|
| 1 | +--- |
| 2 | +title: Adding a new tool |
| 3 | +--- |
| 4 | +This wiki article explains how to add a new tool to the AIDOS toolbox (see [Research](/research).) |
| 5 | + |
| 6 | +### Step 1: Upload an image. |
| 7 | + |
| 8 | +#### Important! Remember to make a branch and open a pull request! All the following changes should be made on a separate branch. |
| 9 | + |
| 10 | +Choose an image that represents your tool. It is best if this image is roughly square-shaped. |
| 11 | + |
| 12 | +Save this image as `your-tool-name.svg` (.png or .jpg is also fine). Upload it to the `/static/tools/` folder. |
| 13 | + |
| 14 | +### Step 2: Create a .toml file |
| 15 | + |
| 16 | +In the `/data/tools` folder, add a new file named `your-tool-name.toml`. |
| 17 | + |
| 18 | +You can use the following as a template: |
| 19 | + |
| 20 | +```html |
| 21 | +name = "SCOTT" |
| 22 | +paper = "Curvature Filtrations for Graph Generative Model Evaluation" |
| 23 | +image = "scott.jpg" |
| 24 | +description = """\ |
| 25 | + SCOTT is a Python package for computing **curvature |
| 26 | + filtrations** for graphs and graph distributions. |
| 27 | + Our method introduces a novel way to compare graph |
| 28 | + distributions by combining discrete curvature on |
| 29 | + graphs with persistent homology, providing |
| 30 | + descriptors of graph sets that are: *robust*, |
| 31 | + *stable*, *expressive*, and *compatible with |
| 32 | + statistical testing*. |
| 33 | + """ |
| 34 | +repo = "https://github.com/aidos-lab/curvature-filtrations" |
| 35 | +pub = "https://doi.org/10.48550/arXiv.2301.12906" |
| 36 | +pip = "pip install curvature-filtrations" |
| 37 | +``` |
| 38 | + |
| 39 | +#### Data dictionary |
| 40 | +Bolded fields appear on the tool cards and thus are mandatory. |
| 41 | +- **Name**: he name of your tool (a clever acronym perhaps?) |
| 42 | +- Paper: If applicable, the name of the paper associated with your tool. |
| 43 | +- **Image**: The name of the image file from Step 1. Note that the field automatically points to `/static/tools`. |
| 44 | +- **Description**: A short description, often similar to what is in a README. Markdown format. |
| 45 | +- **Repo**: A link to the GitHub repo. The GitHub logo on the tool card will link to this url. |
| 46 | +- **Pub**: A link to a publication, blogpost, or other online presence associated with your tool. The document icon on the tool card will link to this url. |
| 47 | +- Pip: If applicable, an installation command (e.g. `pip install your-tool-name`). |
| 48 | + |
| 49 | +### Step 3: Add your tool to tools.md |
| 50 | + |
| 51 | +In the `/content/tools.md` file, add: |
| 52 | + |
| 53 | +```html |
| 54 | +< tool "your-tool-name" > within a pair of double curly braces {{}} |
| 55 | +``` |
| 56 | + |
| 57 | + to the **top** of the list of tools. We want to preserve ordering based on most to least recently developed. |
| 58 | + |
| 59 | + Sanity check: Make sure "your-tool-name" is the same as the name of your `.toml` file. |
| 60 | + |
| 61 | +### Step 3: Final Check & PR Submission |
| 62 | + |
| 63 | +After a quick check to make sure the toolbox is rendering properly with your new addition, you are ready to submit the pull request for Bastian to review. |
| 64 | + |
| 65 | +Nice work! |
0 commit comments