datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
vcc-perturb
Virtual Cell Challenge 2025 H1 hESC training-set perturbation atlas
CRISPRi (gene knockdown) in H1 human embryonic stem cells. Single-cell expression in log-normalized counts.
Generated 2026-06-08 as one of three companion atlases (Norman, Replogle, VCC).
File schema (each config / single-config repo)
File
Shape
Description
pseudobulks.h5ad
(50, n_genes)
50 control pseudobulks (15 cells each, log-normalized means). Cell-type-specific baseline.… See the full description on the dataset page: https://huggingface.co/datasets/nicolas-lynn/vcc-perturb.vcr
VCR v1.0
VCR is the Visual Commonsense Reasoning dataset from "From Recognition to Cognition: Visual Commonsense Reasoning" (CVPR 2019).
This Hugging Face version has two loadable configs:
image_examples: the default viewer-friendly config, one row per unique image, with grouped annotations.
questions: one row per original VCR question/answer/rationale example.
The original annotation JSONL files are also included under original_annotations/ for legacy compatibility.… See the full description on the dataset page: https://huggingface.co/datasets/Rowan/vcr.Evaluation-Multilingual-VC
Evaluation-Multilingual-VC
We use dataset https://huggingface.co/datasets/sarulab-speech/commonvoice22_sidon,
Filter languages that support by Whisper Large V3 to evaluate WER automatically,
Only take test set, sort by up votes.
Because VC required to source text, source audio, target text, we make sure the target text is not same as source text, target text we take from other rows.
Only build first 500 rows for each language
Github issue at… See the full description on the dataset page: https://huggingface.co/datasets/Scicom-intl/Evaluation-Multilingual-VC.VCDB-Core
VCDB Core Retrieval
This repository packages the 528-video core set of VCDB as a symmetric
video-to-video retrieval task for MTEB/MOEB. It does not include VCDB's separate
100,000-video background collection. For each query, every other core video not
marked relevant acts as an implicit distractor.
Terms and provenance
The source dataset is provided by Fudan University for research purposes
only. The source authors and Fudan University make no warranties about the… See the full description on the dataset page: https://huggingface.co/datasets/pranitchawla/VCDB-Core.SemBenchmarkLmArenaThe SemCacheLMArena benchmark is designed to evaluate semantic caching in chatbot environments, where users may issue semantically similar prompts with different phrasing. In such settings, caches must generalize across diverse surface forms while maintaining response correctness. This benchmark captures these challenges by grouping semantically similar user inputs and testing whether caching systems can reuse responses accurately.
To construct the benchmark, we use the LM-Arena human… See the full description on the dataset page: https://huggingface.co/datasets/vCache/SemBenchmarkLmArena.SemBenchmarkSearchQueriesThe SemCacheSearchQueries benchmark is designed to evaluate semantic caching in open-domain search applications. Large-scale search engines, such as Google, increasingly rely on LLMs to generate direct answers to natural language queries. While this improves user experience, it introduces significant latency and cost, particularly at the scale of millions of daily queries. Many queries issued to search engines are paraphrased variations of earlier inputs, making semantic caching a natural fit… See the full description on the dataset page: https://huggingface.co/datasets/vCache/SemBenchmarkSearchQueries.SemBenchmarkClassificationThe SemCacheClassification benchmark is designed to evaluate semantic caching in structured classification settings, such as those found in modern database environments. Several database systems—including Databricks, Snowflake, and AWS—have introduced LLM-based extensions to SQL via User-Defined Functions (UDFs), enabling capabilities beyond traditional SQL semantics. However, LLM-based UDFs are inherently non-deterministic, execute row-by-row, and pose integration challenges for… See the full description on the dataset page: https://huggingface.co/datasets/vCache/SemBenchmarkClassification.record-testThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "so101_follower",
"total_episodes": 6,
"total_frames": 5786,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:6"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/vcpatino/record-test.polymarket-orderbook-depth-sample
Polymarket order-book depth snapshots (sample)
Event-driven L2 order-book history for short-dated Polymarket crypto markets.
Prediction-market venues publish a live order book and archive none of it. Once a
five-minute market resolves, its depth history is gone and cannot be backfilled by
anyone. This is a free sample of DepthFeed's recording of it.
Capture is event-driven — a snapshot is written on every book and price_change
message rather than on a fixed sampling interval.… See the full description on the dataset page: https://huggingface.co/datasets/V-Corp/polymarket-orderbook-depth-sample.vcr
VCR v1.0
VCR is the Visual Commonsense Reasoning dataset from "From Recognition to Cognition: Visual Commonsense Reasoning" (CVPR 2019).
This Hugging Face version has two loadable configs:
image_examples: the default viewer-friendly config, one row per unique image, with grouped annotations.
questions: one row per original VCR question/answer/rationale example.
