Skip to main content
After a run finishes, use the Public API guide to edit config, display name, tags, and notes without re-running the experiment. For tag concepts and UI workflows, see Tags. For config during a run, see Configuration. The examples below use run.update() to persist pending changes in one API call. 
import wandb

api = wandb.Api()
run = api.run("my-entity/my-project/run_id_here")

run.config["post_hoc_label"] = "baseline-v2"
run.name = "resnet50-lr1e-3-aug-v2"
run.tags.append("reviewed")
run.notes = "Final baseline. Val accuracy: 92.4%."
run.update()

​
Bulk updates

To tag or annotate many runs (for example, all runs from a sweep), iterate with filters and call update() per run: 
for run in api.runs("my-entity/my-project", filters={"sweep": "sweep_id_here"}):
    run.tags.append("sweep-reviewed")
    run.update()
Each update() is a separate API call. For large sets, add a short sleep between calls to avoid rate limits.
Runs Experiments API