backfill_forks.py

import sys
import os

# Add BE to path
sys.path.append(os.path.join(os.path.dirname(__file__), 'BE'))

from database import SessionLocal
from models import TrainingRun, ModelFork

def backfill():
    db = SessionLocal()
    try:
        runs = db.query(TrainingRun).filter(TrainingRun.model_fork_id == None).all()
        print(f"Found {len(runs)} runs without a fork.")
        
        for run in runs:
            base_model_name = run.base_model_name or "Legacy / Unknown"
            fork_name = f"Default {base_model_name} Fork"
            
            # Find or create fork
            fork = db.query(ModelFork).filter_by(base_model_name=base_model_name, fork_name=fork_name).first()
            if not fork:
                fork = ModelFork(
                    base_model_name=base_model_name,
                    fork_name=fork_name,
                    description="Auto-generated default fork for legacy runs",
                    active_model_path=run.best_model_path # use the latest run's model path for the fork if newly created
                )
                db.add(fork)
                db.commit()
                db.refresh(fork)
                print(f"Created fork: {fork_name}")
            else:
                # Update active model path if this run is newer (assuming chronological processing)
                fork.active_model_path = run.best_model_path
                db.commit()
                
            run.model_fork_id = fork.id
            run.selected_fork_name = fork.fork_name
            db.commit()
            print(f"Mapped run {run.id} to fork {fork.id}")
            
    except Exception as e:
        print(f"Error: {e}")
    finally:
        db.close()

if __name__ == "__main__":
    backfill()