datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
procedural-pile
Procedural Pile
Verifiable reasoning problems, generated by code, formatted as supervised training data.
Procedural Pile contains 24,806,238 problems from 55 task families: arithmetic, equation systems and inverse word problems, first-order logic and formal proofs in Metamath, program execution and state tracking, belief tracking, planning, graph search, grammars, attribute grammars and regular expressions, SQL over tables, games, causal and probabilistic inference, and more.… See the full description on the dataset page: https://huggingface.co/datasets/reasoning-core/procedural-pile.procedural-typed-decisions
procedural-typed-decisions
Procedurally generated decision problems. Each row is one structured state
(JSON, or a table, CSV, key=value lines, or prose for the arithmetic,
retrieval, and aggregation configs) with several typed questions over that same state, following the
Jev / System One request shape: choice (pick one criterion), noul (a
number in [0, 1]; a probability or a yes/no), and score (an ordered rubric).
Every answer is computed exactly from the state by rules that… See the full description on the dataset page: https://huggingface.co/datasets/tasksource/procedural-typed-decisions.Procedural-City-Multimodal-Dataset-V2
Synthetic Urban Multimodal Dataset - Volume 2 (v0.2)
This is the second major release of the Synthetic Urban Multimodal Dataset, part of an ongoing observation project on AI-generated data ecosystems.
Volume 2 introduces a completely new collection of urban environments to the series, providing 40,000 high-precision files (8,000 completely new unique scenes × 5 modalities) generated entirely via the "Constructive Furnace" (our custom Blender-Python pipeline).
What's New… See the full description on the dataset page: https://huggingface.co/datasets/jp-cypress/Procedural-City-Multimodal-Dataset-V2.mental_rotation_3d_procedural_v2procedural-engine-sounds
Procedural Engine Sounds Dataset (Official)
⚠️ Canonical Source
This is the original and official release of the Procedural Engine Sounds Dataset, created by Robin Doerfler:
Project page (start here): https://rdoerfler.github.io/procedural-engine-sounds-page/
Hugging Face repository: https://huggingface.co/datasets/rdoerfler/procedural-engine-sounds
Zenodo DOI (primary citation): https://doi.org/10.5281/zenodo.16883336
Other versions of this dataset on… See the full description on the dataset page: https://huggingface.co/datasets/rdoerfler/procedural-engine-sounds.factory-manipulation-videos
Factory manipulation videos
Procedural Robotics is open sourcing a small set of our factory data so teams can assess its quality. The videos show workers performing factory tasks.
Contents
Seven continuous takes, 109 minutes in total.
Task
Station
Worker
Duration
File
cardboard manipulation
01
041
23.6 min
cardboard_manipulation_station01_worker041.mp4
cardboard manipulation
04
026
16.5 min
cardboard_manipulation_station04_worker026.mp4
defect… See the full description on the dataset page: https://huggingface.co/datasets/procedural-robotics/factory-manipulation-videos.procedural_2000_10procedural-reasoning-training-pool
Procedural reasoning training pool
Reasoning questions from 101 procedural generators, each of which writes a question, computes its
own answer and ships a verifier that scores an attempt at it, plus a collection of solved Sudoku
puzzles. Every answer is short and exactly checkable, so a trained model can be marked against the
key by a program and no judge is needed. Laid out twice. Train on either layer or on both.
pool.jsonl
Every generator rewritten into one… See the full description on the dataset page: https://huggingface.co/datasets/Emulated-Inc/procedural-reasoning-training-pool.procedural-truth-datasetProcedural-City-Multimodal-Dataset
Synthetic Urban Multimodal Dataset (v0.1)
This dataset contains 8,000 high-quality synthetic images of procedurally generated city models, designed specifically for training and evaluating AI models such as ControlNet and LoRA.
Dataset Description
All assets were generated using a custom Blender-Python automated pipeline. The dataset provides 100 unique urban architectures, each rendered from 8 directions with 10 different artistic styles.
Total Samples: 8,000 (Total 40… See the full description on the dataset page: https://huggingface.co/datasets/jp-cypress/Procedural-City-Multimodal-Dataset.arc-agi-3-procedural-games-v2
[!WARNING]
Deprecated — do not use for training. This corpus matches public action
horizons but not public induction difficulty. Familiar maze semantics make
its causal rules substantially easier than the real public games. It remains
online only for reproducibility and regression audits.
