Closed slvcsl closed 1 year ago
Here's a programmatic way to get the best run from a sweep: https://docs.wandb.ai/guides/track/public-api-guide#get-the-best-run-from-a-sweep
import wandb
api = wandb.Api()
sweep = api.sweep("<entity>/<project>/<sweep_id>")
best_run = sweep.best_run()
Accessing best model and configuration
Hi! I am trying to use sweep starting from the quickstart and sweeps-python/sweep.py example.
How do I access the best model / best config in the code once the selection is finished? I can see the visualization, but what about the code?
Thanks for the incredible work!