README.md

---
language:
- en
license: apache-2.0
size_categories:
- 1M<n<10M
pretty_name: CLRS Text Training Datasets
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: algo_name
    dtype: string
  splits:
  - name: train
    num_bytes: 2150691651
    num_examples: 2150000
  download_size: 531486090
  dataset_size: 2150691651
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---
# CLRS Text Training Datasets
A dataset containing the training data for CLRS-30 Text, generated using our [GitHub code](https://github.com/google-deepmind/clrs/tree/master/clrs/_src/clrs_text).
The full list of the 30 algorithms and lengths included is available [here](https://github.com/google-deepmind/clrs/blob/master/clrs/_src/clrs_text/generate_clrs_text.py#L53).

The full paper available on [ArXiv](https://arxiv.org/abs/2406.04229)

### Licensing Information

We are releasing this dataset under the terms of [Apache-2.0](https://choosealicense.com/licenses/apache-2.0/).

## Bibtex

If you use our dataset, please cite us:
```bibtex
@article{markeeva2024clrs,
  title={The CLRS-Text Algorithmic Reasoning Language Benchmark},
  author={Markeeva, Larisa and McLeish, Sean and Ibarz, Borja and Bounsi, Wilfried and Kozlova, Olga and Vitvitskyi, Alex and Blundell, Charles and Goldstein, Tom and Schwarzschild, Avi and Veli{\v{c}}kovi{\'c}, Petar},
  journal={arXiv preprint arXiv:2406.04229},
  year={2024}
}
```

## Full List of Algorithms
- activity_selector
- articulation_points
- bellman_ford
- bfs
- binary_search
- bridges
- bubble_sort
- dag_shortest_paths
- dfs
- dijkstra
- find_maximum_subarray_kadane
- floyd_warshall
- graham_scan
- heapsort
- insertion_sort
- jarvis_march
- kmp_matcher
- lcs_length
- matrix_chain_order
- minimum
- mst_kruskal
- mst_prim
- naive_string_matcher
- optimal_bst
- quickselect
- quicksort
- segments_intersect
- strongly_connected_components
- task_scheduling
- topological_sort