https://github.com/coatless-rpkg/ucimlrepo
An unofficial R port of the Python package to download data off of the UCI ML repository
https://github.com/coatless-rpkg/ucimlrepo
data data-science machine-learning rstats statistics uci-machine-learning uci-machine-learning-repository web-api
Last synced: 6 months ago
JSON representation
An unofficial R port of the Python package to download data off of the UCI ML repository
- Host: GitHub
- URL: https://github.com/coatless-rpkg/ucimlrepo
- Owner: coatless-rpkg
- License: other
- Created: 2024-07-19T06:14:08.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2024-07-28T05:29:07.000Z (10 months ago)
- Last Synced: 2024-07-28T06:31:09.848Z (10 months ago)
- Topics: data, data-science, machine-learning, rstats, statistics, uci-machine-learning, uci-machine-learning-repository, web-api
- Language: R
- Homepage: http://r-pkg.thecoatlessprofessor.com/ucimlrepo/
- Size: 1.67 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
- License: LICENSE
Awesome Lists containing this project
README
---
output: github_document
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# ucimlrepo 
[](https://github.com/coatless-rpkg/ucimlrepo/actions/workflows/R-CMD-check.yaml)
The goal of `ucimlrepo` is to download and import data sets directly into R
from the [UCI Machine Learning Repository](https://archive.ics.uci.edu/).
> [!IMPORTANT]
>
> This package is an unoffical port of the [Python `ucimlrepo` package](https://github.com/uci-ml-repo/ucimlrepo).
> [!NOTE]
>
> Want to have datasets alongside a help documentation entry?
>
> Check out the [`{ucidata}`](https://github.com/coatless-rpkg/ucidata) R package!
> The package provides a small selection of data sets from the
> UC Irvine Machine Learning Repository alongside of help entries.
## Installation
You can install the development version of ucimlrepo from [GitHub](https://github.com/) with:
``` r
# install.packages("remotes")
remotes::install_github("coatless-rpkg/ucimlrepo")
```
## Usage
To use `ucimlrepo`, load the package using:
```{r}
#| label: load-pkg
library(ucimlrepo)
```
With the package now loaded, we can download a dataset using the `fetch_ucirepo()` function
or use the `list_available_datasets()` function to view a list of available datasets.
### Download data
For example, to download the `iris` dataset, we can use:
```{r}
#| label: download dataset
# Fetch a dataset by name
iris_by_name <- fetch_ucirepo(name = "iris")
names(iris_by_name)
```
There are many levels to the data returned. For example, we can extract the
original data frame containing the `iris` dataset using:
```{r}
iris_uci <- iris_by_name$data$original
head(iris_uci)
```
Alternatively, we could retrieve two data frames, one for the features and one for the targets:
```{r}
iris_features <- iris_by_name$data$features
iris_targets <- iris_by_name$data$targets
```
We can then view the first few rows of each data frame:
```{r}
head(iris_features)
head(iris_targets)
```
Alternatively, you can also directly query by using an ID found by using
`list_available_datasets()` or by looking up the dataset on the UCI ML Repo website:
```{r}
#| label: download dataset by id
#| eval: false
# Fetch a dataset by id
iris_by_id <- fetch_ucirepo(id = 53)
```
### View list of data sets
We can also view a list of data sets available for download
using the `list_available_datasets()` function:
```{r}
#| label: list-packages
#| eval: false
# List available datasets
list_available_datasets()
```
> [!NOTE]
>
> Not all 600+ datasets on UCI ML Repo are available for download using the package.
> The current list of available datasets can be viewed [here](https://archive.ics.uci.edu/datasets?Python=true&skip=0&take=25&sort=desc&orderBy=NumHits).
>
> If you would like to see a specific dataset added, please submit a comment
> on an [issue ticket in the upstream repository](https://github.com/uci-ml-repo/ucimlrepo/issues/9).