README.md
In CamembR/tdvhackaviz2020: ToulouseDataViz Hackaviz2020 data

tdvhackaviz2020

This package contains ready to use R data package of the datasets provided by ToulouseDataViz in the frame of the Hackaviz2020.

This package contains the following datasets with two flavors original and tidy suffixed by _td:

nuitees / nuitees_td
capacites
- capacites_td: whithout weekly data (pop and hbgt_*)
- capacites_sem_td: with only weekly data capa per week (sem_*)
par_origines / par_origines_td
Complements: Split the complements dataset in three datatesets
- comp_evenements
- comp_departemets
- comp_pays

It also contains time series data formatted as a tsibble.


# Install the development version from GitHub

devtools::install_github("CamembR/tdvhackaviz2020")

# Use tidy datasets

tdvhackaviz2020::nuitees_td

# A tibble: 4,745 x 3
# date       dep   nuitees
# <date>     <fct>   <dbl>
# 1 2018-01-01 09      36427
# 2 2018-01-01 11      60093

# Use tsibble

tdvhackaviz2020::tourists_tsbl() %>%
  group_by_key() %>%
  index_by(year_month = ~ year(.)) %>% # monthly aggregates
  summarise(
    capa = mean(capa, na.rm = TRUE),
    nuitees = mean(nuitees, na.rm = TRUE),
    diff = capa - nuitees
  ) %>% 
  arrange(desc(nuitees)) %>% 
  top_n(3, nuitees)

# A tsibble: 3 x 5 [?]
# Key:       dep [3]
# dep   year_month    capa nuitees   diff
# <fct>      <dbl>   <dbl>   <dbl>  <dbl>
# 1 34          2018 135983. 137096. -1112.
# 2 66          2018  86810.  87599.  -789.
# 3 31          2018  78069.  78823.  -754.

Tidy data is a standard way of mapping the meaning of a dataset to its structure. A dataset is messy or tidy depending on how rows, columns and tables are matched up with observations, variables and types. In tidy data:

Each variable forms a column.

Each observation forms a row.

Each type of observational unit forms a table.

--- Tidy data vignette

See help for description and usage.

?nuitees_td

nuitees_td

# A tibble: 4,745 x 3
date       dep    nuitees
<date>     <chr>    <dbl>
1 2018-01-01 dpt_09   36427
2 2018-01-01 dpt_11   60093

# To get total_occitanie
nuitees_td %>%
  group_by(date) %>%
  summarise(total_occitanie = sum(nuitees))

# A tibble: 365 x 2
date       total_occitanie
<date>               <dbl>
1 2018-01-01          927930
2 2018-01-02          444144