BixiData: BIXI Data Class

In BKTR: Bayesian Kernelized Tensor Regression

BIXI Data Class

Description

R6 class encapsulating all BIXI dataframes. It is also possible to use a light version of the dataset by using the is_light parameter. In this case, the dataset is reduced to its first 25 stations and first 50 days. The light version is only used for testing and short examples.

Public fields

departure_df

The departure dataframe

spatial_features_df

The spatial features dataframe

temporal_features_df

The temporal features dataframe

spatial_positions_df

The spatial positions dataframe

temporal_positions_df

The temporal positions dataframe

data_df

The data dataframe

is_light

Whether the light version of the dataset is used

Methods

Public methods

• BixiData$new()

• BixiData$clone()

---

Method new()

Initialize the BIXI data class

Usage

BixiData$new(is_light = FALSE)

Arguments

is_light

Whether the light version of the dataset is used, defaults to FALSE.

Returns

A new BIXI data instance

---

Method clone()

The objects of this class are cloneable with this method.

Usage

BixiData$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Create a light BIXI data collection instance containing multiple dataframes
# This only uses the first 25 stations and 50 days of the full dataset
bixi_data <- BixiData$new(is_light = TRUE)
# Dataframe containing the position (latitude and longitude) of M stations
bixi_data$spatial_positions_df
# Dataframe containing the time position of N days (O to N-1)
bixi_data$temporal_positions_df
# Dataframe with spatial and temporal features for each day and station (M x N rows)
bixi_data$data_df

BKTR documentation built on Sept. 12, 2024, 7:04 a.m.