Skip to content
This repository was archived by the owner on May 1, 2019. It is now read-only.

Latest commit

 

History

History
21 lines (16 loc) · 908 Bytes

File metadata and controls

21 lines (16 loc) · 908 Bytes

The server is build completely inside a docker image.

IF you do not have the docker image yet:

# Get it from Docker Hub
docker pull alband/crayon

# OR build it locally with:
docker build -t alband/crayon:latest .

Then the server can be started with:

docker run -d -p 6006:6006 -p 6007:6007 --name crayon alband/crayon [frontend_refresh] [backend_refresh]

Where frontend_refresh is an optional argument that allows to change the refresh rate of tensorboard to the given value (in seconds). Default is 10 seconds. Where backend_refresh in an optional argument that allows to change how often the tensorflow backend reload the files from disk (in seconds). Default is 0.5 seconds. Reducing this value too much may make tensorboard unstable.

Tensorboard can then be accessed from a browser at localhost:6006. The client should be setup to send the datas at localhost:6007.