Downloads · 30 days
0
Sudoku_superposition
Avra98/Sudoku_superposition
Sudoku_superposition is a machine learning model from Avra98. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Concrete board assignments for a 12-stage Sudoku latent curriculum. Each stage keeps a candidate set per cell; this dataset materializes those sets as ordinary (row, col, value) sequences so training is standard next-…
Access
Public
Updated Aug 26, 2026
Repo size
373 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.tar110 GB · 86%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Aug 14, 2026
- Updated
- Aug 26, 2026
- SHA
- 5bce762f
Model card
From the Hugging Face model README
Sudoku superposition instances
Concrete board assignments for a 12-stage Sudoku latent curriculum.
Each stage keeps a candidate set per cell; this dataset materializes
those sets as ordinary (row, col, value) sequences so training is
standard next-token cross-entropy (no multi-hot candidate head).
Repo: Avra98/Sudoku_superposition
Dataset
| split | puzzles | instances | mean / puzzle |
|---|---|---|---|
| train | 1,804,462 | 89,236,838 | 49.45 |
| test | 99,999 | 4,945,532 | 49.46 |
Mean instances per stage (train, stage 0 → 11):
6.28, 5.96, 5.65, 5.34, 5.02, 4.68, 4.32, 3.88, 3.33, 2.58, 1.41, 1.00
Stage 11 is the unique solution. No empty (puzzle, stage) pair.
Files (data/)
| file | shape | dtype | role |
|---|---|---|---|
{split}_assignments.npy | (M, 81) | uint8 | one full board per instance, cell r*9+c |
{split}_starts.npy | (N, 12) | int32 | first row in assignments for (puzzle, stage) |
{split}_counts.npy | (N, 12) | uint8 | number of instances for (puzzle, stage) |
{split}_index.npy | (M, 3) | int32 | [puzzle_idx, stage, k] (optional; starts/counts are enough) |
The trainer only needs assignments, starts, and counts.
Puzzle clue/solution arrays are not in this repo (they are the
original Sudoku npy files). Candidate masks used to build the
instances live in datasets_multicandidate_s12/.
How a training example is built
- Take the usual solver-order sequence: clue triples, then K latent placeholders, then empty-cell triples.
- Sample one instance for the current curriculum stage.
- Rewrite values only. Location order stays solver-order.
- Predict the output triples with softmax CE. Latent slots are not predicted.
Curriculum stage t (1..12) trains on stage-(t-1) instances.
Stage 12 targets the unique solution.
Code (code/)
code/train/— JAX trainer (data.py,trainer.py,evaluater.py,train_and_evaluate.py,train_backtrack.py,main.py,model.py)code/build_superposition_dataset.py— instance generatorcode/build_instance_offsets.py—starts/countstablescode/superposition_instances.py— per-puzzle instance samplercode/sbatch_instance_latent.sh— Slurm launch (feanor / H200)
Set SUDOKU_INSTANCE_DIR to the data/ folder (or a local copy).
from huggingface_hub import snapshot_download
snapshot_download("Avra98/Sudoku_superposition", local_dir="Sudoku_superposition")
- License
- apache-2.0
- Created
- Aug 14, 2026
- Updated
- Aug 26, 2026