neocr_dataset / README.md
Berzerker's picture
added dataset neocr
ec12f14
|
raw
history blame
218 Bytes
---
dataset_info:
features:
- name: image
dtype: image
- name: output_json_dumpsed
dtype: string
configs:
- config_name: default
data_files:
- split: train
path: data/*.parquet
language:
- en
---