datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Lurcher_10x
Lurcher 10x Microscopy Dataset
Dataset overview
This dataset consists of 2-D microscopy images of histologically stained 3-D structures in tissue sections through the cerebellum of 21 mouse brains. Animals are grouped into wild-type controls (n = 10) and Lurcher mutant mice (n = 11). The classification task is to distinguish Lurcher mutant mice from wild-type controls.
All images were captured at low magnification (10x) and stained with Cresyl violet, a general… See the full description on the dataset page: https://huggingface.co/datasets/USF-CS-Microscopy-Image-Analysis/Lurcher_10x.Openpdf-Analysis-Recognition
Openpdf-Analysis-Recognition
The Openpdf-Analysis-Recognition dataset is curated for tasks related to image-to-text recognition, particularly for scanned document images and OCR (Optical Character Recognition) use cases. It contains over 6,900 images in a structured imagefolder format suitable for training models on document parsing, PDF image understanding, and layout/text extraction tasks.
Attribute
Value
Task
Image-to-Text
Modality
Image
Format
ImageFolder… See the full description on the dataset page: https://huggingface.co/datasets/prithivMLmods/Openpdf-Analysis-Recognition.rocketleague-analysis
Rocket League Analysis
Local Rocket League replay analysis using Ballchasing API exports and plain DuckDB.
The report is meant to answer one practical question: what should I work on next from my saved replay sample?
Quick Start
mise install
mise run setup
mise run test
mise exec -- python scripts/analyze_scenarios.py \
--replay-dir /path/to/Rocket\ League/TAGame/Demos \
--limit 10
Start with CONTRIBUTING.md before changing the pipeline.
Replay files and… See the full description on the dataset page: https://huggingface.co/datasets/edmundmiller/rocketleague-analysis.retina-age-analysis
Retina Age Analysis Dataset
Dataset Description
This dataset contains 9,857 retinal fundus images from 5,393 patients for age prediction tasks.
Dataset Summary
Task: Age prediction from retinal fundus images
Images: 9,857 high-quality retinal images
Patients: 5,393 unique patients
Age Range: 5-97 years
Image Format: JPEG
Average Image Size: ~1 MB
Supported Tasks
Regression: Predict continuous age (5-97 years)
Classification: Predict age group (5… See the full description on the dataset page: https://huggingface.co/datasets/ramankamran/retina-age-analysis.trash-in-river-2025
Street Parade 2025 Dataset
Overview
This dataset was collected by SARA, a student initiative at ETH Zurich, to enable open research on trash presence in aquatic environments. It contains images of litter in the Limmat River in Zurich the day after the Street Parade (August 9, 2025). The dataset is intended for training and evaluating trash classification models.
Dataset summary
Collection date: August 9, 2025
Location: Kornhausbrücke, Zurich… See the full description on the dataset page: https://huggingface.co/datasets/SARA-smartphone-assisted-river-analysis/trash-in-river-2025.lewm-vjepa21-state-moe-shared-residual-wo-tokencl-analysis
LEWM V-JEPA2.1 State-MoE Shared-Residual Analysis
This dataset contains PNG visualizations and adjacent-epoch absolute-delta
summaries for the experiment
lewm_reasoning_vjepa21_vitL_tokens_100tasks_stateMoE_sharedResidual_pair_topkExcess_flat_headaware_woTokenCL_modify.
Contents
metadata.csv: parsed stage, epoch, layer, condition, scope, and image paths.
gallery_manifest.json: manifest consumed by the companion Static HTML Space.
summary.json: aggregate image… See the full description on the dataset page: https://huggingface.co/datasets/zoeloopy/lewm-vjepa21-state-moe-shared-residual-wo-tokencl-analysis.VanGogh_vs_TreeOilPainting_QuantumTorque_EnergyField_Analysis_Phase1_2025
Dataset Policy
VanGogh Vs. Tree Oil Painting: Quantum Torque Energy Field Analysis 2025
Structure Type
Free-form and Semi-structured Narrative
Core Principles
Each file is an independent analytical entity with its own identity.
Each file is the result of Autonomous AI–Human Co-analysis.
The structure is intentionally open, flexible, and adaptive, reflecting the natural reasoning process of the researcher, rather than forcing rigid… See the full description on the dataset page: https://huggingface.co/datasets/HaruthaiAi/VanGogh_vs_TreeOilPainting_QuantumTorque_EnergyField_Analysis_Phase1_2025.diagram-analysis-4Home-Credit-Risk-Analysis
🏦 Home Credit Risk Analysis — EDA, Feature Engineering & ML Models
This project explores the Home Credit Default Risk dataset throughEDA, statistical insights, feature engineering, clustering, regression, and classification modeling.
