imangr / bysykkel

Functions to get Norwegian city bike data

bysykkel

Overview

bysykkel is an R package that provides functions that simplifies the task of gathering Norwegian city bike data for data analysis. bysykkel provides functions to read city bike data directly to R or download it to your R session’s working directory.

• read_trips_data() reads bike trip records to R as a data frame.
• fread_trips_data() fast reads bike trip records to R as a data frame by utilizing fread() from data.table.
• dl_trips_data() downloads bike trip records to your working directory
• get_api_data() gets real-time data from the specified city bike API service.

bysykkel lets you, the user, focus on data exploration, visualization, statistical analysis, and building machine learning models on Norwegian city bike data, by simplifying the task of getting the data. Indeed, the purpose of bysykkel is to reduce time spent on getting Norwegian city bike data, and lower barriers to start analyzing it.

The package name, bysykkel, is the Norwegian word for “city bikes”, where by means “city”, and sykkel means “bike” (or “bicycle”).

Installation

You can install the released version of bysykkel from CRAN with:

install.packages("bysykkel")

Development version

Alternatively, you can install a development version of bysykkel from GitHub to get bug fixes or new features before the next package version is released on CRAN. To install the development version, you can use devtools to install bysykkel from GitHub.

#! install.packages("devtools")
devtools::install_github("imangR/bysykkel")

Bike data

bysykkel currently retrieves data from three[1] city bike services in Norway that make bike data publicly available[2]:

• Oslo City Bike
• Bergen City Bike
• Trondheim City Bike

Each city bike service provide two data-related services:

• Historical trip data
• Real-time data

The historical trip data is available both as a CSV-file, and a JSON-file, that contains monthly anonymized historical bike trip data. Real-time data is available in the GBFS format, and must be accessed with each city bike’s API service, which provide information about

• bike and dock availability;
• stations (including geolocation, addresses and station descriptions); and
• machine readable basic information about the city bike service.

The data is made available under the Norwegian License for Open Data 2.0, abbreviated as NLOD 2.0, which you can read about here.

Examples

Read bike trips data to R

library(bysykkel)

# Get bike trip data for April, 2019 for Oslo as a data frame
oslo_trips <- read_trips_data(year = 2019, month = 04, city = "Oslo")

# Get winter bike data for January, 2019 for Oslo as a data frame
oslo_trips <- read_trips_data(2019, 1, "Oslo")

# Fast read bike data from June to August in 2018 for Bergen with `lapply()`,
# and `rbind()` the resulting `list` with `do.call()` to get a data frame

#! install.packages("data.table")

bergen_trips <- lapply(06:08, fread_trips_data, year = 2018, city = "Bergen")
bergen_trips <- do.call(rbind, bergen_trips)

# Alternatively, use `map_dfr()` from `purrr` instead of `lapply()`,
# `rbind()`, and `do.call() to get the same result: a data frame

#! install.packages("purrr")

library(purrr)

bergen_trips <- map_dfr(6:8, fread_trips_data, year = 2018, city = "Bergen")

NB! I recommend that you use fread_trips_data() to fast read city bike data, especially if you want to read bike data for several months.

NB! data.table is not automatically installed with bysykkel, and must be installed separately with install.packages("data.table") if you want to use fread_trips_data().

Download bike trips data

library(bysykkel)

# Download bike trip data for April 2019 for Trondheim
dl_trips_data(2019, 04, "Trondheim", filetype = "JSON")
#> The CSV-file is downloaded to your R session's current working directory

# Download bike trip data for summer 2018 for Oslo
lapply(06:08, dl_trips_data, year = 2018, city = "Oslo", filetype = "CSV")
#> The CSV-file for each month is downloaded to your R session's working directory

Get real-time data from the API service

NB! Please read each City Bike’s guide on how to correctly use their API service before using get_api_data(). See Oslo City Bike’s guide as an example.

The return_df argument in get_api_data() specifies whether you want to return the result as a data frame. If return_df = FALSE (default), then the function returns a list that contains a data frame, and a number that represents the datetime (in POSIX format) of when you made the API request.

library(bysykkel)

# Get API data on bike stations as a data frame
oslo_stations <- get_api_data(client_id = "myname-myapp", 
                              data = "stations",
                              city = "Oslo",
                              return_df = TRUE)

# Get API data for bike availability as a list that contains a data frame, and
# a number that represents the (POSIX) time of when you made the API request
bergen_availability <- get_api_data(client_id = "mycompany-myservice",
                                    data = "availability",
                                    city = "Bergen",
                                    return_df = FALSE)

# Get API data on bike system information
trondheim_system <- get_api_data("Ola Nordmann-bike dashboard",
                                 "system",
                                 "Trondheim",
                                 return_df = FALSE)

File an issue or suggest an improvement

If you want to report a discovered bug, raise some other issue, or suggest an improvement to bysykkel, then please file an issue on GitHub. For bugs, please file a minimal reproducible example.

Known issues

No issues have been identified at this time in version 0.3.1.

---

1. It used to be four services, but the Oslo Winter Bike service has been shut down, and associated data is now unavailable. The bysykkel package has updated all functions in version 0.3.1 to remove any interface to the Oslo Winter Bike service.

2. Bike data for Bærum City Bike