datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
dronescapes2_annotated_train_set
Dataset Card for DroneScapes2 (annotated train set)
This is a FiftyOne dataset with 218 samples. It's a subset of this split from the original repo.
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 =… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/dronescapes2_annotated_train_set.spacr-example-annotate
spaCR — Annotate and Classify example data
Example input for the Annotate and Classify modules of
spaCR. It is the output of a Measure
run, so both modules can be exercised without segmenting or measuring anything
first.
What is here
Path
What it is
data/
2,341 single-cell PNG crops, foldered by phenotype
measurements.db
The measurements, plus png_list and the annotation tables
measurements/active_learning/
The model card from the first annotation… See the full description on the dataset page: https://huggingface.co/datasets/einarolafsson/spacr-example-annotate.Deepface_Annotated_3K
Deepface Annotated 3K Dataset
Deepface_Annotated_3K is a synthetic facial image dataset containing 3K AI-generated faces from StyleGAN2.Each image is automatically annotated with demographic attributes like:
Age (in years)
Gender (with prediction confidence)
Dominant Race (White, Asian, Latino Hispanic, Indian, etc.)
The dataset is designed for research on fairness, bias detection, demographic classification, and synthetic face representation.
Structure information… See the full description on the dataset page: https://huggingface.co/datasets/Subh775/Deepface_Annotated_3K.open-images-v7-mini
AnnotateIt · Open the app · Models & datasets · Documentation
AnnotateIt Open Images V7 Mini Collection
Eight small, real-world, AnnotateIt-compatible datasets curated from the Open Images V7 validation split. Each archive contains 50–200 images, production-exported annotations, source details, and per-image attribution.
Only images whose official Open Images metadata lists CC BY 2.0 are included. Open Images annotations are CC BY 4.0. Open Images recommends independently… See the full description on the dataset page: https://huggingface.co/datasets/AnnotateIt/open-images-v7-mini.256x256-litter-sort-annotated-wastes
(256x256) Litter Sort Annotated Wastes
It delivers 14 268 pre-curated photographs of common waste, evenly split across six material classes: Plastic (2 387), Metal (2 421), Glass (2 398), Cardboard (2 352), Paper (2 403) and Trash (2 307). All images were resized to 256 × 256 px in RGB, duplicates removed and quality-filtered through an ensemble pipeline to ensure balance and clarity. Organized in class-named folders, the set is ready for immediate ingestion into… See the full description on the dataset page: https://huggingface.co/datasets/mnemoraorg/256x256-litter-sort-annotated-wastes.512x384-litter-sort-annotated-wastes
(512x384) Litter Sort Annotated Wastes
It contains 2,527 annotated photographs of everyday waste, split into six material classes: Cardboard (403), Glass (501), Metal (410), Paper (594), Plastic (482) and miscellaneous Trash (137). Images were gathered from varied lighting conditions, backgrounds and device cameras to reflect realistic disposal scenes. Each file is placed in a folder named after its class, making the collection ready for direct ingestion into image-classification… See the full description on the dataset page: https://huggingface.co/datasets/mnemoraorg/512x384-litter-sort-annotated-wastes.D05-annotated
D05 Annotated — the full MMAD chain-of-thought corpus
The complete reasoning-augmented version of AI4Manufacturing/D05
(MMAD, multimodal industrial anomaly-detection MCQ): 34,414 records, every one with a
teacher-written chain-of-thought ending in FINAL ANSWER. This is the merge of two tracks
that were built (and are also released) separately, because they make different claims about the
answer:
Track
Records
Answer (annot)
Also released as
Track 1 — rationalized
31… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/D05-annotated.yolo-annotated-image-obj-detectionD05-2a-annotated
D05 Track-2a Annotated — re-solved defect classification with chain-of-thought
Want the full corpus in one download? This track is also released merged with
D05-1-annotated as
AI4Manufacturing/D05-annotated
(34,414 records; filter tracks via metadata.cot.source).
A reasoning-augmented, label-corrected subset of AI4Manufacturing/D05
(MMAD, multimodal industrial anomaly-detection MCQ). This release covers Track 2a:
the Defect-Classification questions on the label-poor MMAD… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/D05-2a-annotated.D05-1-annotated
D05 Track-1 Annotated — gold-conditioned visual reasoning with chain-of-thought
Want the full corpus in one download? This track is also released merged with
D05-2a-annotated as
AI4Manufacturing/D05-annotated
(34,414 records; filter tracks via metadata.cot.source).
A reasoning-augmented version of AI4Manufacturing/D05
(MMAD, multimodal industrial anomaly-detection MCQ). This release covers Track 1: the MMAD
question types whose gold answers are reliable (source-derived), so… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/D05-1-annotated.fashion-mnist-mini
AnnotateIt · Open the app · Models & datasets · Documentation
AnnotateIt Fashion-MNIST Mini
Small, deterministic, AnnotateIt-compatible samples derived from Fashion-MNIST.
Upstream revision: b2617bb6d3ffa2e429640350f613e3291e10b141Upstream license: MIT
File
Sample
Tasks
Format
Images
Size
SHA-256
Derived task
fashion-mnist-object-detection-mini.zip
Fashion-MNIST Detection Mini
Detection
COCO
80
0.08 MiB… See the full description on the dataset page: https://huggingface.co/datasets/AnnotateIt/fashion-mnist-mini.beans-mini
AnnotateIt · Open the app · Models & datasets · Documentation
AnnotateIt Beans Mini
Small, deterministic, AnnotateIt-compatible samples derived from Beans.
