Feat/early stopping feature implementation

[Madhavi1108]

Fixes

Closes: #39

Type of Change

• Bug fix
• New feature
• Documentation / Refactor
• Math / Logic correction

Description

This PR adds native early stopping support to the Python API Model.train() method.

The following optional parameters are introduced:

• early_stopping
• patience
• restore_best

The training loop now tracks loss per epoch, maintains a best-loss checkpoint, and stops training when the loss does not improve for a configurable number of consecutive epochs.
When enabled, the model weights from the best-performing epoch are restored after training completes.

The default behavior remains unchanged when early_stopping=False, ensuring full backward compatibility.

How Has This Been Tested?

• Unit Tests: Not added yet
• Manual Testing: Verified behavior using a small dummy training loop and checked that training stops after the specified patience.
• Integration: Not applicable for this change

Screenshots / Logs

Contribution Context

• I am contributing through the SWOC program.

[github-actions]

Thank you for opening this PR! Our automated system is currently verifying the PR requirements.
Internal Discussion: Discord

[github-actions]

Validation Successful!

This pull request has been verified and linked to issue #39. The system is now synchronizing metadata from the referenced issue. Kindly await maintainer review of your changes.

[Aamod007]

Hi @Madhavi1108! Good work on the Rust implementation. Here are the issues:

What's Wrong

Critical:

1. Python API (etna/api.py) doesn't expose the new early_stopping, patience, restore_best parameters
2. These parameters aren't passed to the Rust call
3. No tests for the new feature
4. No documentation/docstring updates

Minor:
5. Missing newline at end of optimizer.rs
6. Silent failure if model restoration fails

What to Fix

• Add the 3 parameters to Python train() method
• Pass them through to Rust
• Write tests
• Update docs

The Rust code is solid, just needs Python integration!

[github-actions]

Validation Successful!

This pull request has been verified and linked to issue #39. The system is now synchronizing metadata from the referenced issue. Kindly await maintainer review of your changes.

[Madhavi1108]

Hi @Aamod007 ,
Thanks for suggesting the changes
I have implemented the changes based on my understanding
Any changes required do let me know
I will make them and update the docs accordingly

[Aamod007]

Everything is good, Excellent job !

[github-actions]

This PR has been successfully merged. We appreciate your effort in improving the library and look forward to your future contributions to the Etsi AI ecosystem.