datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
geometric_shapes
Geometric Shapes Dataset
This dataset contains procedurally generated images of various geometric shapes with corresponding captions. It's designed for educational purposes and testing of diffusion models.
Dataset Overview
Content: 100,000 images of geometric shapes with detailed metadata
Image size: 512x512 pixels
Format: PNG images with CSV metadata
Features: Various shapes, colors, sizes, and descriptive captions
Purpose: Educational use for training and… See the full description on the dataset page: https://huggingface.co/datasets/anokimchen/geometric_shapes.ogbench
OgBench: Benchmarking Graph Neural Networks on Omics Data
OgBench is the first benchmark suite for graph-level prediction in the
n ≪ p regime characteristic of omics data, where the number of
patient samples n is much smaller than the number of nodes (genes or
proteins) p per graph.
Datasets
This repository contains four preprocessed omics graph classification
datasets:
Dataset
Modality
n
p
Task
HERITAGE
Proteomics
654
4,977
Exercise responder… See the full description on the dataset page: https://huggingface.co/datasets/geometric-intelligence/ogbench.geometric-vocab
Research Update 9/13/2025
The MULTITUDE of tests I've ran show that with weighted decay these pentachora are more likely to collapse to zero than retain utility when trained directly. However, when used as a starting point and then only minorly shifted as a trajectory towards a goal, they are more likely to retain full cohesion and even be backtrackable. The constellations show that this is more than a probable solution, it's a likely solution to work.
When the anchor [n, 1… See the full description on the dataset page: https://huggingface.co/datasets/AbstractPhil/geometric-vocab.GeometricShapeDataset
Geometric Shape Dataset
Introduction to Dataset
The Geometric Shape Dataset is a large-scale, synthetically generated computer vision dataset containing 1,680,000 instances of various geometric shapes and lines. It is designed for training image classification, feature extraction, and pattern recognition models across different scales.
The dataset is available in three distinct resolution configurations: 30x30, 40x40, and 50x50 (560,000 images per resolution).… See the full description on the dataset page: https://huggingface.co/datasets/OmerTurk1/GeometricShapeDataset.photo_geometric
Dataset Card for "photo_geometric"
More Information needed
halvest-geometric
HALvest-Geometric
Citation Network of Open Scientific Papers Harvested from HAL
Dataset Summary
overview:
French and English fulltexts from open papers found on Hyper Articles en Ligne (HAL) and its citation network.
You can download the dataset using Hugging Face datasets:
from datasets import load_dataset
ds = load_dataset("Madjakul/HALvest-Geometric", "en")
Details
Nodes
Papers: 18,662,037
Authors: 238,397… See the full description on the dataset page: https://huggingface.co/datasets/almanach/halvest-geometric.geometric-vocab-english-full-a-to-z
Update 9/15/2025
I believe this variation may have some genuinely complex potential beyond the alternatives. The lexical organization and capability of the old-style definitions may have impacted the outcome in a more literal and latin-sense than the wordnet lexical variations concatenated. I'll be running some more tests on this one in the coming days to determine if this formula has a different behavioral response than the more "robust" and "anchored" variations.
I'm open to any… See the full description on the dataset page: https://huggingface.co/datasets/AbstractPhil/geometric-vocab-english-full-a-to-z.geometric_spatial_compose_reasoninggeometric-vocab-4096dViRL39K-GradeSchool__Geometrictask119_semeval_2019_task10_geometric_mathematical_answer_generation
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task119_semeval_2019_task10_geometric_mathematical_answer_generation
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task119_semeval_2019_task10_geometric_mathematical_answer_generation.geometric_optics_physical_consistency_eval
Geometric Optics – Physical Consistency Evaluation Dataset
This dataset contains visual failure cases in geometric optics for multimodal image generation models.
Scope
The dataset focuses on physical and geometric inconsistencies related to:
mirror reflections (law of reflection)
refraction and dispersion in prisms
light ray direction consistency
shadow direction vs light source
camera–object–light spatial coherence
Motivation
Current image generation models… See the full description on the dataset page: https://huggingface.co/datasets/8Planetterraforming/geometric_optics_physical_consistency_eval.geometric-vocab-768dhttps://github.com/AbstractEyes/lattice_vocabulary
geometric-vocab-256dhttps://github.com/AbstractEyes/lattice_vocabulary
The_Geometric_Incompleteness_of_Reasoning
The Geometric Incompleteness of Reasoning
Author: Zixi "Oz" Li
DOI: 10.57967/hf/7080
License: Apache-2.0
Citation
@misc{oz_lee_2025,
author = {Oz Lee},
title = {The_Geometric_Incompleteness_of_Reasoning (Revision 7093c66)},
year = 2025,
url = {https://huggingface.co/datasets/OzTianlu/The_Geometric_Incompleteness_of_Reasoning},
doi = {10.57967/hf/7080},
publisher = {Hugging Face}
}
Core Thesis… See the full description on the dataset page: https://huggingface.co/datasets/OzTianlu/The_Geometric_Incompleteness_of_Reasoning.geometric-vocab-1024dhttps://github.com/AbstractEyes/lattice_vocabulary
geometric-vocab-16dgeometric-vocab-512dhttps://github.com/AbstractEyes/lattice_vocabulary
2d-geometric-shapes-datasetgeometric-vocab-1000dgeometric-vocab-32dhttps://github.com/AbstractEyes/lattice_vocabulary
geometric-vocab-50dgeometric-vocab-64dhttps://github.com/AbstractEyes/lattice_vocabulary
geometric-vocab-100dgeometric-vocab-2048dgeometric-vocab-128dhttps://github.com/AbstractEyes/lattice_vocabulary
geometric-shapes-clipgeometric-shapes-datasetUnroll-Qwen2.5-7B-Instruct_1754687747_eval_3444_gpqadiamond_geometric_num_prune_attn_6
chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754687747_eval_3444_gpqadiamond_geometric_num_prune_attn_6
Precomputed model outputs for evaluation.
Evaluation Results
GPQADiamond
Average Accuracy: 28.45% ± 0.73%
Number of Runs: 3
Run
Accuracy
Questions Solved
Total Questions
1
28.79%
57
198
2
29.80%
59
198
3
26.77%
53
198
Unroll-Qwen2.5-7B-Instruct_1754916227_eval_6a28_math500_geometric_num_prune_ffn_4_run-002
chengfu0118/Unroll-Qwen2.5-7B-Instruct_1754916227_eval_6a28_math500_geometric_num_prune_ffn_4_run-002
Precomputed model outputs for evaluation.
Evaluation Results
MATH500
Accuracy: 38.20%
Accuracy
Questions Solved
Total Questions
38.20%
191
500