Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/egenn/rtemis
Advanced Machine Learning and Visualization
https://github.com/egenn/rtemis
data-science data-visualization machine-learning machine-learning-library r rstats visualization
Last synced: 1 day ago
JSON representation
Advanced Machine Learning and Visualization
- Host: GitHub
- URL: https://github.com/egenn/rtemis
- Owner: egenn
- License: gpl-3.0
- Created: 2019-02-13T11:23:36.000Z (almost 6 years ago)
- Default Branch: main
- Last Pushed: 2024-10-16T02:15:07.000Z (2 months ago)
- Last Synced: 2024-10-17T12:32:49.593Z (2 months ago)
- Topics: data-science, data-visualization, machine-learning, machine-learning-library, r, rstats, visualization
- Language: R
- Homepage: https://rtemis.org
- Size: 7.92 MB
- Stars: 141
- Watchers: 9
- Forks: 19
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: NEWS.md
- License: LICENSE
Awesome Lists containing this project
README
[](https://github.com/egenn/rtemis/actions/workflows/R-CMD-check.yaml)
[](https://egenn.r-universe.dev/rtemis)
# **_rtemis_** Machine Learning and Visualization
A platform for advanced Machine Learning research and applications.
The goal of **rtemis** is to make data science efficient and accessible with no compromise on flexibility.
## Documentation
* [**Documentation and vignettes**](https://rtemis.org/rtemis)
## Requirements
R version 4.1 or higher
## Installation
You can install `rtemis` from `r-universe` or using `pak`, `remotes`, or `devtools`.
* `r-universe`:
```r
install.packages('rtemis', repos = c('https://egenn.r-universe.dev', 'https://cloud.r-project.org'))
```
* `pak`:
```r
pak::pkg_install("egenn/rtemis")
```
* `remotes`:
```r
remotes::install_github("egenn/rtemis")
```
* `devtools`:
```r
devtools::install_github("egenn/rtemis")
```
### Note about Fortran support in MacOS
To allow compilation from source of any dependencies that require Fortran, you
will need to install the GNU Fortran compiler. The easiest way to do this is
with [Homebrew](https://brew.sh/):
```bash
brew install gcc
```
Then, you will need to add the following to your `~/.R/Makevars` file:
```bash
FC = usr/local/opt/gcc/bin/gfortran
F77 = /usr/local/opt/gcc/bin/gfortran
FLIBS = -L/usr/local/opt/gcc/lib
```
### Note about using `d_UMAP()`
`d_UMAP()` requires the `uwot` package, which currently requires that the `Matrix` and
`irlba` dependencies be installed from source. See more in the `uwot` issue
[here](https://github.com/jlmelville/uwot/issues/115).
### More setup info
See [here](https://rtemis.org/rtemis/Setup.html) for more setup and
installation instructions.
**Note:** Make sure to keep your installation updated by running
`remotes::install_github("egenn/rtemis")` regularly: it will only proceed if
there are updates available.
## 30-second intro to **rtemis**
Install dependencies if they are not already installed:
```r
packages <- c("future.apply", "ranger")
.add <- !packages %in% installed.packages()
install.packages(packages[.add])
```
Get cross-validated random forest performance on the iris dataset:
```r
library(rtemis)
mod <- train_cv(iris)
```
## What's new
We are working towards the 1.0 release, which will feature updates to the
API as well as the backend, and preparing for CRAN submission.
This will be accompanied by expansion of the [documentation](https://rtemis.org/rtemis).
For all updates, please see the [NEWS](NEWS.md) file.
The Python and Julia ports, `rtemispy` and `Rtemis.jl`, which are not yet
publicly available, are in parallel development. With the upcoming 1.0 release
of rtemis, the aim is to offer a unified API across all three languages.
## Features
* **Visualization**
* Static: **_mplot3_** family (base graphics)
* Dynamic: **_dplot3_** family ([plotly](https://plotly.com/r/))
* **Unsupervised Learning**
* Clustering: **_c_\*_**
* Decomposition: **_d_\*_**
* **Supervised Learning**
* Classification, Regression, Survival Analysis: **_s_\*_**
* **Cross-Decomposition**
* Sparse Canonical Correlation / Sparse Decomposition: **_x_\*_**
* **Meta-Models**
* Model Stacking: **_metaMod()_**
* Modality Stacking: **_metaFeat()_**
* Group-weighted Stacking: **_metaGroup()_**
(metaFeat and metaGroup have been removed for updating)
---
---
## rtemislive
**rtemislive** is rtemis' web interface / GUI.
It makes advanced visualization and modeling instantly accessible by all.
It is currently available for beta testing at UCSF,
and will be made publicly available once funding is secured for a hosting server.
## VS Code theme
Get the [rtemis-dark VS Code theme](https://marketplace.visualstudio.com/items?itemName=egenn.rtemis-dark).
Recommended font is Fira Code with its pretty ligatures.
## Python & Julia APIs
Python and Julia APIs are in development. The goal is to delliver a unified API across
all three languages by the time of the 1.0 release.