https://github.com/hypnosapos/visdom-docker
Visdom Docker
https://github.com/hypnosapos/visdom-docker
machine-learning metrics plotly visdom visdom-docker visualisation
Last synced: 11 months ago
JSON representation
Visdom Docker
- Host: GitHub
- URL: https://github.com/hypnosapos/visdom-docker
- Owner: hypnosapos
- License: mit
- Created: 2018-04-03T13:11:48.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2021-04-19T14:45:24.000Z (about 5 years ago)
- Last Synced: 2024-07-31T20:50:11.310Z (almost 2 years ago)
- Topics: machine-learning, metrics, plotly, visdom, visdom-docker, visualisation
- Language: Python
- Homepage:
- Size: 24.4 KB
- Stars: 12
- Watchers: 3
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Visdom [](https://circleci.com/gh/hypnosapos/visdom-docker/tree/master)
[](https://hub.docker.com/r/hypnosapos/visdom)
If you want to run [visdom server](https://github.com/facebookresearch/visdom) by a docker container this is for you.
## Build your own image
Not all versions of original project are released, so if you want to build a particular version (I mean "commit-ish") of sources
then update the **commitish** file:
```bash
$ docker build -t hypnosapos/visdom:$(cat commitish) .
```
Default version of python is **3.9** (build arg PY_VERSION) and distribution is **slim** (build arg DIST),
thus add your custom values if your want other base docker image, here an example:
```bash
$ docker build --build-arg PY_VERSION=3.9 --build-arg DIST=slim -t hypnosapos/visdom:3.9-slim-$(cat commitish) .
```
## Play
Before running the container take a look at the following env variables ( `-e NAME=VALUE`):
| Name | Default |
|----------------------|-------------------|
| `PORT` | 8097 |
| `ENV_PATH` | $HOME/.visdom |
| `LOGGING_LEVEL` | INFO |
| `HOSTNAME` | localhost |
| `BASE_URL` | / |
| `READONLY` | True |
| `ENABLE_LOGIN` | False |
| `FORCE_NEW_COOKIE` | False |
Now we're ready to run the container
```bash
$ docker run -it -p 8097:8097 [-e NAME=VALUE] --name visdom hypnosapos/visdom:$(cat commitish)
```
In order to preserve data or sessions we recommend you attach a volume to persist them (obviously):
```bash
$ docker run -it -v : -p 8097:8097 -e ENV_PATH= --name visdom hypnosapos/visdom:$(cat commitish)
```
Using default values server would be ready at http://localhost:8097
## Trying visdom
To play with some examples once our docker container is running we recommend you try:
```bash
docker exec -it visdom bash -c "python /root/visdom/example/demo.py"
```
And then check out examples at: http://localhost:8097
I hope it helps :satisfied: !!