Skip to content

Testing different configuration for running ERA5 experiments (Draft PR not to be merged) #387

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: develop
Choose a base branch
from

Conversation

ankitpatnala
Copy link
Contributor

Description

This is a dummy PR to keep track of the workable configuration for training on ERA5.

  • config/default_config.yml
  • forecasting_finetuning_config.yml ( overwrite over default config for finetuning forecast experiments)
  • config/streams/streams_anemoi/era5.yml

Issue Number #382

The motivation behind this PR to keep track on the config that is working for all the collaborators across different HPCs

@FussyDuck
Copy link

CLA assistant check
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
You have signed the CLA already but the status is still pending? Let us recheck it.

@ankitpatnala ankitpatnala changed the title added default forecasting config for finetuning Testing different configuration for running ERA5 experiments (Draft PR not to be merged) Jun 25, 2025
@ankitpatnala ankitpatnala marked this pull request as draft July 1, 2025 13:15
@tjhunter
Copy link
Collaborator

tjhunter commented Jul 3, 2025

@ankitpatnala do we still need this PR?

@ankitpatnala
Copy link
Contributor Author

Somehow this was meant for config. Would be nice if it is left open and latest config is documented here. I do not know if the config in the develop is the desired config or still in testing phase.
@clessig ??

@clessig
Copy link
Collaborator

clessig commented Jul 3, 2025

@sophie-xhonneux @Jubeku : do we use this config? If yes, we should keep it open. Otherwise, how to we track the default config to use?

@sophie-xhonneux
Copy link
Contributor

I haven't used it as I primarily focused on stability and so simply did masking runs based on the failing config from @Jubeku

@Jubeku
Copy link
Contributor

Jubeku commented Jul 3, 2025

Somehow this was meant for config. Would be nice if it is left open and latest config is documented here. I do not know if the config in the develop is the desired config or still in testing phase. @clessig ??

I have never used it. I think it's from @ankitpatnala.

@clessig
Copy link
Collaborator

clessig commented Jul 3, 2025

We need to converge on one config for the scientific experiments. I think a draft PR is a good way to track it but we can use something else if you prefer. But we do need to converge on a common config to have comparable results.

@sophie-xhonneux
Copy link
Contributor

Sure, we can put it here. I don't mind, but it should be the full default_config and streams config

Can you put it here?

@sophie-xhonneux
Copy link
Contributor

Then I will send the large scale run!

@sophie-xhonneux
Copy link
Contributor

@clessig can you update this PR or confirm that I can use this config to test performance changes for the new prediction heads?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
Status: No status
Development

Successfully merging this pull request may close these issues.

6 participants