scripts/dataset.py

# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
#
# NVIDIA CORPORATION and its licensors retain all intellectual property
# and proprietary rights in and to this software, related documentation
# and any modifications thereto. Any use, reproduction, disclosure or
# distribution of this software and related documentation without an express
# license agreement from NVIDIA CORPORATION is strictly prohibited.
#
'''
Dataset class for loading and processing grasp data from a WebDataset.

Installation:
    pip install trimesh==4.5.3 objaverse==0.1.7 meshcat==0.0.12 webdataset==0.2.111

'''
import os
import json
import time
import webdataset as wds
import glob
from pathlib import Path
from tqdm import tqdm
from typing import Dict, Optional

def load_uuid_list(uuid_list_path: str):
    if not os.path.exists(uuid_list_path):
        raise FileNotFoundError(f"UUID list file not found: {uuid_list_path}")
    if uuid_list_path.endswith(".json"):
        with open(uuid_list_path, 'r') as f:
            uuids = json.load(f)
        if type(uuids) == list:
            return uuids
        elif type(uuids) == dict:
            return list(uuids.keys())
        else:
            raise ValueError(f"UUID list is not a list or dict: {uuids}")
    elif uuid_list_path.endswith(".txt"):
        with open(uuid_list_path, 'r') as f:
            uuids = [line.strip() for line in f.readlines()]
    else:
        raise ValueError(f"Unsupported file format: {uuid_list_path}")
    return uuids

class GraspWebDatasetReader:
    """Class to efficiently read grasps data using a pre-loaded index."""
    
    def __init__(self, dataset_path: str):
        """
        Initialize the reader with dataset path and load the index.
        
        Args:
            dataset_path (str): Path to directory containing WebDataset shards
        """
        self.dataset_path = dataset_path
        self.shards_dir = self.dataset_path
        
        # Load the UUID index
        index_path = os.path.join(self.shards_dir, "uuid_index.json")
        with open(index_path, 'r') as f:
            self.uuid_index = json.load(f)
        
        # Cache for open datasets
        self.shard_datasets = {}

    def read_grasps_by_uuid(self, object_uuid: str) -> Optional[Dict]:
        """
        Read grasps data for a specific object UUID using the index.
        
        Args:
            object_uuid (str): UUID of the object to retrieve
        
        Returns:
            Optional[Dict]: Dictionary containing the grasps data if found, None otherwise
        """
        if object_uuid not in self.uuid_index:
            return None
            
        shard_idx = self.uuid_index[object_uuid]
        
        # Get or create dataset for this shard
        if shard_idx not in self.shard_datasets:
            shard_path = f"{self.shards_dir}/shard_{shard_idx:03d}.tar"
            self.shard_datasets[shard_idx] = wds.WebDataset(shard_path)
        
        dataset = self.shard_datasets[shard_idx]
        
        # Search for the UUID in the specific shard
        for sample in dataset:
            if sample["__key__"] == object_uuid:
                return json.loads(sample["grasps.json"])
        
        return None