Skip to content
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

Hydra yaml config #81

Merged
merged 30 commits into from
Jun 28, 2024
Merged

Hydra yaml config #81

merged 30 commits into from
Jun 28, 2024

Conversation

kks32
Copy link
Contributor

@kks32 kks32 commented Jun 25, 2024

Use yaml to configure settings rather than command line arguments:

  • Define input file schema and validation
  • Support resume flag
  • Update README
  • move constants to data section

Related Issues/PRs
#80

@kks32 kks32 added the Priority: Medium Priority: Medium label Jun 25, 2024
@kks32 kks32 added this to the GNS v2.0 milestone Jun 25, 2024
@kks32 kks32 requested a review from yjchoi1 June 25, 2024 14:32
@kks32 kks32 self-assigned this Jun 25, 2024
@kks32 kks32 marked this pull request as ready for review June 25, 2024 23:20
gns/train.py Outdated Show resolved Hide resolved
@yjchoi1
Copy link
Collaborator

yjchoi1 commented Jun 28, 2024

Would it be better to add the configurations from metadata.json to config.yml? However, this might cause compatibility issues with the previous version, which explicitly reads from metadata.json.

Currently, metadata.json controls some of the important hyperparameters like n_message_passing, connectivity_raidus, nnode_in (optional) etc.

Other than these, everything looks good. I tested PR on my actual data for training, rollout, and rendering.

@kks32
Copy link
Contributor Author

kks32 commented Jun 28, 2024

Currently, metadata.json controls some of the important hyperparameters like n_message_passing, connectivity_raidus, nnode_in (optional) etc.

MLP configuration will be part of Configuration file in future PRs. This is planned for v2 to be able to switch between different types of MLPs/KANs

@kks32 kks32 merged commit a28603e into v2 Jun 28, 2024
1 check passed
@kks32 kks32 deleted the config branch June 28, 2024 12:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Priority: Medium Priority: Medium
Projects
Development

Successfully merging this pull request may close these issues.

2 participants