metadata

license: cc-by-4.0
dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': airplane
            '1': automobile
            '2': bird
            '3': cat
            '4': deer
            '5': dog
            '6': frog
            '7': horse
            '8': ship
            '9': truck
  splits:
    - name: train
      num_bytes: 178662714
      num_examples: 90000
    - name: validation
      num_bytes: 180126542
      num_examples: 90000
    - name: test
      num_bytes: 178913694
      num_examples: 90000
  download_size: 771149160
  dataset_size: 537702950
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*
task_categories:
  - image-classification
size_categories:
  - 100K<n<1M

Dataset Card for CINIC-10

CINIC-10 has a total of 270,000 images equally split amongst three subsets: train, validate, and test. This means that CINIC-10 has 4.5 times as many samples than CIFAR-10.

Dataset Details

In each subset (90,000 images), there are ten classes (identical to CIFAR-10 classes). There are 9000 images per class per subset. Using the suggested data split (an equal three-way split), CINIC-10 has 1.8 times as many training samples as in CIFAR-10. CINIC-10 is designed to be directly swappable with CIFAR-10. To understand the motivation behind the dataset creation please visit the GitHub repository.

Dataset Sources

Repository: https://github.com/BayesWatch/cinic-10
Paper: https://arxiv.org/abs/1810.03505
Dataset: http://dx.doi.org/10.7488/ds/2448
Benchmarking, Papers with code: https://paperswithcode.com/sota/image-classification-on-cinic-10

Use in FL

In order to prepare the dataset for the FL settings, we recommend using Flower Dataset (flwr-datasets) for the dataset download and partitioning and Flower (flwr) for conducting FL experiments.

To partition the dataset, do the following.

Install the package.

pip install flwr-datasets[vision]

Use the HF Dataset under the hood in Flower Datasets.

from flwr_datasets import FederatedDataset
from flwr_datasets.partitioner import IidPartitioner

fds = FederatedDataset(
    dataset="flwrlabs/cinic10",
    partitioners={"train": IidPartitioner(num_partitions=10)}
)
partition = fds.load_partition(partition_id=0)

Dataset Structure

Data Instances

The first instance of the train split is presented below:

{
  'image': <PIL.PngImagePlugin.PngImageFile image mode=RGB size=32x32>,
  'label': 0
}

Data Split

DatasetDict({
    train: Dataset({
        features: ['image', 'label'],
        num_rows: 90000
    })
    validation: Dataset({
        features: ['image', 'label'],
        num_rows: 90000
    })
    test: Dataset({
        features: ['image', 'label'],
        num_rows: 90000
    })
})

Citation

When working with the CINIC-10 dataset, please cite the original paper. If you're using this dataset with Flower Datasets and Flower, cite Flower.

BibTeX:

Original paper:

@misc{darlow2018cinic10imagenetcifar10,
      title={CINIC-10 is not ImageNet or CIFAR-10}, 
      author={Luke N. Darlow and Elliot J. Crowley and Antreas Antoniou and Amos J. Storkey},
      year={2018},
      eprint={1810.03505},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/1810.03505}, 
}

Flower:

@article{DBLP:journals/corr/abs-2007-14390,
  author       = {Daniel J. Beutel and
                  Taner Topal and
                  Akhil Mathur and
                  Xinchi Qiu and
                  Titouan Parcollet and
                  Nicholas D. Lane},
  title        = {Flower: {A} Friendly Federated Learning Research Framework},
  journal      = {CoRR},
  volume       = {abs/2007.14390},
  year         = {2020},
  url          = {https://arxiv.org/abs/2007.14390},
  eprinttype    = {arXiv},
  eprint       = {2007.14390},
  timestamp    = {Mon, 03 Aug 2020 14:32:13 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-2007-14390.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Dataset Card Contact

If you have any questions about the dataset preprocessing and preparation, please contact Flower Labs.