SwanLab

Setting Experiment Tags ​

Experiment tags allow you to quickly label the methods, datasets, models, hyperparameters, Git repositories, etc. used in the experiment, and can later be used for grouping and filtering experiments.

Configured tags will appear below the experiment name:

Regular Tags ​

Method 1: Programmatic Setup

Use the tags parameter in swanlab.init to set experiment tags.

python
swanlab.init(
    tags=["tag1", "tag2"],
)

Method 2: GUI Setup

In the web interface, navigate to the top section of the experiment and click the "Add Tag" button to edit tags:

Git Tags ​

Supports recognizing GitHub and Gitee repository links in tags, displaying them in a special format with clickable redirects.

python
swanlab.init(
    tags=[
        "https://github.com/SwanHubX/SwanLab",
        "https://gitee.com/SwanHubX/SwanLab",
    ],
)

Arxiv Tags ​

Supports recognizing Arxiv links in tags, displaying them in a special format with clickable redirects.

python
swanlab.init(
    tags=["https://arxiv.org/abs/1706.03762"],
)

AI Community Tags ​

Supports recognizing links from AI open-source communities (HuggingFace, ModelScope, Modelers), displaying them in a special format with clickable redirects.

python
swanlab.init(
    tags=[
        "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528",
        "https://modelscope.cn/models/deepseek-ai/DeepSeek-R1-0528",
        "https://modelers.cn/models/Modelers_Park/DeepSeek-R1-0528-Qwen3-8B",
    ],
)

Hardware Tags ​

Supports recognizing hardware accelerator brands (nvidia, ascend, apple) in tags.

python
swanlab.init(
    tags=["nvidia", "ascend", "apple"],
)