Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception: SplitsNotFoundError
Message: The split names could not be parsed from the dataset config.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
for split_generator in builder._split_generators(
~~~~~~~~~~~~~~~~~~~~~~~~~^
StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/folder_based_builder/folder_based_builder.py", line 246, in _split_generators
raise ValueError(
"`file_name`, `*_file_name`, `file_names` or `*_file_names` must be present as dictionary key in metadata files"
)
ValueError: `file_name`, `*_file_name`, `file_names` or `*_file_names` must be present as dictionary key in metadata files
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 68, in compute_split_names_from_streaming_response
for split in get_dataset_split_names(
~~~~~~~~~~~~~~~~~~~~~~~^
path=dataset,
^^^^^^^^^^^^^
config_name=config,
^^^^^^^^^^^^^^^^^^^
token=hf_token,
^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
info = get_dataset_config_info(
path,
...<6 lines>...
**config_kwargs,
)
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
RoboCraftsman Synthetic Inspection Dataset (100-Sample Benchmark)
Executive Summary
The RoboCraftsman Synthetic Inspection Dataset addresses the data-scarcity bottleneck in software-defined sheet-metal forming cells. Training robust 3D vision and defect-detection models for aerospace/defense structures (fuselage panels, wing ribs, ducts) requires thousands of labeled defect examples that are extremely expensive to produce physically.
This dataset simulates the physical and kinematic envelope of a twin 7-axis robotic forming cell on linear rails, with structured-light 3D scanning, material-specific optics, and ground-truth deformation anomalies.
Dataset Benchmark Overview
| Parameter | Specification |
|---|---|
| Sample Count | 100 fully-annotated synthetic inspection scenes |
| Cell Environment | Dual 7-axis industrial arms on 3.5 m linear rails, central clamp frame |
| Sensors | End-effector structured-light 3D scanner + RGB camera |
| Point Cloud Density | 25,600 points per scan over a 500 x 500 mm FOV (~3.1 mm spacing); full-density available on request |
| Materials Simulated | Al 2024-T3, Al 7075-T6, Ti-6Al-4V, Stainless Steel 304 |
| Surface Finishes | Mill finish, brushed, polished, oxidized |
| Defect Classes | Springback deviation, wrinkling, excessive thinning, micro-cracks, surface scoring |
| Annotations | 3D bounding boxes, profile deviation, QC Pass/Fail verdict, joint angles, lighting state |
Multimodal Schema Architecture
Each sample links the 3D surface scan with exact kinematic and lighting states:
{
"sample_id": "ROBO_INSP_001",
"cell_environment": {
"cell_id": "RC_CELL_01",
"central_fixture_clamped": true,
"ambient_temp_c": 84.5,
"lighting": { "intensity_lux": 512.4, "color_temp_k": 4000, "shadow_occlusion_factor": 0.18 }
},
"robot_kinematics": {
"left_arm_rail_m": 2.145,
"left_arm_joints_deg": [12.4, -45.2, 89.1, 0.0, 32.1, -12.8, 180.0],
"right_arm_rail_m": 1.820,
"right_arm_joints_deg": [-10.1, 30.5, -75.0, 12.0, 45.0, 90.0, 0.0],
"scanner_end_effector": "StructuredLight_RGB_v2",
"standoff_distance_m": 0.998
},
"part_specifications": {
"part_type": "fuselage_skin_panel",
"material": "Al_2024",
"surface_finish": "brushed",
"nominal_thickness_mm": 1.5,
"sheet_dimensions_mm": [1200, 800, 1.5]
},
"sensor_specifications": {
"fov_mm": [500, 500],
"point_spacing_mm": 3.14,
"noise_model": "gaussian",
"noise_sigma_mm": 0.025,
"point_cloud_ref": "point_clouds/ROBO_INSP_001.pcd"
},
"ground_truth_inspection": {
"qc_verdict": "FAIL",
"max_profile_deviation_mm": 1.842,
"primary_defect_type": "springback_deviation",
"defect_severity_index": 0.742,
"defect_bounding_box_3d": {
"x_center_mm": 42.10, "y_center_mm": -12.50, "z_center_mm": 4.10,
"size_x_mm": 65.0, "size_y_mm": 48.0, "size_z_mm": 6.2
},
"affected_surface_area_mm2": 3120.0
}
}
Repository Contents
| Path | Description |
|---|---|
data/metadata.parquet |
100-row viewer table (full schema above) |
data/records/ |
Per-sample JSON inspection records |
data/point_clouds/ |
Structured-light point clouds (.pcd + .ply) |
data/renders/ |
RGB + depth renders (.png) |
data/masks/ |
Binary defect segmentation masks (.png) |
data/kinematics/ |
Twin 7-axis robot kinematics sidecars (.json) |
data/dataset_index_100.json |
Full dataset index |
generate_dataset.py |
Parametric generator (reproduce / scale) |
Quick Start Usage
from datasets import load_dataset
dataset = load_dataset("tryforge/robocraftsman-inspection-100")
sample = dataset["train"][0]
print("Sample ID:", sample["sample_id"])
print("Part Type:", sample["part_type"])
print("QC Verdict:", sample["qc_verdict"])
print("Max Profile Deviation:", sample["max_profile_deviation_mm"], "mm")
Contact & Technical Enquiries
Email: ravi@getforge.tech
LinkedIn: https://www.linkedin.com/in/raveekumar1/
- Downloads last month
- 47