metadata

dataset_info:
  features:
    - name: image
      dtype:
        image:
          decode: false
    - name: manual_label
      dtype: int64
    - name: logits
      dtype: float64
    - name: prob
      dtype: float64
    - name: model_label
      dtype: int64
    - name: start_datetime
      dtype: timestamp[ns]
    - name: antenna
      dtype: string
  splits:
    - name: test
      num_bytes: 6796135624.293
      num_examples: 30549
    - name: val
      num_bytes: 6793230118.642
      num_examples: 30533
    - name: train
      num_bytes: 54294424548.544
      num_examples: 243668
  download_size: 67497182964
  dataset_size: 67883790291.479004
configs:
  - config_name: default
    data_files:
      - split: test
        path: data/test-*
      - split: val
        path: data/val-*
      - split: train
        path: data/train-*
license: mit
task_categories:
  - text-classification
size_categories:
  - 100K<n<1M

E-Callisto Radio Sunburst Dataset

The E-Callisto Radio Sunburst Dataset is a collection of 300,000 images captured from the e-CALLISTO (Compact Astronomical Low-frequency, Low-cost Instrument for Spectroscopy and Transportable Observatory) network. This dataset focuses on solar radio bursts and is designed to aid researchers and enthusiasts in the study and detection of these phenomena.

Dataset Description

Total Images: 300,000
Burst Images: Approximately 30'000

The dataset exhibits a class imbalance with a ratio of 1:10 between burst and non-burst images.

Labels, Logits and Probability

There are manual labels, found here and labels created by a ResNet Model on a clean subset of the data. The probability is the output of the softmax function on logits. To make the probability more meaningful, the temperature was tuned.

More information here about the model: Identification of Radio Solar Bursts with ResNet Architecture and Augmented Datasets.

Citation

If you use this dataset in your research, please cite it as:

@dataset{e_callisto_radio_sunburst_2023,
  author       = {Vincenzo Timmel},
  title        = {E-Callisto Radio Sunburst Dataset},
  year         = {2024},
  publisher    = {\url{https://www.fhnw.ch/de/die-fhnw/hochschulen/ht/institute/institut-fuer-data-science}},
  howpublished = {\url{https://huggingface.co/datasets/i4ds/ecallisto_radio_sunburst/}},
}

Contact

For questions or feedback, please contact [email protected].