Skip to main content
To add metrics to a run that has already finished, W&B offers several approaches depending on the complexity of your workflow. The following sections describe two patterns to continue logging to an existing experiment.

​
Group multiple runs for complex workflows

For complex workflows, use multiple runs and set the group parameters in wandb.init() to a unique value for all processes within a single experiment. The Runs tab groups the table by group ID so visualizations work as expected. This approach enables concurrent experiments and training runs while logging results in one location.

​
Resume a run for simpler workflows

For simpler workflows, call wandb.init() with resume=True and id=[UNIQUE-ID], then call wandb.init() again with the same id=[UNIQUE-ID]. Log as usual with run.log() or run.summary(), and the run values update.
Runs Metrics