In Jofou/jofou.lib: Useful Personal Function

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "80%"
)

jofou.lib

The goal of jofou.lib is to regroup all the functions that are useful for me to work efficiently.

Installation

You can install the lastest version of jofou.lib with:

devtools::install_github("jofou/jofou.lib")

EDA Functions

Example

This are basics examples which shows you how use the my_inspect group function:

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_cat()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_cor()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_imb()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_na()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_num()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_inspect_types()

I also have a couples of other functions to show distributions of numeric and categorical variables:

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_num_dist()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_corr_num_graph()

library(tidyverse)
library(jofou.lib)

iris %>% 
  my_cat_dist()

Utilities Functions

Example

These are basic examples that show you how to use my utilities functions:

library(tidyverse)
library(jofou.lib)

iris %>%
  mutate(cat_Sepal.Length=round(Sepal.Length, digits = 0)) %>%
  group_by(cat_Sepal.Length) %>%
  summarise(mode_species=calculate_mode(Species))

library(tidyverse)
library(jofou.lib)

iris %>%
  filter(Species %ni% "setosa") %>%
  group_by(Species) %>%
  summarise(nb=dplyr::n())

ML Functions

Example

These are basic examples that show you how to use my machine learning utilities functions:

library(tidyverse)
library(lubridate)
library(timetk)
library(parsnip)
library(rsample)
library(modeltime)

# Data
data_prepared_tbl <- m4_monthly %>% filter(id == "M750")

# Split Data 80/20
splits <- initial_time_split(data_prepared_tbl, prop = 0.9)

# Model: auto_arima
model_fit_arima <- arima_reg() %>%
   set_engine(engine = "auto_arima") %>%
   fit(value ~ date, data = training(splits))

# Calibrate and plot
calibrate_and_plot(model_fit_arima, type="testing")