|
1 | 1 | # Machine Learning in R <img src="man/figures/logo_navbar.png" align="right" />
|
2 | 2 |
|
3 | 3 | [](https://travis-ci.org/mlr-org/mlr)
|
4 |
| -[](https://cran.r-project.org/package=mlr) |
| 4 | +[](https://cran.r-project.org/package=mlr)j |
5 | 5 | [](https://cran.r-project.org/web/checks/check_results_mlr.html)
|
6 | 6 | [](https://cran.rstudio.com/web/packages/mlr/index.html)
|
7 | 7 | [](https://stackoverflow.com/questions/tagged/mlr)
|
8 | 8 | [](https://www.tidyverse.org/lifecycle/#stable)
|
9 | 9 | [](https://cran.r-project.org/package=mlr)
|
10 | 10 |
|
11 | 11 | * [CRAN release site](https://CRAN.R-project.org/package=mlr)
|
12 |
| -* [Online Tutorial](https://mlr.mlr-org.com/index.html) |
13 |
| -* [mlr cheatsheet](https://github.com/mlr-org/mlr/blob/master/addon/cheatsheet/MlrCheatsheet.pdf) |
| 12 | +* [Online tutorial](https://mlr.mlr-org.com/index.html) |
| 13 | +* [Cheatsheet](https://github.com/mlr-org/mlr/blob/master/addon/cheatsheet/MlrCheatsheet.pdf) |
14 | 14 | * [Changelog](https://mlr.mlr-org.com/news/index.html)
|
15 | 15 |
|
16 |
| -### Install the development version |
| 16 | +**We are actively working on [mlr3](https://github.com/mlr-org/mlr3) as a successor of *mlr*. This implies that we have less time to reply to *mlr* issues.** |
17 | 17 |
|
18 |
| -```R |
19 |
| -remotes::install_github("mlr-org/mlr") |
| 18 | +* Stackoverflow: [`mlr`](https://stackoverflow.com/questions/tagged/mlr) |
| 19 | +* [Slack](https://mlr-org.slack.com/) |
| 20 | +* [We have a blog on mlr](https://mlr-org.com/). |
| 21 | + |
| 22 | +### Installation |
| 23 | + |
| 24 | +**Release** |
| 25 | + |
| 26 | +```r |
| 27 | +install.packages("mlr") |
20 | 28 | ```
|
21 | 29 |
|
22 |
| -* There is a `mlr` tag on [Stackoverflow](https://stackoverflow.com/questions/tagged/mlr). |
23 |
| -* [We are on Slack](https://mlr-org.slack.com/) (Request invitation: code{at}jakob-r.de). |
24 |
| -* [We have a blog on mlr](https://mlr-org.com/). |
25 |
| -* We are in the top 20 of the most starred R packages on Github, as reported by [metacran](https://www.r-pkg.org/starred). |
| 30 | +**Development** |
26 | 31 |
|
27 |
| -**We are actively working on [mlr3](https://github.com/mlr-org/mlr3) as a successor of *mlr*. This implies that we have less time to reply to *mlr* issues.** |
| 32 | +```R |
| 33 | +remotes::install_github("mlr-org/mlr") |
| 34 | +``` |
28 | 35 |
|
29 | 36 | # mlr - How to Cite and Citing Publications
|
30 | 37 |
|
@@ -52,38 +59,37 @@ Furthermore, the package is nicely connected to the [**OpenML**](https://github.
