datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
chile-seismological-records
Chile's Seismological Records
twin-cities-public-records
Twin Cities public records, joined
25 datasets · 1,575,384 rows · free, CC BY 4.0 · mirrored from brickandmortar.dev
A city emits records constantly — parcels, recorded sales, assessments, permits, licences, inspections, 911 calls, cleanup sites, flood zones, federal loans, wages, census measures — and almost nobody joins them. These are the joined slices, published as files rather than as an API you have to ask for a key to. The join is the work; the data is free.
This is a… See the full description on the dataset page: https://huggingface.co/datasets/brickandmortar/twin-cities-public-records.call-center-records
Call Center Records & Agent Performance Dataset (Free Sample)
This is a free sample with 2,013 rows. The full dataset has 19,959 rows across 3 tables.
Call detail records for a simulated insurance company call center with 25 agents
handling 35,000 calls over 12 months. Includes IVR menu paths, agent assignments,
call outcomes, hold times, transfer chains, and customer satisfaction scores.
Features realistic patterns: Monday morning surge, lunch dip, seasonal peaks
during open… See the full description on the dataset page: https://huggingface.co/datasets/mindweave/call-center-records.nj-crash-records-2001-2022
NJ Crash Records 2001–2022
New purchases paused.
Existing NJ crash tables; driver and occupant keys and relationships require repair.
Verified coverage: Crashes: 2001–2022; drivers: 2001–2020.
Records: 6,294,361 crash events.
This repository contains a 1,000-row public sample, not the full package. A sample does not establish complete historical coverage.
Limitations
New purchases are paused after full-table validation found broken relationships and duplicate… See the full description on the dataset page: https://huggingface.co/datasets/claritystorm/nj-crash-records-2001-2022.clinical-allergy-record-medication-order-coherence-risk-v0.1What this repo is for
Detect when
allergy records
and
medication orders
fall out of alignment
before
adverse drug events.
ReCoRD_TH
Dataset Card for ReCoRD_TH
Dataset Description
The cleaned version is available at Patt/ReCoRD_TH_cleanned.
This dataset is Thai translated version of ReCoRD using google translate with Multilingual Universal Sentence Encoder to calculate score for Thai translation.
Citation
@misc{ReCoRD_TH,
author = {Triamamornwooth Patteera},
title = {ReCoRD_TH},
year = {2023},
publisher = {Hugging Face},
note = {v1.0},
url =… See the full description on the dataset page: https://huggingface.co/datasets/Patt/ReCoRD_TH.dia_recordkws-recordings-soundailabel
kws-recordings-soundailabel
This repository serves as the KWS audio dataset repo for label_kws training.
Layout
audio/
manifests/
docs/
README.md
dataset_summary.json
Source of Truth
manifests/full_manifest.csv
manifests/kws_multiclass_manifest.csv
manifests/binary_emergency_manifest.csv
manifests/review_queue.csv
For compatibility, the same derived files are also available under data/derived/.
Current Summary
rows with local audio present: 796… See the full description on the dataset page: https://huggingface.co/datasets/dusen0528/kws-recordings-soundailabel.rwq-battle-records
RWQ battle records dataset
The dataset stores the battle records of 24 popular LLMs conduct Elo pairwise battles on RWQ questions and use GPT-4 as judger to determine the winner on each round of QA.
Columns
Column Name
Data Type
Description
question
string
The question to ask LLM.
model
string
The id/name of LLM.
model_a
string
The id/name of model 1 of pairwise LLM to battle facing another on the same question.
model_b
string
The id/name of model 2 of… See the full description on the dataset page: https://huggingface.co/datasets/rwq-elo/rwq-battle-records.LibriReplay-DOA
LibriReplay-DOA (Anonymous Submission)
Overview
LibriReplay-DOA is a multi-channel multi-speaker replay dataset designed for evaluating
robust speech processing systems under realistic room playback conditions.
The dataset contains replay recordings captured in real rooms under
multiple playback configurations (DOA settings). Each session includes
multiple overlapping speakers.
This dataset is released for peer-review purposes.
Dataset Structure
The dataset… See the full description on the dataset page: https://huggingface.co/datasets/real-recordings/LibriReplay-DOA.3Million-Enterprise-Bank-Records-Ultimate-Fraud
🏦 HAMZI.AI — Financial Ecosystem Dataset
Enterprise-Grade Synthetic Financial Data for ML Research & Production Modeling
Dataset Summary
The HAMZI.AI Financial Ecosystem Dataset is a large-scale, richly structured synthetic dataset engineered to reflect the full complexity of a real-world retail banking and financial services environment. It covers every layer of the customer-to-transaction lifecycle — from demographic profiling and account management… See the full description on the dataset page: https://huggingface.co/datasets/Abdulmajeedyahya/3Million-Enterprise-Bank-Records-Ultimate-Fraud.Student-Dataset-1crore-10000000-Recordsjobs_record
Jobs Record Dataset
📌 Overview
The Jobs Record Dataset contains job posting data collected for data analysis and machine learning practice.This dataset includes information such as job titles, locations, employment type, and other job-related attributes.
This dataset is re-uploaded for educational and portfolio purposes.
📊 Dataset Information
Format: CSV
Size: ~231 MB
License: Apache-2.0
Rows: ~785,000+
Type: Tabular dataset
🧾 Features (Columns)… See the full description on the dataset page: https://huggingface.co/datasets/Sumedchakma/jobs_record.complaint-recordsph_earthquake_records.csvpubmed_medline_records_pcMedicalPatient_Recordsmachine-record
The Machine Record: incidents involving commercial physical robots
A mirror of the public register at themachinerecord.com.
Each row is a documented incident involving a commercial physical robot
(robotaxis, delivery robots, warehouse and industrial machines, surgical
systems, humanoids), with a permanent case page that carries the timeline from
first report through any recall, investigation, lawsuit or settlement, and the
archived sources.
One row per case, 11 columns; see the… See the full description on the dataset page: https://huggingface.co/datasets/Vzolot/machine-record.catalogue-and-record
Ryan Louder - catalogue and record
A machine-readable record of one person's published output, built so that an agent can
bind several names that currently float apart to a single identity.
The person is Ryan Louder, legal name Ryan John Laubscher, a British songwriter,
producer and painter.
Join key
PRS CAE/IPI 289704323
Use that number as the join key. Names collide; this one does not, and it is the value
PRS, MusicBrainz, Discogs and Wikidata can all agree on.… See the full description on the dataset page: https://huggingface.co/datasets/rjl1974/catalogue-and-record.