I’m running experiments on an environment with no internet connection, and I have some issues running wand locally. I can save the run in offline mode.
I’d like to access the logs/artifacts/tables using the public API (I have that already implemented), but by giving a path to the relevant experiment’s wandb directory (and not with the
entity/project/run_id run_id format)?
If you would like to use the API to fetch data, unfortunately the only way right now is to log the files in the directory to a W&B instance (local or our cloud). Seeing as you do not have access to the internet, the only solution in this case would be to set up a local instance.
Could you share what issues you are seeing with trying to set up your local instance? We can help get it going for you.
We wanted to follow up with you regarding your support request as we have not heard back from you. Please let us know if we can be of further assistance or if your issue has been resolved.
Weights & Biases