> For the complete documentation index, see [llms.txt](https://pytorchhair.gitbook.io/project/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://pytorchhair.gitbook.io/project/utils/visdom.md).

# Visdom

[Visdom](https://github.com/facebookresearch/visdom)은 데이터를 시각화하기 위한 툴입니다. torch, numpy를 지원하고 있습니다.

![from https://github.com/facebookresearch/visdom](https://camo.githubusercontent.com/d69475a01f9f327fc42931a21df8134d1fbdfc19/68747470733a2f2f6c68332e676f6f676c6575736572636f6e74656e742e636f6d2f2d62714839555843772d42452f574c3255736472726241492f41414141414141416e59632f656d727877436d6e7257345f434c54797955747442305359524a2d693443436951434c63422f73302f53637265656e2b53686f742b323031372d30332d30362b61742b31302e35312e30322b414d2e706e67253232766973646f6d5f626967253232)

#### 쉬운 시작

1.Visdom server 켜기

```
pip install visdom
python -m visdom.server
# localhost:8097에 접속합니다.
```

2\. Visdom에 로그 남기기

```python
DEFAULT_PORT = 8097
DEFAULT_HOSTNAME = "http://localhost"

vis = visdom.Visdom(port=DEFAULT_PORT, server=DEFAULT_HOSTNAME)
vis.text('텍스트 써보기') vis.images(img:numpy_type) # numpy type의 이미지를 변수에 할당 vis.matplot(plt) # matplotlib의 plot type의 변수에 할당
```

 더 자세한 사항은 [Visdom 문서](https://github.com/facebookresearch/visdom)를 참조해주세요