metadata.yaml

---
license: mit
task_categories:
  - reinforcement-learning
  - human-computer-interaction
task_ids:
  - mouse-behavior-modeling
  - rl-from-demonstration
language:
  - en
size_categories:
  - n<1K
tags:
  - mouse-input
  - human-in-the-loop
  - reinforcement-learning
  - webxos-2026
  - space-zombies-rl
pretty_name: "Space Zombies RL Mouse Behavior Dataset"
dataset_info:
  features:
    - name: state
      dtype: 
        playerHealth: float32
        playerPosition: 
          x: float32
          z: float32
        zombiesCount: int32
        wave: int32
        datasetLevel: int32
    - name: action
      dtype: string
    - name: reward
      dtype: float32
    - name: next_state
      dtype: 
        playerHealth: float32
        playerPosition: 
          x: float32
          z: float32
        zombiesCount: int32
        wave: int32
        datasetLevel: int32
    - name: done
      dtype: bool
    - name: timestamp
      dtype: int64
  splits:
    train:
      num_examples: 716
    test:
      num_examples: 179
---
# Space Zombies RL Mouse Behavior Dataset

This dataset contains mouse interaction data collected from the "Space Zombies RL" game, designed for reinforcement learning research on human input patterns.

## Dataset Description

- **Total Samples**: 896
- **Collection Date**: 2026-01-05
- **Game Version**: Space Zombies RL v1.0 by webXOS 2026
- **Purpose**: RL training from human demonstrations, mouse behavior modeling, adaptive interface research

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("your-username/space-zombies-rl")
```

## Citation

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

```
@dataset{space_zombies_rl_2026,
  title={Space Zombies RL Mouse Behavior Dataset},
  author={webXOS Research},
  year={2026},
  publisher={Hugging Face}
}
```