---
license: apache-2.0
task_categories:
- image-to-text
tags:
- chess
- HCS
- handwriting
- scoresheets
- forms
size_categories:
- 10K<n<100K
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype: string
  splits:
  - name: train
    num_bytes: 517848025.198
    num_examples: 13731
  download_size: 642552652
  dataset_size: 517848025.198
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---
# Disclaimer
This is not data that I created. It originally came from the Paper [_Digitization of Handwritten Chess Scoresheets with a BiLSTM Network_](https://www.mdpi.com/1477430)

You can also find the dataset [here](https://sites.google.com/view/chess-scoresheet-dataset) and [here](https://tc11.cvc.uab.es/datasets/HCS_1)

# Datasets
There are 2 versions of this dataset
- processed_hcs Dataset where you are right now
- unprocessed_hcs where the whole scoresheet can be seen [here](https://huggingface.co/datasets/BenjaminKost/unprocessed_hcs)

# Desciption
The Handwritten Chess Scoresheet Datase (HCS) contains a set of single and double paged chess scoresheet images with ground truth labels. 
The processed images are a list of the extracted move boxes with the corresponding labels. 
The labels are strings of the chess moves shown in the move boxes.

# Usage
## In Python script
```python
import datasets
unprocessed_hcs = datasets.load_dataset("Benjaminkost/unprocessed_hcs")

# Show first image
first_image = unprocessed_hcs["train"][0]["image"]
first_label = unprocessed_hcs["train"][0]["labels"]

first_image.show()
print(first_label)
```