|
52 | 59 |
|
53 | 60 | # Features
|
54 | 61 |
|
55 |
| -* Clear S3 interface to R classification, regression, clustering and survival analysis methods |
| 62 | +* Clear **S3** interface to R **classification, regression, clustering and survival** analysis methods |
56 | 63 | * Possibility to fit, predict, evaluate and resample models
|
57 | 64 | * Easy extension mechanism through S3 inheritance
|
58 | 65 | * Abstract description of learners and tasks by properties
|
59 | 66 | * Parameter system for learners to encode data types and constraints
|
60 | 67 | * Many convenience methods and generic building blocks for your machine learning experiments
|
61 |
| -* Resampling methods like bootstrapping, cross-validation and subsampling |
| 68 | +* Resampling methods like **bootstrapping, cross-validation and subsampling** |
62 | 69 | * Extensive visualizations for e.g. ROC curves, predictions and partial predictions
|
63 | 70 | * Benchmarking of learners for multiple data sets
|
64 |
| -* Easy hyperparameter tuning using different optimization strategies, including potent configurators like iterated F-racing (irace) or sequential model-based optimization |
65 |
| -* Variable selection with filters and wrappers |
| 71 | +* Easy hyperparameter tuning using different optimization strategies, including potent configurators like **iterated F-racing (irace) or sequential model-based optimization** |
| 72 | +* **Variable selection with filters and wrappers** |
66 | 73 | * Nested resampling of models with tuning and feature selection
|
67 |
| -* Cost-sensitive learning, threshold tuning and imbalance correction |
| 74 | +* **Cost-sensitive learning, threshold tuning and imbalance correction** |
68 | 75 | * Wrapper mechanism to extend learner functionality in complex and custom ways
|
69 | 76 | * Combine different processing steps to a complex data mining chain that can be jointly optimized
|
70 |
| -* OpenML connector for the Open Machine Learning server |
71 |
| -* Extension points to integrate your own stuff |
72 |
| -* Parallelization is built-in |
| 77 | +* **OpenML** connector for the Open Machine Learning server |
| 78 | +* **Parallelization** is built-in |
73 | 79 | * Unit-testing
|
74 |
| -* Detailed tutorial |
| 80 | +* **Detailed tutorial** |
| 81 | + |
75 | 82 |
|
76 | 83 | # Get in Touch
|
77 | 84 |
|
78 |
| -Please use the issue tracker for problems, questions and feature requests. |
| 85 | +Please use the issue tracker for bugs, questions and feature requests. |
79 | 86 |
|
80 |
| -We also do not hate beginners and it is perfectly valid to mark an issue as "Question". |
81 |
| -However, simple usage questions are better suited at [Stackoverflow using the 'mlr' tag](https://stackoverflow.com/questions/tagged/mlr). |
| 87 | +Simple usage questions are better suited at [Stackoverflow using the 'mlr' tag](https://stackoverflow.com/questions/tagged/mlr). |
82 | 88 |
|
83 |
| -Please don't forget that all of us work in academia and put a lot of work into this project, simply because we like it, not because we are specifically paid for it. |
| 89 | +Please don't forget that all of us work in academia and put a lot of work into this project - simply because we like it, not because we are paid for it. |
84 | 90 |
|
85 |
| -We also welcome pull requests or new developers. |
86 |
| -Just make sure that you have a glance at our [**mlr** coding guidelines](https://github.com/mlr-org/mlr/wiki/Developer-Instructions#mlr-coding-guidelines) before. |
| 91 | +New development efforts should go into [`mlr3`](https://github.com/mlr-org/mlr3). |
| 92 | +We have a [developer guide](https://github.com/mlr-org/mlr/wiki/Developer-Instructions#mlr-coding-guidelines) and our own coding style which can easily applied by using the `mlr_style` from the [styler](https://github.com/r-lib/styler) package. |
87 | 93 |
|
88 | 94 | # mlr-tutorial
|
89 | 95 |
|
@@ -122,6 +128,3 @@ Important: Do not commit any file in `docs/` as the rendering will be done by Tr
|
122 | 128 |
|
123 | 129 | Edit `_pkgdown.yml` and add the new section at the appropriate place.
|
124 | 130 |
|
125 |
| -**Issues and Pull Requests** |
126 |
| - |
127 |
| -If you want to open an issue or pull request that is related to `mlr-tutorial`, label it with `tutorial` and mention [jakob-r](https://github.com/jakob-r) or [pat-s](https://github.com/pat-s) if you need help. |
|
0 commit comments