# In which libraries does R search for packages? -------------------------------

# your personal computing environment may return one or more locations

.libPaths()

# What is your default library? ------------------------------------------------

.Library

# Confirm the library supplied is, in fact, the default library. ---------------

identical(.Library, .libPaths())

identical(.Library, .libPaths()[2])

# Huh? Maybe this is an symbolic link issue?

identical(path_real(.Library), path_real(.libPaths()))

identical(path_real(.Library), path_real(.libPaths()[2]))

# Create a tibble of all installed packages ------------------------------------

df_pkgs <- installed.packages() |>

as_tibble()

View(df_pkgs)

# How many packages are installed? ---------------------------------------------

nrow(df_pkgs)

# Explore the packages - count some things. ------------------------------------

# For example, tabulate by LibPath, Priority, or both.

df_pkgs |>

count(LibPath, Priority)

# What proportion need compilation?

df_pkgs |>

count(NeedsCompilation) |>

mutate(prop = n / sum(n))

# What version of R they were built on?

df_pkgs |>

count(Built) |>

mutate(prop = n / sum(n))

# Is every installed package either base or recommended?

df_pkgs |>

count(Priority %in% c("base", "recommended"))

# Explore package naming conventions (all lower case, contains '.', etc)

df_pkgs |>

count(str_detect(Package, "\\."))

# Use `fields` argument to installed.packages() to get more info.

# ?installed.packages

df_pkgs_url <- installed.packages(fields = "URL") |>

as_tibble()

# What proportion of packages have a URL that point to github?

df_pkgs_url |>

mutate(github = str_detect(URL, "github")) |>

count(github) |>

mutate(prop = n / sum(n))

View(df_pkgs_url)

.libPahts()

.libPaths()

.Library

?rm()

?options

?Sys.setenv

# Use a blank slate to clear environment and remove packages

usethis::use_blank_slate()

# Create a new folder to make a project with code

usethis::create_project("C:/Users/larn_/Documents/GitHub/Workshops-Certifications/R-workflow")

here::dr_here()

library(shiny)

library(dplyr)

# Import Data

setwd("C:/Users/larn_/Documents/GitHub/Workshops-Certifications/DataScienceCertification")

gym_attendance_data <- read.csv("fitness_class_2212-clean.csv")

## Build App

ui <- fluidPage(

titlePanel("Gym Attendance Tracker"),

#Create a sidebar with different filters

sidebarLayout(

sidebarPanel(

# Create a checkbox group to filter by class category

checkboxGroupInput("categories", "Class Category:",

gym_attendance_data$category,

selected = gym_attendance_data$category[1]),

# Create a slider to filter by number of months as a member

sliderInput("months_as_member", "Months as Member:", 1,

max(gym_attendance_data$months_as_member),

value = range(gym_attendance_data$months_as_member)),

# Create a drop-down menu to filter by day of the week

selectInput("days_of_week", "Day of the Week:",

c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"),

selected = gym_attendance_data$days_of_week[1]),

# Create a drop-down menu to filter by time of class

selectInput("time", "Time:", c("Morning", "Afternoon", "Evening"),

selected = gym_attendance_data$time[1])),

mainPanel(

# Create a table to show the filtered gym attendance data

tableOutput("gym_attendance_table"),

# Create a plot to show the overall gym attendance rate

plotOutput("gym_attendance_plot")

)

)

)

server <- function(input, output) {

#Create a reactive expression to filter the gym attendance data

filtered_gym_attendance_data <- reactive({

gym_attendance_data %>%

filter(category %in% input$categories,

months_as_member >= input$months_as_member,

days_of_week == input$days_of_week,

time == input$time)

})

#Create a reactive expression to calculate the overall gym attendance rate

overall_gym_attendance_rate <- reactive({

mean(filtered_gym_attendance_data()$attended)

})

#Create a table to show the filtered gym attendance data

output$gym_attendance_table <- renderTable({

filtered_gym_attendance_data()

})

#Create a plot to show the overall gym attendance rate

output$gym_attendance_plot <- renderPlot({

plot(overall_gym_attendance_rate(), type = "p", main = "Overall Gym Attendance Rate")

})

}

shinyApp(ui, server)

View(gym_attendance_data)

runApp('ClassAttendance')

class(gym_attendance_data$category)

unique(gym_attendance_data$category)

runApp('ClassAttendance')

runApp('ClassAttendance')

runApp('ClassAttendance')

knitr::opts_knit$set(root.dir = 'C:/Users/larn_/Documents/GitHub/Predicting-Gym-Attendance')

# Load necessary libraries

library(tidyverse)

# Load necessary libraries

library(tidyverse)

library(randomForest)

library(caret)

library(tibble)

df <- read.csv("fitness_class_2212.csv")

which(is.na(df), arr.ind=TRUE)

unique(df$months_as_member)

# Make months numeric

df$months_as_member <- as.numeric(df$months_as_member)

class(df$weight)

min(df$weight, na.rm = TRUE)

max(df$weight, na.rm = TRUE)

# Replace weight NAs with column average

df <- df %>% mutate(weight = coalesce(weight, mean(weight)))

min(df$days_before)

max(df$days_before)

class(df$days_before)

# Remove 'days' in some rows of days_before

df$days_before <- gsub("days", "", as.character(df$days_before))

# Make days_before numeric

df$days_before <- as.numeric(df$days_before)

unique(df$day_of_week)

length(df$day_of_week[which(df$day_of_week == "Fri.")])

