Nadeem-Shoukath115113's picture
download
raw
1.28 kB
import sys
import argparse
import hydra
from omegaconf import DictConfig
from hydra.utils import instantiate
# Parse --mode before Hydra sees sys.argv
parser = argparse.ArgumentParser()
parser.add_argument(
"--mode",
type=str,
default="eval",
choices=["eval", "raw_data", "label", "label_mp", "clear"],
)
args, remaining_argv = parser.parse_known_args()
# Replace sys.argv with only the remaining args for Hydra
sys.argv = [sys.argv[0]] + remaining_argv
@hydra.main(config_path="config", config_name=None, version_base="1.1")
def main(cfg: DictConfig):
workspace = instantiate(cfg)
if args.mode == "eval":
workspace.eval()
elif args.mode == "raw_data":
workspace.eval_raw_data()
elif args.mode == "label":
# Label a directory of raw episodes with per-frame reward.npy / progress.npy
workspace.label_reward()
elif args.mode == "label_mp":
# Same as label, but sharded across GPUs (multi-process)
workspace.label_reward_multi_process()
elif args.mode == "clear":
# Remove previously written reward.npy / progress.npy from the label dir
workspace.clear_labels()
else:
raise ValueError(f"Unsupported mode: {args.mode}")
if __name__ == "__main__":
main()

Xet Storage Details

Size:
1.28 kB
·
Xet hash:
e021f882f61f592a12e0e8e4799bbd810e8dfecb3b1c17103868131a7f9b67ed

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.