0

I'm checking the KubeFlow guide here and it mentions using some pre-made images. I'm using the JupyterLab + PyTorch (CPU) + common packages one. I'm running the Docker image locally, and in the logs it says it's running JupyterLab:

[I 2022-10-11 11:24:07.526 ServerApp] Jupyter Server 1.8.0 is running at:
[I 2022-10-11 11:24:07.526 ServerApp] http://575f25bebe96:8888/lab
[I 2022-10-11 11:24:07.526 ServerApp]     http://127.0.0.1:8888/lab

However, I have no idea how to connect to it. Opening http://127.0.0.1:8888/lab doesn't work. Some suggest I should get the Docker container's IP address and connect to it directly, but doing:

C:\Users\me>docker inspect --format "{{ .NetworkSettings.IPAddress }}" ml_kubeflow
172.17.0.2

and trying to open 172.17.0.2:8888/lab still won't work.

How can I connect to this container's running JupyterLab?

lte__
  • 7,175
  • 25
  • 74
  • 131

0 Answers0