ARC-AGI-3 Procedural Games v2
This is a fresh replacement for the earlier transformation-classification
datasets. It contains complete executable multi-level games, not isolated
pixel… See the full description on the dataset page: https://huggingface.co/datasets/ritwika96/arc-agi-3-procedural-games-v2.mental_rotation_3d_proceduralaec-procedural-blenderprocedural-engine-sounds
Procedural Engine Sounds Dataset
Dataset Description
The Procedural Engine Sounds Dataset is a comprehensive collection of synthetically generated and annotated engine audio samples. This dataset contains procedurally generated high-resolution engine sounds free of confounding noises, with detailed time-aligned annotations, designed for research in audio processing, vehicle acoustics, and synthetic sound generation.
Associated Publication
The dataset and… See the full description on the dataset page: https://huggingface.co/datasets/huggingbear12/procedural-engine-sounds.procedural_2000_10
procedural_2000_10 (TsFile)
Apache TsFile version of sniper-qian/procedural_2000_10.
Converted rows: 1,716,102
Data files: ['data/procedural_2000_10.tsfile']
Usage
Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:
from pathlib import Path
from tsfile import TsFileReader
path = Path("data/procedural_2000_10.tsfile")
with TsFileReader(str(path)) as reader:
schemas = reader.get_all_table_schemas()
print("tables:"… See the full description on the dataset page: https://huggingface.co/datasets/THULab/procedural_2000_10.procedural-warmupprocedural-choices
Procedural Choices
Multiple-choice questions over procedural reasoning tasks. Every question is a prompt from a procedural
dataset with its verified reference answer plus up to 12 wrong options. Every wrong option is rejected by the
task's own scorer.
config
source
tasks
train
test
reasoning_core (default)
procedural-pretraining-pile @ rc14
54
49,092
4,909
reasoning_gym
reasoning-gym @ 531e215
68
47,073
4,581
from datasets import load_dataset
ds =… See the full description on the dataset page: https://huggingface.co/datasets/reasoning-core/procedural-choices.so101_vials_dr_procedural_layout_aware_250ep_20261009This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"observation.state": {
"dtype": "float32",
"shape": [
6
],
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos"… See the full description on the dataset page: https://huggingface.co/datasets/liorbenhorin-nv/so101_vials_dr_procedural_layout_aware_250ep_20261009.procedural-engine-sounds
Procedural Engine Sounds Dataset
Dataset Description
The Procedural Engine Sounds Dataset is a comprehensive collection of synthetically generated and annotated engine audio samples. This dataset contains procedurally generated high-resolution engine sounds free of confounding noises, with detailed time-aligned annotations, designed for research in audio processing, vehicle acoustics, and synthetic sound generation.
Associated Publication
The dataset and… See the full description on the dataset page: https://huggingface.co/datasets/trungdev/procedural-engine-sounds.Procedural_QuestionsProceduralclinical-5node-sedation-buf-lag-cpl-procedural-sedation-resp-suppression-v0.1
What this repo does
This dataset models a procedural sedation respiratory suppression cascade by measuring when sedation pressure rises, respiratory buffer capacity erodes, monitoring and escalation lags increase, and cross-team coupling tightens across procedure settings, crossing a phase transition into an unrecoverable respiratory suppression cascade state.
Core quad
sedationbuflagcpl
Prediction target
label_cascade_state
Row structure
One row… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/clinical-5node-sedation-buf-lag-cpl-procedural-sedation-resp-suppression-v0.1.procedural_gen_4operands
Dataset Card for "procedural_gen_4operands"
More Information needed
procedural_gen
Dataset Card for "procedural_gen"
More Information needed
Skillex-Procedural_memory_Matrixprocedural_gen_3operands
Dataset Card for "procedural_gen_3operands"
More Information needed
procedural-shapeprocedural_gen_2operands_1to5
Dataset Card for "procedural_gen_2operands_1to5"
More Information needed
procedural_gen_2operands
Dataset Card for "procedural_gen_2operands"
More Information needed
ML-Based-Procedural-Sound-Design
ML-Based Procedural Sound Design Artifacts
This repository stores large experiment artifacts for the ML-Based-Procedural-Sound-Design project.
Contents
explosion_dataset.zip: Full explosion dataset
embeddings_630k_audioset.zip: CLAP-based embedding outputs used in the project
kmeans_results.zip: K-means clustering outputs and related result files
Notes
This repository is used for artifact storage only. Source code, evaluation figures, documentation, and the… See the full description on the dataset page: https://huggingface.co/datasets/krvieni/ML-Based-Procedural-Sound-Design.