The goals of the project are to:
Understand borrower behavior and key risk drivers
Predict missing creditworthiness scores (EXT_SOURCE_2)
Develop a classification model to predict loan default
Deploy both models via HuggingFace for… See the full description on the dataset page: https://huggingface.co/datasets/guyshilo12/Home-Credit-Risk-Analysis.diagram-analysis-8risk-analysisiclr-rebuttal-analysis
ICLR Rebuttal Dynamics Dataset
Companion dataset for the paper "Rebuttal dynamics in machine-learning peer review: an LLM-instrumented analysis of ICLR 2024-2025".
Overview
This dataset contains 164,974 peer reviews from ICLR 2023-2026, enriched with LLM-extracted features for studying how author rebuttals influence reviewer score changes. Each review includes the full review text, pre- and post-rebuttal scores, Gemini-predicted scores, score-change verdicts, and… See the full description on the dataset page: https://huggingface.co/datasets/MlouisBE/iclr-rebuttal-analysis.qwen36-mtp-turbo-kv-analysis
Qwen3.6 MTP Turbo KV Runtime Analysis
This repository is a curated analysis artifact for local Qwen3.6-35B-A3B MTP GGUF inference experiments on Windows CUDA. It compares clean MTP llama.cpp, QuinsZouls llama-next TurboQuant, and the completed subset of Atomic TurboQuant runs under a fixed 64k context, MoE CPU offload, and Unsloth-aligned sampling settings.
The raw benchmark runs included incomplete and capability-incompatible rows. This repo keeps only completed, comparable… See the full description on the dataset page: https://huggingface.co/datasets/jakeatx/qwen36-mtp-turbo-kv-analysis.Mining-Analysis
Mining-Analysis Dataset
A research dataset for error analysis of satellite land-use classifiers, with model predictions, ground truth, RGB imagery, raw spectral bands, and per-month cloud masks. Used in a study of how cloud occlusion affects model predictions of West African land cover (mining, oil palm, rubber).
Background
We use satellite images to detect land-use changes in West Africa — specifically mining sites, oil palm plantations, and rubber plantations.… See the full description on the dataset page: https://huggingface.co/datasets/mqraitem/Mining-Analysis.VanGogh_vs_TreeOilPainting_QuantumTorque_EnergyField_Analysis_2025⚠️ Legacy Notice (Phase 1)
This dataset represents the early Center Index / Phase 1 design of the Tree Oil Painting × Van Gogh forensic framework.
It is kept online for historical transparency, methodology reference, and interface testing.
For all current physics-locked baselines, biomechanical signatures, and production-ready files, please refer to:
➡️ VanGogh_vs_TreeOilPainting_QuantumTorque_EnergyField_Analysis_2025
🌿 Interactive Image Space (Phase 1)ForensicImageGallery_Phase1… See the full description on the dataset page: https://huggingface.co/datasets/HaruthaiAi/VanGogh_vs_TreeOilPainting_QuantumTorque_EnergyField_Analysis_2025.trash-in-river-2024
Trash in River 2024 — Street Parade 2024 Dataset
Overview
This dataset was collected by SARA, a student initiative at ETH Zurich, to enable open research on trash presence in aquatic environments. It contains visual records of litter observed in the Limmat River in Zurich the day after the Street Parade (August 10, 2024). The dataset is intended for training and evaluating trash classification models.
Dataset summary
Collection date: August 10… See the full description on the dataset page: https://huggingface.co/datasets/SARA-smartphone-assisted-river-analysis/trash-in-river-2024.diagram-analysis-5SNAP25_ESM2_OpenFold3_Structural_Analysis
🧬 SNAP25 OpenFold3 Structural Analysis Dataset
Comprehensive structural predictions and therapeutic discovery data for 677 SNAP25 missense variants
🎯 Overview
This dataset provides the first comprehensive structural analysis of SNAP25 (Synaptosomal-Associated Protein 25 kDa) missense variants, generated to support therapeutic discovery for SNAP25-related developmental and epileptic encephalopathy (DEE-SNAP25).
SNAP25 is a critical component of the neuronal… See the full description on the dataset page: https://huggingface.co/datasets/SkyWhal3/SNAP25_ESM2_OpenFold3_Structural_Analysis.lewm-state-moe-shared-residual-wo-tokencl-jepa-top2-analysis
LEWM JEPA Top-2 State-MoE Analysis
This dataset contains PNG visualizations and adjacent-epoch absolute-delta
summaries for the experiment
100tasks_stateMoE_sharedResidual_pair_topkExcess_flat_headaware_woTokenCL_modify_jepa_top2.
