---
license: other
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
source_datasets:
- 0jl/SPair-71k
dataset_info:
  features:
  - name: images
    list:
      image:
        decode: false
  - name: inputs
    dtype: string
  - name: labels
    dtype:
      class_label:
        names:
          '0': r0c0
          '1': r0c1
          '2': r0c2
          '3': r0c3
          '4': r0c4
          '5': r0c5
          '6': r0c6
          '7': r1c0
          '8': r1c1
          '9': r1c2
          '10': r1c3
          '11': r1c4
          '12': r1c5
          '13': r1c6
          '14': r2c0
          '15': r2c1
          '16': r2c2
          '17': r2c3
          '18': r2c4
          '19': r2c5
          '20': r2c6
          '21': r3c0
          '22': r3c1
          '23': r3c2
          '24': r3c3
          '25': r3c4
          '26': r3c5
          '27': r3c6
          '28': r4c0
          '29': r4c1
          '30': r4c2
          '31': r4c3
          '32': r4c4
          '33': r4c5
          '34': r4c6
          '35': r5c0
          '36': r5c1
          '37': r5c2
          '38': r5c3
          '39': r5c4
          '40': r5c5
          '41': r5c6
          '42': r6c0
          '43': r6c1
          '44': r6c2
          '45': r6c3
          '46': r6c4
          '47': r6c5
          '48': r6c6
  - name: metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 44059042998
    num_examples: 100000
  - name: validation
    num_bytes: 43287942
    num_examples: 100
  - name: test
    num_bytes: 43615519
    num_examples: 100
  download_size: 44053985482
  dataset_size: 44145946459
---

# spair71k-grid

SPair-71k native correspondences: mark the source, classify the target 7x7 cell.

Native train/validation/test image sets must be disjoint. Every supplied
keypoint correspondence is eligible; a bounded mirror samples keypoints
uniformly rather than selecting one fixed point per pair. Target images
retain their original encoded bytes. PASCAL/Flickr source terms apply.

Original data: [0jl/SPair-71k](https://huggingface.co/datasets/0jl/SPair-71k). Repackaged as parquet for [tasksource](https://github.com/sileod/tasksource) by [scripts/repackage_dataset/regions.py](https://github.com/sileod/tasksource/blob/main/scripts/repackage_dataset/regions.py#L426).

[Conversion provenance](provenance.json): source revisions, hashes, configuration and exclusions.

## Release scope

Bounded samples of the eligible native annotations; full classification ontologies are retained.

- default: train: 100,000, validation: 100, test: 100.

See [provenance.json](provenance.json) for source pins, selection seeds, rendering configuration and exclusions.