metadata
dataset_info:
features:
- name: db_id
dtype: string
- name: schema
dtype: string
splits:
- name: train
num_bytes: 217487
num_examples: 166
download_size: 78111
dataset_size: 217487
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
The purpose of this dataset is creating schema generation for databases, which collected from https://github.com/VinAIResearch/ViText2SQL/data/[syllable/word]-level/tables.json
The usage of this dataset is at https://huggingface.co/datasets/TeeA/ViText2SQL-2