--- dataset_info: features: - name: inputs list: uint8 length: 81 - name: labels list: class_label: names: '0': '0' '1': '1' '2': '2' '3': '3' '4': '4' '5': '5' '6': '6' '7': '7' '8': '8' '9': '9' length: 81 - name: puzzle_index dtype: int32 splits: - name: train num_bytes: 733733000 num_examples: 1001000 - name: test num_bytes: 309902138 num_examples: 422786 download_size: 1407088809 dataset_size: 1043635138 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* license: mit tags: - trm - sudoku --- The `sudoku-extreme-1k-aug-1000` used in the Tiny Recursion Models (TRM) [paper](https://arxiv.org/abs/2510.04871) hosted here for convenience. > [!NOTE] > **NOTE: Value encoding.** The original TRM data uses 1..10. > We shift everything down by 1, so final values are **0..9** > where **0 = empty** and **1..9 = digits 1..9**. Generated with the [official code](https://github.com/SamsungSAILMontreal/TinyRecursiveModels) by running: ```bash python dataset/build_sudoku_dataset.py --output-dir data/sudoku-extreme-1k-aug-1000 --subsample-size 1000 --num-aug 1000 # 1000 examples, 1000 augments ``` and then ```python from pathlib import Path import numpy as np from datasets import Dataset, DatasetDict, Features, Sequence, Value data_dir = Path("./data/sudoku-extreme-1k-aug-1000/") def load_split(split): d = data_dir / split # note: shift down by 1 to have values in 0..9 instead of 1..10 X = np.load(d / "all__inputs.npy") - 1 y = np.load(d / "all__labels.npy") - 1 # exclude last entry to match shapes idx = np.load(d / "all__puzzle_indices.npy")[:-1] # ids = np.load(d / "all__puzzle_identifiers.npy") # omitting since all zeros return X, y, idx.astype(np.int32) X_train, y_train, idx_train = load_split("train") X_test, y_test, idx_test = load_split("test") features = Features( { "inputs": Sequence(Value("uint8"), length=81), "labels": Sequence(Value("uint8"), length=81), "puzzle_index": Value("int32"), } ) ds_train = Dataset.from_dict( {"inputs": X_train, "labels": y_train, "puzzle_index": idx_train}, features=features ) ds_test = Dataset.from_dict( {"inputs": X_test, "labels": y_test, "puzzle_index": idx_test}, features=features ) dset = DatasetDict({"train": ds_train, "test": ds_test}) repo_id = "emiliocantuc/sudoku-extreme-1k-aug-1000" dset.push_to_hub(repo_id, private=False) ``` Example usage in [this repo](https://github.com/ecntu/tiny-trm).