Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hendersontrent/theftdlc
Analyse and interpret time-series features calculated from the theft R package
https://github.com/hendersontrent/theftdlc
data-science data-visualization machine-learning r statistics time-series
Last synced: 2 months ago
JSON representation
Analyse and interpret time-series features calculated from the theft R package
- Host: GitHub
- URL: https://github.com/hendersontrent/theftdlc
- Owner: hendersontrent
- License: other
- Created: 2024-02-23T02:13:16.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-10-04T08:54:01.000Z (3 months ago)
- Last Synced: 2024-10-05T07:15:04.129Z (3 months ago)
- Topics: data-science, data-visualization, machine-learning, r, statistics, time-series
- Language: R
- Homepage: https://hendersontrent.github.io/theftdlc/
- Size: 11.8 MB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.Rmd
- License: LICENSE
Awesome Lists containing this project
README
---
output: rmarkdown::github_document
---
# theftdlc 
[](https://www.r-pkg.org/pkg/theftdlc)
[](https://www.r-pkg.org/pkg/theftdlc)
Analyse and Interpret Time Series Features
```{r, include = FALSE}
knitr::opts_chunk$set(
comment = NA, fig.width = 14, fig.height = 10, cache = FALSE)
```
## Installation
You can install the stable version of `theftdlc` from CRAN:
```{r eval = FALSE}
install.packages("theftdlc")
```
You can install the development version of `theftdlc` from GitHub using the following:
```{r eval = FALSE}
devtools::install_github("hendersontrent/theftdlc")
```
## General purpose
The [`theft`](https://hendersontrent.github.io/theft/) package for R facilitates user-friendly access to a structured analytical workflow for the extraction of time-series features from six different feature sets (or a set of user-supplied features): `"catch22"`, `"feasts"`, `"Kats"`, `"tsfeatures"`, `"tsfresh"`, and `"TSFEL"`.
`theftdlc` extends this feature-based ecosystem by providing a suite of functions for analysing, interpreting, and visualising time-series features calculated using `theft`. Functionality including data quality assessments and normalisation methods, low dimensional projections (linear and nonlinear), data matrix and feature distribution visualisations, time-series classification machine learning procedures, statistical hypothesis testing, and various other statistical and graphical tools.
A high-level overview of how the `theft` ecosystem for R is typically accessed by users is shown below. Many more functions and options for customisation are available within the packages.
### What's in a name?
`theftdlc` means 'downloadable content' (DLC) for `theft`---just like you get [DLCs and expansions](https://en.bandainamcoent.eu/elden-ring/elden-ring/shadow-of-the-erdtree) for video games.
## Quick tour
`theft` and `theftdlc` combine to create an intuitive and efficient tidy feature-based workflow. Here is an example of a single code chunk that calculates features using [`catch22`](https://github.com/hendersontrent/Rcatch22) and a custom set of mean and standard deviation, and projects the feature space into an interpretable two-dimensional space using principal components analysis:
```{r, message = FALSE, warning = FALSE, fig.height=6, fig.width=6}
library(dplyr)
library(theft)
library(theftdlc)
calculate_features(data = theft::simData,
group_var = "process",
feature_set = "catch22",
features = list("mean" = mean, "sd" = sd)) %>%
project(norm_method = "RobustSigmoid",
unit_int = TRUE,
low_dim_method = "PCA") %>%
plot()
```
In that example, `calculate_features` comes from `theft`, while `project` and the `plot` generic come from `theftdlc`.
Similarly, we can perform time-series classification using a similar simple workflow to compare the performance of `catch22` against our custom set of the first two moments of the distribution:
```{r, message = FALSE, warning = FALSE}
calculate_features(data = theft::simData,
group_var = "process",
feature_set = "catch22",
features = list("mean" = mean, "sd" = sd)) %>%
classify(by_set = TRUE,
n_resamples = 5,
use_null = TRUE) %>%
compare_features(by_set = TRUE,
hypothesis = "null") %>%
head()
```
In this example, `classify` and `compare_features` come from `theftdlc`.
Please see the vignette for more information and the full functionality of both packages.
## Citation
If you use `theft` or `theftdlc` in your own work, please cite both the paper:
T. Henderson and Ben D. Fulcher. [Feature-Based Time-Series Analysis in R using the theft Package](https://arxiv.org/abs/2208.06146). arXiv, (2022).
and the software:
```{r, echo = FALSE}
citation("theft")
citation("theftdlc")
```