How to use the seml.queuing.queue_experiments function in seml

To help you get started, we’ve selected a few seml examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github TUM-DAML / seml / seml / View on Github external
hdlr = logging.StreamHandler(sys.stderr)

    if args.func == mongodb_credentials_prompt:  # launch SEML configure.
        del args.db_collection_name
    else:  # otherwise remove the flag as it is not used elsewhere.
        if not args.db_collection_name:
            parser.error("the following arguments are required: db_collection_name")
            if os.path.isfile(args.db_collection_name):
                logging.warning("Loading the collection name from a config file. This has been deprecated. "
                                "Please instead provide a database collection name in the command line.")
                seml_config, _, _ = read_config(args.db_collection_name)
                if args.func == queue_experiments:
                    args.config_file = args.db_collection_name
                args.db_collection_name = seml_config['db_collection']
            elif args.func == queue_experiments and not args.config_file:
                parser_queue.error("the following arguments are required: config_file")

    f = args.func
    del args.func
    del args.verbose
    if 'filter_states' in args:
        args.filter_states = [state.upper() for state in args.filter_states]
github TUM-DAML / seml / seml / View on Github external
help="Path to the YAML configuration file for the experiment.")
            '-nh', '--no-hash', action='store_true',
            help="Do not use the hash of the config dictionary to filter out duplicates (by comparing all"
                 "dictionary values individually). This is much  slower, so use only if you have a good reason not to"
                 " use the hash.")
            '-nc', '--no-config-check', action='store_true',
            help="Do not check the config for missing/unused arguments. "
                 "Use this if the check fails unexpectedly when using "
                 "advanced Sacred features or to accelerate queueing.")
            '-f', '--force-duplicates', action='store_true',
            help="Add experiments to the database even when experiments with identical configurations "
                 "are already in the database.")

    parser_start = subparsers.add_parser(
            help="Fetch queued experiments from the database and run them (by default via Slurm).")
            '-l', '--local', action='store_true',
            help="Run the experiments locally (not via Slurm).")
            '-n', '--num-exps', type=int, default=-1,
            help="Only start the specified number of experiments.")
            '-u', '--unobserved', action='store_true',
            help="Run the experiments without Sacred observers (no changes to the database). "
                 "This also disables output capturing by Sacred, facilitating the use of debuggers (pdb, ipdb).")
            '-pm', '--post-mortem', action='store_true',