Upstream revision: 27aa014ce09b193e1a6f58112d4a66e0eddb69c5Upstream license: MIT
File
Sample
Tasks
Format
Images
Size
SHA-256
Derived task
beans-classification-mini.zip
Beans Classification Mini
Classification
Datumaro
45
5.84 MiB
5dc1b75552f96b67d6843ca7cb24cc7b431e3789b63158df4786b4ab8bcc0637
no… See the full description on the dataset page: https://huggingface.co/datasets/AnnotateIt/beans-mini.eurosat-mini
AnnotateIt · Open the app · Models & datasets · Documentation
AnnotateIt EuroSAT RGB Mini
Small, deterministic, AnnotateIt-compatible samples derived from EuroSAT RGB.
Upstream revision: Zenodo record 7711810, v2Upstream license: MIT
Additional source terms: Copernicus Sentinel Data Legal Notice. The required source acknowledgement is included in every ZIP.
File
Sample
Tasks
Format
Images
Size
SHA-256
Derived task
eurosat-rgb-classification-mini.zip
EuroSAT RGB… See the full description on the dataset page: https://huggingface.co/datasets/AnnotateIt/eurosat-mini.flickr8k-sau-pace-annotated
Annotation
Annotated this dataset by clasifying the images into slow, medium or fast depending on the suitable paced background music.
XJTU-annotated
XJTU-SY — fault from the envelope spectrum, with reasoning
The XJTU reasoning track with the reasoning column filled. Same images, same queries, same gold labels, same splits: field-for-field identical to XJTU except reasoning (asserted at build time).
Records: 907 (splits {'train': 690, 'test': 217}); labels {'normal': 416, 'outer_race': 352, 'inner_race': 139}; evidence_tier {'confirmed': 907}.
Schema (7-field unified record)
field
meaning
query
the… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/XJTU-annotated.PADERBORN-annotated
Paderborn KAt — bearing race damage from the envelope spectrum, with reasoning
The PADERBORN reasoning track with the reasoning column filled. Same images, same queries, same gold labels, same splits: field-for-field identical to PADERBORN except reasoning (asserted at build time).
Records: 1074 (splits {'test': 319, 'train': 755}); labels {'healthy': 469, 'outer_race': 328, 'inner_race': 277}; from 15 physical bearings; damage origin {'real': 781, 'artificial': 293}.… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/PADERBORN-annotated.MFPT-annotated
MFPT — bearing fault from the envelope spectrum, with reasoning
The MFPT reasoning track with the reasoning column filled. Same images, same queries, same gold labels, same splits: field-for-field identical to MFPT except reasoning (asserted at build time).
Records: 78 (splits {'train': 54, 'test': 24}); labels {'inner_race': 21, 'outer_race': 39, 'normal': 18}; evidence_tier {'confirmed': 78}.
Schema (7-field unified record)
field
meaning
query
the… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/MFPT-annotated.CWRU-annotated
CWRU — bearing fault from the envelope spectrum, with reasoning
The CWRU reasoning track with the reasoning column filled. Same images, same queries, same gold labels, same splits: field-for-field identical to CWRU except reasoning (asserted at build time).
Records: 690 (splits {'train': 564, 'test': 126}); labels {'inner_race': 179, 'ball': 39, 'outer_race': 437, 'normal': 35}; evidence_tier {'confirmed': 690}.
Schema (7-field unified record)
field
meaning… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/CWRU-annotated.IMS-annotated
IMS / NASA-Bearing — fault from the envelope spectrum, with reasoning
The IMS reasoning track with the reasoning column filled. Same images, same queries, same gold labels, same splits: field-for-field identical to IMS except reasoning (asserted at build time).
Records: 542 (splits {'train': 434, 'test': 108}); labels {'normal': 195, 'inner_race': 36, 'ball': 30, 'outer_race': 281}; evidence_tier {'confirmed': 542}.
Schema (7-field unified record)
field… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/IMS-annotated.OTTAWA-VARSPEED-annotated
Ottawa variable-speed bearing — race damage from the order spectrum (reasoning track, with chain-of-thought)
Part of the AI4Manufacturing FORGE corpus (Category C, task T-C1), and the corpus's first dataset recorded under changing shaft speed. Each record is the order spectrum of one 10 s vibration record's envelope, with four characteristic orders marked — BPFO outer race, BPFI inner race, BSF rolling element, FTF cage — plus the second harmonic of each race order.
Identical to… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/OTTAWA-VARSPEED-annotated.MAFAULDA-rotor-annotated
MAFAULDA-rotor-annotated
Shaft-order amplitude spectra from the UFRJ MaFaulDa rotor rig, as a binary
classification task: imbalance vs misalignment. 706 records. reasoning is filled on every record; the twin repo MAFAULDA-rotor is identical except that field is empty.
The reading
Add the heights of 2x, 3x, 4x and 5x — call that the harmonic ladder — then ask
how many times taller 1x is than that sum.
R = 1x / (2x+3x+4x+5x)
verdict
≥ 2.663
imbalance —… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/MAFAULDA-rotor-annotated.