README.md

---
license: cc-by-4.0
thumbnail: https://huggingface.co/datasets/ArkadiumInc/public-assets/resolve/main/backgammon/example_move_01.webp
language:
- en
tags:
- backgammon
- games
- game-logs
- human-gameplay
- board-games
- reinforcement-learning
- game-ai
- strategy-games
pretty_name: Arkadium Backgammon Dataset
size_categories:
- 1K<n<10K
configs:
- config_name: gamelogs_with_images
  data_files:
  - split: train
    path: "data/gamelogs_with_images/train.parquet"
  default: true
- config_name: gamelogs
  data_files:
  - split: train
    path: "data/gamelogs/train.parquet"
default: gamelogs_with_images
---
<div align="center">
<table width="100%" cellpadding="0" cellspacing="0" 
       style="table-layout: fixed; border-collapse: collapse; border: none;">
  <tr>
    <td width="88" align="left" 
        style="vertical-align: middle; padding: 0 0px 0 0; border: none;">
      <img src="https://cdn-avatars.huggingface.co/v1/production/uploads/69455179a5c08dc77d1777a4/s1IcnLC_jDjyWpKYvnHrB.png" alt="Arkadium" width="72" 
           style="border-radius: 12px; display: block;">
    </td>
    <td align="left" style="vertical-align: middle; border: none;">
      <h1 style="margin: 0 0 4px; font-size: 28px;">Arkadium Backgammon Dataset</h1>
      <div style="margin: 0 0 6px;">High-Quality Human Backgammon Gameplay (Logs + Optional Board Screenshots)</div>
      <div>
        <a href="#-dataset-overview"><img src="https://img.shields.io/badge/Total_Matches-10,000-1a1a2e?style=for-the-badge&labelColor=2d2d44" alt="Matches"></a>
        &nbsp;
        <a href="#-dataset-overview"><img src="https://img.shields.io/badge/Screenshots-81,420-1a1a2e?style=for-the-badge&labelColor=2d2d44" alt="Screenshots"></a>
        &nbsp;
        <a href="#-dataset-overview"><img src="https://img.shields.io/badge/Total_Size-3.92_GB-1a1a2e?style=for-the-badge&labelColor=2d2d44" alt="Size"></a>
      </div>
    </td>
  </tr>
</table>
</div>

<!-- Game Screenshots Gallery -->
<div align="center">
<table width="100%" cellpadding="4" cellspacing="0" style="table-layout: fixed; border-collapse: collapse;">
<tr>
<td align="center" width="33%" style="padding: 4px 8px;">
<img src="https://huggingface.co/datasets/ArkadiumInc/public-assets/resolve/main/backgammon/example_move_01.webp" alt="Opening Position" style="width: 100%; height: auto; border-radius: 8px; box-shadow: 0 4px 6px rgba(0,0,0,0.3);">
<br><sub><b>Opening Position</b></sub>
</td>
<td align="center" width="33%" style="padding: 4px 8px;">
<img src="https://huggingface.co/datasets/ArkadiumInc/public-assets/resolve/main/backgammon/example_move_02.webp" alt="Mid-Game" style="width: 100%; height: auto; border-radius: 8px; box-shadow: 0 4px 6px rgba(0,0,0,0.3);">
<br><sub><b>Mid-Game Strategy</b></sub>
</td>
<td align="center" width="33%" style="padding: 4px 8px;">
<img src="https://huggingface.co/datasets/ArkadiumInc/public-assets/resolve/main/backgammon/example_move_03.webp" alt="Bear-Off Phase" style="width: 100%; height: auto; border-radius: 8px; box-shadow: 0 4px 6px rgba(0,0,0,0.3);">
<br><sub><b>Bear-Off Phase</b></sub>
</td>
</tr>
</table>
</div>

## Dataset Overview

This dataset contains **real human gameplay** from Arkadium's backgammon platform, providing researchers and developers with high-quality training data for game AI, reinforcement learning, and computer vision applications.

**RAW data only.** The primary dataset is the raw ZIP archives (`.mat`/`.webp`). Parquet files in `data/` contain only 10 example records for the HuggingFace Dataset Viewer and are not a full export. We do not provide a full Parquet version.

**Totals (both ZIPs):** 10,000 matches, ~3.92 GB raw ZIP size.

<table width="100%" cellpadding="6" cellspacing="0" style="table-layout: fixed; border-collapse: collapse;">
<thead>
<tr>
<th style="text-align: left;">Metric</th>
<th align="center"><code>gamelogs</code></th>
<th align="center"><code>gamelogs_with_images</code></th>
</tr>
</thead>
<tbody>
<tr>
<td>Raw match logs</td>
<td align="center">8,472</td>
<td align="center">1,528</td>
</tr>
<tr>
<td>Raw screenshots</td>
<td align="center">-</td>
<td align="center">81,420</td>
</tr>
<tr>
<td>Raw ZIP size</td>
<td align="center">~9.7 MB</td>
<td align="center">~3.9 GB</td>
</tr>
<tr>
<td>Raw format</td>
<td align="center">MAT (ZIP)</td>
<td align="center">MAT + WebP (ZIP)</td>
</tr>
</tbody>
</table>

## Use Cases

- **Game AI Training**: Train neural networks to predict optimal moves using supervised learning on human gameplay.
- **Reinforcement Learning**: Use as demonstration data for imitation learning or to initialize RL agents.
- **Computer Vision**: Board state recognition, piece detection, and game state extraction from images.
- **Strategy Analysis**: Study human decision-making patterns, opening strategies, and endgame techniques.

