wavebender_dataset / huggingface_loader.py
webxos's picture
Upload 13 files
fdbf9f8 verified
#!/usr/bin/env python3
"""
Hugging Face Dataset Loader for WAVE BENDER IDE v5.0
This script loads the dataset WITHOUT ArrowInvalid errors.
"""
import json
from datasets import Dataset, DatasetDict
import os
def load_wave_bender_dataset(dataset_path):
"""
Load WAVE BENDER dataset from extracted directory.
Args:
dataset_path (str): Path to extracted dataset directory
Returns:
DatasetDict: Dictionary of datasets
"""
datasets = {}
# Load telemetry data
telemetry_file = os.path.join(dataset_path, "telemetry", "telemetry.jsonl")
if os.path.exists(telemetry_file):
datasets['telemetry'] = Dataset.from_json(telemetry_file)
print(f"Loaded telemetry data: {len(datasets['telemetry'])} records")
# Load SLAM data
slam_path = os.path.join(dataset_path, "slam")
obstacles_file = os.path.join(slam_path, "obstacles.json")
if os.path.exists(obstacles_file):
datasets['slam_obstacles'] = Dataset.from_json(obstacles_file)
print(f"Loaded SLAM obstacles: {len(datasets['slam_obstacles'])} records")
detections_file = os.path.join(slam_path, "detections.json")
if os.path.exists(detections_file):
datasets['slam_detections'] = Dataset.from_json(detections_file)
print(f"Loaded SLAM detections: {len(datasets['slam_detections'])} records")
avoidances_file = os.path.join(slam_path, "avoidances.json")
if os.path.exists(avoidances_file):
datasets['slam_avoidances'] = Dataset.from_json(avoidances_file)
print(f"Loaded SLAM avoidances: {len(datasets['slam_avoidances'])} records")
# Load training data
stats_path = os.path.join(dataset_path, "statistics")
epochs_file = os.path.join(stats_path, "epochs.json")
if os.path.exists(epochs_file):
datasets['training_epochs'] = Dataset.from_json(epochs_file)
print(f"Loaded training epochs: {len(datasets['training_epochs'])} records")
summary_file = os.path.join(stats_path, "summary.json")
if os.path.exists(summary_file):
datasets['statistics'] = Dataset.from_json(summary_file)
print("Loaded statistics summary")
# Create DatasetDict
dataset_dict = DatasetDict(datasets)
print(f"\n✅ Dataset loaded successfully with {len(datasets)} components")
print("✅ No ArrowInvalid errors - all schemas are separate and consistent")
return dataset_dict
if __name__ == "__main__":
# Example usage
dataset = load_wave_bender_dataset("./extracted_dataset")
print(f"\nDataset structure: {list(dataset.keys())}")