---
license: apache-2.0
dataset_info:
  features:
  - name: problem
    dtype: string
  - name: answer
    dtype: string
  - name: database
    dtype: string
  splits:
  - name: train
    num_bytes: 37542213
    num_examples: 9428
  - name: test
    num_bytes: 4916219
    num_examples: 1534
  download_size: 4591783
  dataset_size: 42458432
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
task_categories:
- reinforcement-learning
language:
- en
---

# BIRD-RL

<!-- Provide a quick summary of the dataset. -->

This dataset is a processed dataset of BIRD-SQL for Post-Training.