Contents
metadata.csv: parsed stage, epoch, layer, condition, scope, and image paths.
gallery_manifest.json: manifest consumed by the companion Static HTML Space.
summary.json: aggregate image counts.
summaries/*.csv: train, validation… See the full description on the dataset page: https://huggingface.co/datasets/zoeloopy/lewm-state-moe-shared-residual-wo-tokencl-jepa-top2-analysis.LMOD-Cataract-1K-surgical-analysis-cot
Cataract-1K LLM-Generated Surgical Instructions
Dataset Overview
This dataset is derived from the Cataract-1K dataset (part of the LMOD benchmark) and enhanced using Qwen3-VL-30B-A3B-Thinking, a large vision-language model with reasoning capabilities. It is designed for training medical AI systems to provide actionable surgical guidance with transparent reasoning.
Generation Process
Source Data: Cataract-1K processed frames with segmentation annotations… See the full description on the dataset page: https://huggingface.co/datasets/mehti/LMOD-Cataract-1K-surgical-analysis-cot.startup-Investments-analysis
📊 StartUp Investments EDA
1. Background & Objectives
This project explores a comprehensive dataset of startup investments (sourced from Crunchbase) to uncover the primary factors that predict a startup's survival and trajectory in a competitive market.
Through this Exploratory Data Analysis (EDA), we analyze historical funding data, investment rounds, and market categories to determine which variables drive specific company outcomes - namely, whether a business… See the full description on the dataset page: https://huggingface.co/datasets/lia-prop13/startup-Investments-analysis.diabetes_eda_analysis
Diabetes Dataset — Exploratory Data Analysis (EDA)
This repository contains a diabetes-related tabular dataset and a complete Exploratory Data Analysis (EDA).The main objective of this project was to learn how to conduct a structured EDA, apply best practices, and extract meaningful insights from real-world health data.
The analysis includes correlations, distributions, group comparisons, class balance exploration, and statistical interpretations that illustrate how different… See the full description on the dataset page: https://huggingface.co/datasets/guyshilo12/diabetes_eda_analysis.Resume-Analysis-CoTR
Resume Reasoning and Feedback Dataset
Dataset Description
This dataset contains approximately 417 examples designed to facilitate research and development in automated resume analysis and feedback generation. Each data point consists of a user query regarding their resume, a simulated internal analysis (chain-of-thought) performed by an expert persona, and a final, user-facing feedback response derived solely from that analysis.
The dataset captures a two-step reasoning… See the full description on the dataset page: https://huggingface.co/datasets/Akhil-Theerthala/Resume-Analysis-CoTR.Deforest-Analysis
NRT Forest-Loss Test Set for Student Analysis
This package contains the fixed held-out test split used for a study of
near-real-time forest-loss detection from four HLS observations. It is an
analysis release: it includes inputs, labels, model outputs, and visual
renders, but no checkpoints or GPU-dependent code.
The intended analyses are prediction-shape comparison, per-connected-component
performance, and seasonal performance. Do not use this test set to select model… See the full description on the dataset page: https://huggingface.co/datasets/mqraitem/Deforest-Analysis.STARE-structured-analysis-of-the-retinaarXiv:2501.18921https://arxiv.org/abs/2501.18921
Gym-Exercise-Video-Analysis
Gym-Exercise-Video-Analysis
Gym-Exercise-Video-Analysis is a specialized multimodal video understanding dataset comprising 500 annotated gym workout and exercise clips. It is designed for fine-tuning and evaluating Video-Language Models (Video-LLMs), visual fitness coaches, and temporal exercise analysis systems. Each entry pairs exercise videos and extracted frame sequences with in-depth textual descriptions, biomechanical observations, form evaluations, and routine tracking.… See the full description on the dataset page: https://huggingface.co/datasets/prithivMLmods/Gym-Exercise-Video-Analysis.Visual_Emotional_Analysisreligious-artwork-analysis-data
Data
Download from Kaggle (needs an API token from https://www.kaggle.com/settings):
pip install kaggle
python data/download.py
Expected layout after download:
data/artwork_metadata.csv 3,997 rows — filename, religion (1,000 each of
buddhism / christianity / hinduism; 997 islam),
sub_religion, artist, title, year, place,
source, source_id, source_url, image_url
data/images/ the… See the full description on the dataset page: https://huggingface.co/datasets/cvikl/religious-artwork-analysis-data.diagram-analysis-6cable-representation-analysis
Cable representation analysis — ACT encoder, DG-5F / UR5e cable sorting
What the ACT encoder's internal representation holds about the cable being
handled. All figures come from a forward hook on policy.model.encoder: the
transformer tokens are mean-pooled into a single 512-d vector per frame.
Every split in this directory is episode-level. Frames inside one episode
are near-duplicates, so a frame-level split leaks the answer and inflates every
number below.
Model… See the full description on the dataset page: https://huggingface.co/datasets/Kaz55/cable-representation-analysis.