README.md

SCRCdataAPI

Functions to generate and process data files for the SCRC data pipeline.

• Installation
• Create array
• Create table
• Create distribution
• Create point-estimate

Installation

Note to Linux users: installing devtools may require libcurl4-openssl-dev, libhdf5-dev libudunits2-dev, and libgdal-dev.

library(devtools)
install_github("ScottishCovidResponse/SCRCdataAPI")

and load it into R:

library(SCRCdataAPI)

Create table

In the following example, we will populate test_table.h5 with table type data downloaded from https://github.com/ScottishCovidResponse/simple_network_sim/blob/master/sample_output_files/sample-1591198608.csv.

download.file("https://github.com/ScottishCovidResponse/simple_network_sim/raw/master/sample_output_files/sample-1591198608.csv", "sample.csv")
sample <- read.csv("sample.csv")

We want to put this data in a directory called sample1:

# Create *.h5 file
create_table(filename = "test_table.h5", component = "sample1", df = sample)

Note that the filename argument can take the name of a file you want to create, or an existing *.h5 file.

To read the data file:

read_table(filename = "test_table.h5", component = "sample1")

Create array

In the following example, we will populate "array.h5" with array type data.

First we generate an array:

array <- matrix(1:10, 5)
colnames(array) <- paste0("age", 1:2)
rownames(array) <- paste0("dz", 1:5)

Then we extract row and column names (and specify descriptors):

dimension_names <- list(`area names` = rownames(array), 
`age classes` = colnames(array))

We want to put this data in a directory called dz, in a subdirectory called total:

# Create *.h5 file
create_array(filename = "test_array.h5", component = "dz/total", array = array, dimension_names = dimension_names)

To read the data file:

read_array(filename = "test_array.h5", component = "dz/total")

Create distribution

In the following example, we populate "test_distribution.toml":

# Create *.toml file
create_distribution(filename = "test_distribution.toml", path = "data-raw", name = "latency", distribution = "gamma", parameters = list(shape = 2.0, scale = 3.0))

To read the toml:

read_distribution(filename = "data-raw/test_distribution.toml")

Create point-estimate

In the following example, we populate "test_number.toml" with a single point-estimate:

# Create *.toml file
create_estimate(filename = "test_number.toml", path = "data-raw", parameters = list(asymptomatic_period = 192.0))

To include multiple point-estimates:

# Create *.toml file
create_estimate(filename = "test_number.toml", path = "data-raw", parameters = list(asymptomatic_period = 192.0, latent_period = 123.12))

To read the toml:

read_estimate(filename = "data-raw/test_number.toml")