CODEWITHSUNDEEP
dockertensorboard
AKE
AKE

Reputation: 111

Run tensorboard to visualise docker container running reinforcement learning in Ray Rllib

I have a docker container running a custom simulation for reinforcement learning. I have placed all of this code in a docker container.

This code takes a long time to run, and I would like to be able to visualise the algorithm in real time.

However, I am unsure on how to start tensorboard from this container. I was thinking of saving the data to a volume and then accessing that from another container.

Upvotes: 0

Views: 850

Answers (2)

TkrA
TkrA

Reputation: 666

The steps I followed and I could visualise the results with tensorboard:

  • when creating a the container, open/map an external port for tensorboard:

nvidia-docker run -d --name tkra_tensorb --ipc=host -it -p 8513:8090 -p 3014:6006 -v /data:/data tkra_tb

  • inside the container, run tensorboard:

tensorboard --logdir /data/tkra/MyDatasets/resnet101/checkpoints/ --host 0.0.0.0 --port 6006

Open tensorboard in my browser: <server_address>:3014

Upvotes: 0

AKE
AKE

Reputation: 111

I ended up doing as a I proposed, saving the data in a volume and accessing the external data source via tensorboard.

docker run -it -v <path/to/tensorflow_data>:</path/to/vol> -v <docker_container>:latest

Where code inside <> should be replaced with where the tensorflow data is stored, the path where you saved the volume inside the container and the name of your container.

Upvotes: 1

Related Questions