|
Download README.md from djdumpling/fruit-box: direct link, hf CLI and curl.
- Browser
- Download file 726 Bytes
-
https://huggingface.co/datasets/djdumpling/fruit-box/resolve/main/README.md
- Command line
-
hf download hf://datasets/djdumpling/fruit-box/README.md
-
curl -L -o README.md https://huggingface.co/datasets/djdumpling/fruit-box/resolve/main/README.md
726 Bytes
| tags: | |
| - agent | |
| size_categories: | |
| - 100K<n<1M | |
| Trajectories for the fruit box game (play [here](https://en.gamesaien.com/game/fruit_box/)). An agent (scripted policy) repeatedly selects axis-aligned rectangles on a fixed 10×17 board of digits 1...9 so that the sum over the rectangle equals 10. Selected cells are cleared (set to 0). There is no gravity and no refill. | |
| Reward per step = number of non-zero cells cleared. Episodes end when no legal move remains. | |
| Data is formatted as follows: | |
| ``` | |
| { | |
| episode_id: string, | |
| step: int, | |
| grid: Arr[10][17], | |
| action: {"c1": int, "c2": int, "r1": int, "r2": int}, | |
| num_legal_actions: int, | |
| legal: bool, | |
| reward: int, | |
| done: bool, | |
| agent_tag: string, | |
| rng_seed: 1 | |
| } | |
| ``` |