datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
mrqa
Dataset Card for MRQA 2019
Dataset Summary
The MRQA 2019 Shared Task focuses on generalization in question answering. An effective question answering system should do more than merely interpolate from the training set to answer test examples drawn from the same distribution: it should also be able to extrapolate to out-of-distribution examples — a significantly harder challenge.
The dataset is a collection of 18 existing QA dataset (carefully selected subset of them) and… See the full description on the dataset page: https://huggingface.co/datasets/mrqa-workshop/mrqa.InsPLAD-workshop-pool
Dataset Card for InsPLAD Workshop Pool
This is a FiftyOne dataset with 1754 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("harpreetsahota/InsPLAD-workshop-pool")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/harpreetsahota/InsPLAD-workshop-pool.ICCV_workshop_testCVPR_workshop_efficiencyVLMCLERC-generation-workshopasi-fimsa-workshop-2026
ASI-FIMSA 2026 spatial omics workshop — staged data
Small, Colab-sized artifacts derived from the public 10x Genomics Atera whole-transcriptome
Xenium preview dataset of FFPE human breast cancer, prepared for a two-hour hands-on
workshop at the ASI-FIMSA meeting.
Notebooks and build scripts: https://github.com/xiao233333/ASI-FIMSA-workshop-2026
Attribution
Derived from Preview Data: Atera In Situ Gene Expression, FFPE Human Breast
Cancer.
Data © 10x Genomics, used… See the full description on the dataset page: https://huggingface.co/datasets/xiao233333/asi-fimsa-workshop-2026.referential-gesture-challenge
Referential Gesture Challenge Data Structure
Training data for the Referential Gesture Challenge at the 1st HSI Workshop (ECCV 2026). Contains ~2K pointing gesture clips from MM-Conv with synchronized speech, motion, and 3D scene graphs and ~1K pointing gesture from synthetic dataset with synchronized speech, motion, and 3D targets.
Current top-level structure:
data/
annotations/ # released training annotations
audio/ # released training audio… See the full description on the dataset page: https://huggingface.co/datasets/hsi-workshop/referential-gesture-challenge.real-fake-news-workshopOBLI_QA-generation-workshopfo_video_workshop_enriched
Dataset Card for harpreetsahota/fo_video_workshop_enriched
This is a FiftyOne dataset with 1144 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("harpreetsahota/fo_video_workshop_enriched")
# Launch the App
session =… See the full description on the dataset page: https://huggingface.co/datasets/harpreetsahota/fo_video_workshop_enriched.synthea-workshop-data
synthea-workshop-data
Workshop data for Transformer Architectures for Computational Social Science
(ICSC 2026, Oxford). A 50,000-patient subsample of a locally generated
Synthea population (Massachusetts,
ages 35–65, 10 years of history each), split 40,000 train / 10,000 test.
These are not real patients. Synthea is a rule-based simulator. Any structure
recovered from this data reflects the generator's rules rather than epidemiology,
and no result here says anything about human… See the full description on the dataset page: https://huggingface.co/datasets/carlomarxx/synthea-workshop-data.workshop3uncannyelevationWorkshop-CarDD-Dataset
Dataset Card for cardd_from_hub
This is a FiftyOne dataset with 2816 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Arvind1403/Workshop-CarDD-Dataset")
# Launch the App
session = fo.launch_app(dataset)
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/Arvind1403/Workshop-CarDD-Dataset.VISION-Datasets
Dataset Card for VISION Datasets
Dataset Summary
The VISION Datasets are a collection of 14 industrial inspection datasets, designed to explore the unique challenges of vision-based industrial inspection. These datasets are carefully curated from Roboflow and cover a wide range of manufacturing processes, materials, and industries. To further enable precise defect segmentation, we annotate each dataset with polygon labels based on the provided bounding box labels.… See the full description on the dataset page: https://huggingface.co/datasets/VISION-Workshop/VISION-Datasets.DL-hf-workshop-Persian_sentiment-datasetECHR_QA-generation-workshopWorkshop-CarDD-Dataset-Subset
Dataset Card for Workshop-CarDD-Dataset
This is a FiftyOne dataset with 500 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Arvind1403/Workshop-CarDD-Dataset-Subset")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/Arvind1403/Workshop-CarDD-Dataset-Subset.organize_stationery_cosmos_workshopThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_yaw.pos",
"wrist_roll.pos",
"gripper.pos"
]… See the full description on the dataset page: https://huggingface.co/datasets/manavgoel4/organize_stationery_cosmos_workshop.marimo-workshop-catalogue
marimo workshop product catalogue
A small, ready-to-search product catalogue for a hands-on marimo workshop, where students build a multimodal
(text and photo) product search engine that runs on a CPU.
