wandb.agent

Search…
wandb.agent
​​
View source on GitHub​
Generic agent entrypoint, used for CLI or jupyter.
agent(
    sweep_id, function=None, entity=None, project=None, count=None
)
Will run a function or program with configuration parameters specified by server.
Arguments
Text
sweep_id
(dict) Sweep ID generated by CLI or sweep API
function
(func, optional) A function to call instead of the "program" specifed in the config.
entity
(str, optional) W&B Entity
project
(str, optional) W&B Project
count
(int, optional) the number of trials to run.
Examples:
Run a sample sweep over a function:
import wandb
sweep_configuration = {
    "name": "my-awesome-sweep",
    "metric": {"name": "accuracy", "goal": "maximize"},
    "method": "grid",
    "parameters": {
        "a": {
            "values": [1, 2, 3, 4]
        }
    }
}
​
def my_train_func():
    # read the current value of parameter "a" from wandb.config
    wandb.init()
    a = wandb.config.a
​
    wandb.log({"a": a, "accuracy": a + 1})
​
sweep_id = wandb.sweep(sweep_configuration)
​
# run the sweep
wandb.agent(sweep_id, function=my_train_func)
wandb.Run
wandb.controller
Last modified 1mo ago
Copy link