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/webdataset/webdataset.py", line 81, in _split_generators
                  first_examples = list(islice(pipeline, self.NUM_EXAMPLES_FOR_FEATURES_INFERENCE))
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 32, in _get_pipeline_from_tar
                  fs: fsspec.AbstractFileSystem = fsspec.filesystem("memory")
                                                  ~~~~~~~~~~~~~~~~~^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/fsspec/registry.py", line 302, in filesystem
                  cls = get_filesystem_class(protocol)
                File "/usr/local/lib/python3.14/site-packages/fsspec/registry.py", line 239, in get_filesystem_class
                  raise ValueError(f"Protocol not known: {protocol}")
              ValueError: Protocol not known: memory
              
              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 71, 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.

Light-trap insects: two illumination domains

Two YOLO object-detection datasets built from light-trap photographs:

  • archives/full_illumination_yolo.tar — full illumination;
  • archives/round_spot_yolo.tar — a circular light spot.

Both domains use the same single training class: insect.

Structure

Each TAR archive expands to:

annotations/       COCO 1.0 annotations
images/train/
images/val/
images/test/
labels/train/
labels/val/
labels/test/
dataset.yaml        Ultralytics YOLO configuration

SHA-256:

2EBFC768882C03E2AC54F0AA18AB5EAA553D8CBF8B67809AC1B8F988ED45CD3D  full_illumination_yolo.tar
10D11070617A307D9B4E947F81C01C5CC2DA11E5D51E9CE299CAF13713C7ED51  round_spot_yolo.tar

The split is assigned by complete capture nights, not by individual tiles. Capture-night, batch, and paired-frame leakage between train, validation, and test was checked and not detected.

Dataset summary

Domain Source images Retained tiles Tile annotations
Full illumination 2,402 26,240 41,618
Round spot 1,778 20,756 30,119
Total 4,180 46,996 71,737

Tiles are 1280 x 1280 with 20% overlap. Negative tiles were retained at a target ratio of 0.25.

Annotation provenance and limitations

The annotations are pseudo-labels produced by a bootstrap YOLO detector and automated QA. Coarse CLIP morphogroup suggestions from the source annotation were collapsed to a single insect class for training; debris was removed.

These annotations have not been fully reviewed by an entomologist. They should not be treated as scientifically verified species-level labels. Reported model metrics measure agreement with the current pseudo-labels. The independent test split contains one capture night, so seasonal generalization is not established.

Geometry QA reported no invalid, out-of-bounds, tiny, or duplicate boxes in the canonical source annotation.

Ultralytics usage

mkdir full_illumination
tar -xf archives/full_illumination_yolo.tar -C full_illumination
yolo detect train data=full_illumination/dataset.yaml model=yolo11n.pt imgsz=1024

Extract the second archive into round_spot/ and replace the dataset path with round_spot/dataset.yaml for the second domain.

Downloads last month
24