def test_meta_train_splitting_v2_multi_gpu_recurrent() -> None:
"""
Runs meta training with meta splitting networks v2 and compares reward curves
against saved baseline for a MetaWorld ML10 benchmark, running multiple processes
and on the GPU with a recurrent policy.
"""
# Load default training config.
with open(META_SPLITTING_V2_CONFIG_PATH, "r") as config_file:
config = json.load(config_file)
# Modify default training config.
config["meta_train_config"]["num_updates"] //= MP_FACTOR
config["meta_train_config"]["num_processes"] *= MP_FACTOR
config["meta_test_config"]["num_updates"] //= MP_FACTOR
config["meta_test_config"]["num_processes"] *= MP_FACTOR
config["cuda"] = True
config["meta_train_config"]["architecture_config"]["recurrent"] = True
config["meta_train_config"]["architecture_config"][
"recurrent_hidden_size"] = 64
config["meta_train_config"][
"baseline_metrics_filename"] = "meta_splitting_v2_multi_gpu_recurrent"
# Run training.
meta_train(config)
def test_meta_train_splitting_v2() -> None:
"""
Runs meta training with meta splitting networks v2 and compares reward curves
against saved baseline for a MetaWorld ML10 benchmark.
"""
# Load default training config.
with open(META_SPLITTING_V2_CONFIG_PATH, "r") as config_file:
config = json.load(config_file)
# Modify default training config.
config["meta_train_config"][
"baseline_metrics_filename"] = "meta_splitting_v2"
# Run training.
meta_train(config)
def test_meta_train_splitting_v1_recurrent() -> None:
"""
Runs meta training with meta splitting networks v1 and compares reward curves
against saved baseline for a MetaWorld ML10 benchmark, with a recurrent policy.
"""
# Load default training config.
with open(META_SPLITTING_V1_CONFIG_PATH, "r") as config_file:
config = json.load(config_file)
# Modify default training config.
config["meta_train_config"]["architecture_config"]["recurrent"] = True
config["meta_train_config"]["architecture_config"][
"recurrent_hidden_size"] = 64
config["meta_train_config"][
"baseline_metrics_filename"] = "meta_splitting_v1_recurrent"
# Run training.
meta_train(config)
def test_meta_train_splitting_v1_multi() -> None:
"""
Runs meta training with meta splitting networks v1 and compares reward curves
against saved baseline for a MetaWorld ML10 benchmark, running multiple processes.
"""
# Load default training config.
with open(META_SPLITTING_V1_CONFIG_PATH, "r") as config_file:
config = json.load(config_file)
# Modify default training config.
config["meta_train_config"]["num_updates"] //= MP_FACTOR
config["meta_train_config"]["num_processes"] *= MP_FACTOR
config["meta_test_config"]["num_updates"] //= MP_FACTOR
config["meta_test_config"]["num_processes"] *= MP_FACTOR
config["meta_train_config"][
"baseline_metrics_filename"] = "meta_splitting_v1_multi"
# Run training.
meta_train(config)
if __name__ == "__main__":
# Parse config filename from command line arguments.
parser = argparse.ArgumentParser()
parser.add_argument(
"command",
type=str,
help="Command to run. Either 'train' or 'tune'.",
)
parser.add_argument(
"config_filename",
type=str,
help="Name of config file to load from.",
)
args = parser.parse_args()
# Load config file.
with open(args.config_filename, "r") as config_file:
config = json.load(config_file)
# Run specified command.
if args.command == "train":
train(config)
elif args.command == "tune":
tune(config)
elif args.command == "meta_train":
meta_train(config)
else:
raise ValueError("Unsupported command: '%s'" % args.command)
