The dataset viewer is not available for this subset.
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/webdataset/webdataset.py", line 80, in _split_generators
raise ValueError(
...<2 lines>...
)
ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
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.
NDB v1: Noble Detection Base
NDB is a synthetic object-detection corpus from Noble, built to pretrain Turbo, a compact NMS-free CNN detector. Turbo is then fine-tuned on COCO.
Each image is a composite. Real photographic cut-outs of 20 object classes are pasted onto real background photos together with unlabelled distractor objects. Compositing includes scale and placement priors, colour harmonisation, ground shadows, occlusion, and blur/noise augmentation.
Status: generation in progress, target 1,000,000 images. Shards are added as they finish.
Contents
shards/shard-NNNNNN.tar.gz: about 10 GB each. Every shard containsimages/NNNNNNNNNNNN.jpgandlabels/NNNNNNNNNNNN.yaml.- Image sizes: 640x480, 480x640, 640x640, 640x360.
ATTRIBUTIONS.csv: every source photo used, with creator, license and source page.
Label format
Each label file is a YAML mapping from class name to normalised [x1, y1, x2, y2] boxes:
car:
- [0.432812, 0.898438, 0.848437, 1.0]
suitcase:
- [0.157812, 0.576562, 0.434375, 0.748437]
An empty file ({}) means the image contains no labelled object.
Classes (20)
person, car, bicycle, motorcycle, bus, truck, animal, bird, backpack, handbag, suitcase, bottle, cup, chair, table, laptop, phone, book, food, traffic light
animal groups cat, dog, horse, sheep, cow, elephant, bear, zebra and giraffe. food groups
banana, apple, sandwich, orange, broccoli, carrot, hot dog, pizza, donut and cake.
Sources and license
The pipeline drew 2,465 source images (targets, distractors and backgrounds) from Wikimedia Commons, Openverse and Poly Haven, under these licenses:
| License | Assets |
|---|---|
| CC BY | 1,666 |
| CC BY-SA | 600 |
| CC0 | 146 |
| Public domain | 53 |
The composites are adaptations of CC BY-SA material, so this dataset is released under
CC BY-SA 4.0. When you redistribute images, credit the original creators as listed in
ATTRIBUTIONS.csv.
- Downloads last month
- 56