# Make day_of_week labels consistent

df$day_of_week[which(df$day_of_week == "Monday")] <- "Mon"

df$day_of_week[which(df$day_of_week == "Wednesday")] <- "Wed"

df$day_of_week[which(df$day_of_week == "Fri.")] <- "Fri"

unique(df$time)

class(df$time)

unique(df$category)

length(df$category[which(df$category == "-")])

class(df$category)

# Replace '-' categories with 'unknown'

df$category[which(df$category == "-")] <- "unknown"

unique(df$attended)

class(df$attended)

# Make attended a factor

df$attended <- as.factor(df$attended)

# Split the data into training and testing sets

set.seed(123)

train_indices <- sample(1:nrow(df), 0.7 * nrow(df)) # 70% for training

train_data <- df[train_indices, ]

test_data <- df[-train_indices, ]

# Train the random forest model

rf_model <- randomForest(attended ~ . - booking_id, data = train_data)

View(train_data)

View(train_data)

train_data$attended <- as.factor(train_data$attended)

# Train the random forest model

rf_model <- randomForest(attended ~ . - booking_id, data = train_data)

is.na(train_data$attended)

is.na(train_data$attended) == TRUE

which(is.na(train_data$attended))

which(is.na(train_data))

# Make months numeric

df$months_as_member <- as.numeric(df$months_as_member)

class(df$weight)

min(df$weight, na.rm = TRUE)

max(df$weight, na.rm = TRUE)

# Replace weight NAs with column average

df <- df %>% mutate(weight = coalesce(weight, mean(weight)))

min(df$days_before)

max(df$days_before)

class(df$days_before)

# Remove 'days' in some rows of days_before

df$days_before <- gsub("days", "", as.character(df$days_before))

# Make days_before numeric

df$days_before <- as.numeric(df$days_before)

unique(df$day_of_week)

length(df$day_of_week[which(df$day_of_week == "Fri.")])

# Make day_of_week labels consistent

df$day_of_week[which(df$day_of_week == "Monday")] <- "Mon"

df$day_of_week[which(df$day_of_week == "Wednesday")] <- "Wed"

df$day_of_week[which(df$day_of_week == "Fri.")] <- "Fri"

unique(df$time)

class(df$time)

unique(df$category)

length(df$category[which(df$category == "-")])

class(df$category)

# Replace '-' categories with 'unknown'

df$category[which(df$category == "-")] <- "unknown"

unique(df$attended)

class(df$attended)

# Make attended a factor

df$attended <- as.factor(df$attended)

# Split the data into training and testing sets

set.seed(123)

train_indices <- sample(1:nrow(df), 0.7 * nrow(df)) # 70% for training

train_data <- df[train_indices, ]

test_data <- df[-train_indices, ]

# Train the random forest model

rf_model <- randomForest(attended ~ . - booking_id, data = train_data)

which(is.na(train_data))

View(train_data)

# Replace weight NAs with column average

df <- df %>% mutate(weight = coalesce(weight, mean(weight)))

which(is.na(df))

View(df)

which(is.na(df$weight))

View(df)

# Replace weight NAs with column average

df$weight <- df %>% mutate(weight = coalesce(weight, mean(weight)))

which(is.na(df$weight))

df %>% mutate(weight = coalesce(weight, mean(weight)))

# Replace weight NAs with column average

df <- df %>% mutate(weight = replace_na(weight, mean(weight, na.rm = TRUE)))

# Replace weight NAs with column average

df <- df %>% mutate(weight = replace_na(weight, mean(weight)))

# Replace weight NAs with column average

df <- df %>% mutate(weight = coalesce(weight, mean(weight)))

df$weight <- as.numeric(df$weight)

class(df$weight)

df$weight <- as.numeric(df$weight)

df <- read.csv("fitness_class_2212.csv")

class(df$weight)

# Replace weight NAs with column average

df <- df %>% mutate(weight = coalesce(weight, mean(weight)))

# Remove 'days' in some rows of days_before

df$days_before <- gsub("days", "", as.character(df$days_before))

# Make days_before numeric

df$days_before <- as.numeric(df$days_before)

# Make day_of_week labels consistent

df$day_of_week[which(df$day_of_week == "Monday")] <- "Mon"

df$day_of_week[which(df$day_of_week == "Wednesday")] <- "Wed"

df$day_of_week[which(df$day_of_week == "Fri.")] <- "Fri"

# Replace '-' categories with 'unknown'

df$category[which(df$category == "-")] <- "unknown"

# Make attended a factor

df$attended <- as.factor(df$attended)

which(is.na(df))

df[3115, ]

df[3212, ]

df <- na.omit(df)

which(is.na(df))

?na.omit