How do I launch multiple runs from one script? - Weights & Biases Documentation

To log multiple runs within a single script, you must finish each previous run before you start a new one. The following sections describe two patterns for sequencing runs in one script and a separate pattern for keeping multiple runs active at the same time.

Use a context manager

Use wandb.init() as a context manager. This pattern finishes the run and marks it as failed if your script raises an exception:

import wandb

for x in range(10):
    with wandb.init() as run:
        for y in range(100):
            run.log({"metric": x + y})

Call run.finish explicitly

You can also call run.finish() explicitly:

import wandb

for x in range(10):
    run = wandb.init()

    try:
        for y in range(100):
            run.log({"metric": x + y})

    except Exception:
        run.finish(exit_code=1)
        raise

    finally:
        run.finish()

Multiple active runs

In wandb 0.19.10 and later, if you need more than one run active at the same time rather than running them sequentially, set the reinit setting to "create_new" to create multiple simultaneously active runs.

import wandb

with wandb.init(reinit="create_new") as tracking_run:
    for x in range(10):
        with wandb.init(reinit="create_new") as run:
            for y in range(100):
                run.log({"x_plus_y": x + y})

            tracking_run.log({"x": x})

For more information about reinit="create_new", including caveats about W&B integrations, see Multiple runs per process.

Experiments

⌘I

​Use a context manager

​Call run.finish explicitly

​Multiple active runs

Use a context manager

Call run.finish explicitly

Multiple active runs