Skip to main content
The ability to fork a run is in private preview. Contact W&B Support at support@wandb.com to request access to this feature.
You can explore different parameters or models from a specific point in an experiment without affecting the original run. To do this, fork an existing W&B run. When you fork a run, W&B creates a new run using the source run’s run ID and a specified step. The forked run includes all data from the source run up to that step. After the fork step, you can log new data to the forked run independently of the original run.
  • Forking a run requires wandb SDK version >= 0.16.5
  • Forking a run requires monotonically increasing steps. You cannot fork from a run that uses non-monotonic steps defined with define_metric(). Non-monotonic steps break the chronological order of run history and system metrics.
Use the fork_from parameter in wandb.init() to fork from an existing run. Specify the source run’s unique run ID and the step you want to start the forked run from.
See Unique run identifiers to learn more about run IDs and how to locate them.

​
Fork from a previously logged run

The following code snippet shows how to fork from a run that was previously logged to W&B. Use this approach to fork from a run in a different script or notebook, or from a run logged by a different user or service account. Replace <source-run-id>, <project>, and <entity> with your own values. 
import wandb

# The unique ID of the source run to fork from
source_run_id = "<source-run-id>"

# Specify the step to fork from
fork_step = 200

# Fork the run
with wandb.init(
    project="<project>",
    entity="<entity>",
    fork_from=f"{source_run_id}?_step={fork_step}",
) as forked_run:
    pass

​
Fork from a run in the same script

The following code snippet shows how to fork from a run within the same script. This might occur if you want explore different parameters or models from a specific point within the same script or notebook. 
import wandb

# Initialize a run
with wandb.init(
    project="<project>",
    entity="<entity>"
) as original_run:
    # ...training logic goes here ...
    pass

# Specify the step to fork from
fork_step = 200

# Use the original run's ID and specify the step to fork from
with wandb.init(
    project="<project>",
    entity="<entity>",
    fork_from=f"{original_run.id}?_step={fork_step}",
) as forked_run:
    # ...training logic goes here ...
    pass
Use the original_run.id property to obtain the unique run ID of the original run.

​
Example script

For example, the following code example shows how to first fork a run and then how to log metrics to the forked run starting from a training step of 200. Copy and paste the following code into a Python script or notebook cell. Replace <project> and <entity> with your own values. 
import wandb
import math

# Initialize the first run and log some metrics
with wandb.init(
    project="<project>",
    entity="<entity>"
) as run1:
    for i in range(300):
        run1.log({"metric": i})

# Fork from the first run at a specific step and log the
# metric starting from step 200
with wandb.init(
    project="<project>", 
    entity="<entity>", 
    fork_from=f"{run1.id}?_step=200"
) as run2:
    # Continue logging in the new run
    # For the first few steps, log the metric as is from run1
    # After step 250, start logging the spikey pattern
    for i in range(200, 300):
        if i < 250:
            # Continue logging from run1 without spikes
            metric_value = i
        else:
            # Introduce the spikey behavior starting from step 250
            metric_value = i + (2 * math.sin(i / 3.0))  # Apply a subtle spikey pattern

        # Log both metrics in a single call to ensure they're
        # logged at the same step
        run2.log({
            "metric": metric_value,
            "additional_metric": i * 1.1
        })
Rewind and forking compatibilityForking compliments a rewind by providing more flexibility in managing and experimenting with your runs.When you fork from a run, W&B creates a new branch off a run at a specific point to try different parameters or models.When you rewind a run, W&B let’s you correct or modify the run history itself.