## Data Format

### File Structure

```
Arkadium Backgammon Dataset
│
├── Arkadium_Backgammon_full_data_gamelogs_001.zip
│   │   └── 8,472 match logs
│   ├── 1200001.mat
│   ├── 1200002.mat
│   └── ...
│
└── Arkadium_Backgammon_full_data_gamelogs_and_images_002.zip
    │   └── 1,528 matches with screenshots
    ├── 0217550.mat
    ├── attachments_0217550/
    │   ├── 0217550_0001.webp
    │   ├── 0217550_0002.webp
    │   └── ... (20-176 screenshots per match)
    └── ...
```

### Game Log Format (`.mat`)

Each `.mat` file contains a complete match record in a human-readable format:

```
; [Place "Arkadium"]
; [Player 1 "pl1"]
; [Player 2 "pl2"]
; [EventDate "2025.06.05"]
; [EventTime "15.46"]
; [CubeLimit "64"]
; [GP "987b847a-9ed5-9e0e-1ef4-b46757caeb39"]
; [RE "pl1+6"]

 3 point match

 Game 1
 pl1 : 0                  pl2 : 0
  1) 35: 8/3 6/3                 43: 24/20 24/21
  2) 61: 13/7 7/6                26: 21/15 15/13
  3) 33: 8/5* 8/5 24/21 24/21    54: 25/21 13/8
      ...
      Wins 6 points and the match
```

<br>

<details>
<summary><b>Header Field Reference</b></summary>

<br>

| Field | Description | Example |
|:------|:------------|:--------|
| `[Place]` | Gaming platform | `"Arkadium"` |
| `[Player 1]` | First player ID | `"pl1"` |
| `[Player 2]` | Second player ID | `"pl2"` |
| `[EventDate]` | Match date (YYYY.MM.DD) | `"2025.06.05"` |
| `[EventTime]` | Match time (HH.MM) | `"15.46"` |
| `[CubeLimit]` | Maximum doubling cube value | `"64"` |
| `[GP]` | Unique game UUID | `"987b847a-..."` |
| `[RE]` | Match result (winner + points) | `"pl1+6"` |

</details>

<details>
<summary><b>Move Notation Guide</b></summary>

<br>

| Notation | Meaning | Example |
|:---------|:--------|:--------|
| `X/Y` | Move checker from point X to point Y | `24/20` |
| `X/Y*` | Move and hit opponent's blot | `8/5*` |
| `X/0` | Bear off checker from point X | `6/0` |
| `Doubles => N` | Offer doubling cube at value N | `Doubles => 2` |
| `Takes` | Accept doubled stakes | — |
| `Drops` | Decline and forfeit game | — |

</details>

---

<details>
<summary><b>Parquet Preview (10 records)</b></summary>
<br>
Parquet files in `data/` are **only a 10-record preview** for the HuggingFace Dataset Viewer.  
They exist **exclusively for demo/browsing** and **do not represent the full dataset**.  
All primary data is in the raw ZIP archives above, and no full Parquet release is provided.

### Quick Start (Preview)

### Installation

```bash
pip install datasets pillow
```

### Load the Preview

```python
from datasets import load_dataset

# Load game logs preview (10 records)
ds = load_dataset("ArkadiumInc/ArkadiumBackgammon", "gamelogs")

# Load with board screenshots preview (10 records)
ds_img = load_dataset("ArkadiumInc/ArkadiumBackgammon", "gamelogs_with_images")
```

### Basic Usage

```python
# Access a single match
match = ds["train"][0]

print(f"Players: {match['player1']} vs {match['player2']}")
print(f"Date: {match['event_date']}")
print(f"Result: {match['result']}")
print(f"\nMoves:\n{match['moves']}")
```

### Working with Screenshots

```python
from PIL import Image
import io

# Get a match with images
match = ds_img["train"][0]

print(f"This match has {match['num_images']} board screenshots")

# Display the first board state
if match['images']:
    img = Image.open(io.BytesIO(match['images'][0]))
    img.show()
```

### Parquet Schema (Preview)

These columns describe the **Parquet preview** (10 records).

<div align="center">

| Column | Type | Description | Config |
|:-------|:-----|:------------|:------:|
| `data_id` | `string` | Unique record identifier | Both |
| `file_id` | `string` | Source file reference | Both |
| `place` | `string` | Gaming platform | Both |
| `num_images` | `int32` | Screenshot count | Images only |
| `images` | `list[Image]` | WebP board screenshots (stored as bytes/path in Parquet preview) | Images only |
| `player1` | `string` | First player ID | Both |
| `player2` | `string` | Second player ID | Both |
| `event_date` | `string` | Match date (YYYY.MM.DD) | Both |
| `event_time` | `string` | Match time (HH.MM) | Both |
| `cube_limit` | `int32` | Maximum cube value | Both |
| `match_length` | `int32` | Points needed to win | Both |
| `game_uuid` | `string` | Unique game identifier | Both |
| `result` | `string` | Match outcome | Both |
| `moves` | `string` | Parsed move sequence | Both |
| `raw_log` | `string` | Original .mat content | Both |

</div>

</details>

## Citation

If you use this dataset in your research, please cite:

```bibtex
@dataset{arkadium_backgammon_2026,
  title     = {Arkadium Backgammon Dataset: Human Gameplay with Board Screenshots},
  author    = {Arkadium Inc.},
  email     = {ai@arkadium.com},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/ArkadiumInc/ArkadiumBackgammon}
}
```

<div align="center">
<a href="https://creativecommons.org/licenses/by/4.0/">
<img src="https://img.shields.io/badge/License-CC_BY_4.0-lightgrey.svg?style=for-the-badge" alt="CC BY 4.0">
</a>

</div>

<div align="center">
<br />

<sub>

[Report Issues](https://huggingface.co/datasets/ArkadiumInc/ArkadiumBackgammon/discussions) · [Request Features](https://huggingface.co/datasets/ArkadiumInc/ArkadiumBackgammon/discussions)


</div>