wandb launch-agent

GitHub source

Run a W&B launch agent.

Usage

wandb launch-agent [OPTIONS]

Options

queues

The name of a queue for the agent to watch. Multiple -q flags supported.

Flag	Default	Type
-q, --queue	None	STRING

entity

The entity to use. Defaults to current logged-in user

Flag	Default	Type
-e, --entity	None	STRING

log_file

Destination for internal agent logs. Use - for stdout. By default all agents logs will go to debug.log in your wandb/ subdirectory or WANDB_DIR if set.

Flag	Default	Type
-l, --log-file	None	STRING

max_jobs

The maximum number of launch jobs this agent can run in parallel. Defaults to 1. Set to -1 for no upper limit

Flag	Default	Type
-j, --max-jobs	None	STRING

config

path to the agent config yaml to use

Flag	Default	Type
-c, --config	None	STRING

url

a wandb client registration URL, this is generated in the UI

Flag	Default	Type
-u, --url	None	STRING

verbose

Display verbose output

Flag	Default	Type
-v, --verbose	0

help

Show this message and exit.

Flag	Default	Type
--help	False	BOOL

CLI Help

Usage: launch-agent [OPTIONS]

  Run a W&B launch agent.

Options:
  -q, --queue TEXT     The name of a queue for the agent to watch. Multiple -q
                       flags supported.
  -e, --entity TEXT    The entity to use. Defaults to current logged-in user
  -l, --log-file TEXT  Destination for internal agent logs. Use - for stdout.
                       By default all agents logs will go to debug.log in your
                       wandb/ subdirectory or WANDB_DIR if set.
  -j, --max-jobs TEXT  The maximum number of launch jobs this agent can run in
                       parallel. Defaults to 1. Set to -1 for no upper limit
  -c, --config TEXT    path to the agent config yaml to use
  -v, --verbose        Display verbose output
  --help               Show this message and exit.