File
What it is
catalogue.csv
3,000 products, 150 in each of 20 categories
images.zip
images/<product_id>.jpg, 240×320 JPEG
image_vectors.npy
(3000, 512) float32 CLIP image vectors, one per catalogue row, each of length 1
products.csv
a random 500-row subset used… See the full description on the dataset page: https://huggingface.co/datasets/PS4Research/marimo-workshop-catalogue.spot-telluride-workshop-dataset
Spot Telluride Workshop Dataset
Multimodal sensor data from a Boston Dynamics Spot D02 (Marble backpack compute), extracted from ROS bags for the Telluride Neuromorphic + AI Workshop. The dataset currently covers two locations, each with its own extraction tool and Hub layout root:
Location
Runs
Hub layout root
Extraction tool
Classroom demo
run1
classroom/run1/<modality>/
extract_demo_data.py
School
run1, run2
school/run1/<modality>/, school/run2/<modality>/… See the full description on the dataset page: https://huggingface.co/datasets/lorinachey/spot-telluride-workshop-dataset.cng-workshop-materialsso101_dual_handkerchief_v6_add_for_workshopcvpr_workshop_sam_prompt_clustersworkshop-materials-gsi
GenAI Workshop Materials (v2)
Portable workshop source for a fresh GPU machine. Runtime artifacts (.venv/, work/, Docker layers, checkpoints) are not included — notebooks recreate them on first run.
Layout
workshop-Materials/ — notebooks, small seed data, recipes, shared helpers
See workshop-Materials/M7-model_training/README.md for CPT → SFT → DPO chain
New machine setup
Clone/copy this tree (e.g. under /home/ubuntu/repo-content_v2).
Ensure a… See the full description on the dataset page: https://huggingface.co/datasets/s4sarath/workshop-materials-gsi.nutrition-workshop-sample
Nutrition Workshop Sample
A small food object detection set for teaching. It is a sample of the
Food Portion Benchmark (FPB),
cut down to 20 dishes so that a beginner can train a working detector on a free
Colab GPU in under ten minutes.
Built for the Nutrition and AI workshop at ISSAI, Nazarbayev University, for
participants with no programming background.
What is in it
Task
object detection, YOLO format
Dishes
20
Photos
1,127
Boxes
1,699… See the full description on the dataset page: https://huggingface.co/datasets/issai/nutrition-workshop-sample.droid-frames-workshop
Dataset Card for droid-frames
This is a FiftyOne dataset with 5172 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("dgural/droid-frames-workshop")
# Launch the App
session = fo.launch_app(dataset)
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/dgural/droid-frames-workshop.seqworkshop_dataset_20260919_152614This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"shape": [
6… See the full description on the dataset page: https://huggingface.co/datasets/Hin125/workshop_dataset_20260919_152614.workshop_dataset_1_20260919_152923This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"shape": [
6… See the full description on the dataset page: https://huggingface.co/datasets/ajnibu/workshop_dataset_1_20260919_152923.workshop_dataset_1_20260919_153940This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"shape": [
6… See the full description on the dataset page: https://huggingface.co/datasets/arivu10/workshop_dataset_1_20260919_153940.