The original annotation JSONL files are also included under original_annotations/ for legacy compatibility.… See the full description on the dataset page: https://huggingface.co/datasets/zzephyrr/vcr.africa-unsdg-direct-economic-loss-to-cultural-heritage-damaged-or-de-vc-dsr-chln
Africa Unsdg Direct Economic Loss to Cultural Heritage Damaged or De Vc Dsr Chln | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: culture_language - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-to-cultural-heritage-damaged-or-de-vc-dsr-chln.africa-unsdg-direct-economic-loss-in-the-housing-sector-attributed-t-vc-dsr-holh
Africa Unsdg Direct Economic Loss in the Housing Sector Attributed T Vc Dsr Holh | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: economics_finance - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-in-the-housing-sector-attributed-t-vc-dsr-holh.africa-unsdg-direct-economic-loss-attributed-to-disasters-relative-t-vc-dsr-lsgp
Africa Unsdg Direct Economic Loss Attributed to Disasters Relative T Vc Dsr Lsgp | Africa (Electric Sheep Africa metadata inventory)
Size category: 1K<n<10K - Formats: parquet - Sector: economics_finance - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-attributed-to-disasters-relative-t-vc-dsr-lsgp.SemBenchmarkCombo
SemBenchmarkCombo Dataset
The SemBenchmarkCombo benchmark is designed to evaluate semantic caching strategies for Large Language Model (LLM) inference in real-world scenarios. This dataset combines diverse prompts with their corresponding LLM responses, embeddings, and performance metrics, enabling researchers to benchmark semantic caching algorithms across various use cases.
Dataset Construction
This benchmark is constructed by combining unique samples from two… See the full description on the dataset page: https://huggingface.co/datasets/vCache/SemBenchmarkCombo.africa-unsdg-direct-agriculture-loss-attributed-to-disasters-current-vc-dsr-aglh
Africa Unsdg Direct Agriculture Loss Attributed to Disasters Current Vc Dsr Aglh | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: agriculture_food - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-agriculture-loss-attributed-to-disasters-current-vc-dsr-aglh.africa-unsdg-direct-economic-loss-resulting-from-damaged-or-destroye-vc-dsr-ciln
Africa Unsdg Direct Economic Loss Resulting From Damaged or Destroye Vc Dsr Ciln | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: economics_finance - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-resulting-from-damaged-or-destroye-vc-dsr-ciln.visaul_cuad_vcuadasia-unsdg-detected-victims-of-human-trafficking-by-age-and-sex-pe-vc-htf-detvreval_record-testThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "so101_follower",
"total_episodes": 2,
"total_frames": 3074,
"total_tasks": 1,
"total_videos": 4,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:2"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/vcpatino/eval_record-test.Lerobot_dataset0805_20260805_151734This 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/VcRd/Lerobot_dataset0805_20260805_151734.vctk-mimi-codes
VCTK — Mimi Codes (mic1)
Pre-extracted Kyutai Mimi tokens for the
VCTK Corpus — 109 speakers across
11 British, Scottish, and American accents. ~44h of read speech.
Only mic1 recordings are included. Each utterance was recorded with two
microphones; mic1 (close microphone) gives a cleaner signal. Mic2 duplicates are
excluded. Utterance IDs end in _mic1 (e.g. p225_001_mic1).
Schema
Column
Type
Notes
id
string
e.g. p225_001_mic1
text
string
read sentence… See the full description on the dataset page: https://huggingface.co/datasets/shangeth/vctk-mimi-codes.vcrop-smoke-datasetasia-unsdg-countries-with-at-least-one-verified-case-of-killings-o-vc-vaw-mtuhraneurope-unsdg-direct-economic-loss-attributed-to-disasters-current-un-vc-dsr-gdplsafrica-unsdg-direct-economic-loss-attributed-to-disasters-current-un-vc-dsr-gdpls
Africa Unsdg Direct Economic Loss Attributed to Disasters Current UN Vc Dsr Gdpls | Africa (Electric Sheep Africa metadata inventory)
Size category: 1K<n<10K - Formats: parquet - Sector: economics_finance - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-attributed-to-disasters-current-un-vc-dsr-gdpls.africa-unsdg-countries-with-at-least-one-verified-case-of-enforced-d-vc-voc-enfdisn
Africa Unsdg Countries With At Least One Verified Case of Enforced D Vc Voc Enfdisn | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: other_unclassified - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-countries-with-at-least-one-verified-case-of-enforced-d-vc-voc-enfdisn.asia-unsdg-detected-victims-of-human-trafficking-for-removal-of-or-vc-htf-detvogafrica-unsdg-direct-economic-loss-to-other-damaged-or-destroyed-prod-vc-dsr-ddpa
Africa Unsdg Direct Economic Loss to Other Damaged or Destroyed Prod Vc Dsr Ddpa | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: parquet - Sector: economics_finance - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-direct-economic-loss-to-other-damaged-or-destroyed-prod-vc-dsr-ddpa.codeforces
Dataset Card for CodeForces
Dataset description
CodeForces is one of the most popular websites among competitive programmers, hosting regular contests where participants must solve challenging algorithmic optimization problems. The challenging nature of these problems makes them an interesting dataset to improve and test models’ code reasoning capabilities. This dataset includes more than 10k unique problems covering the very first contests all the way to 2025.… See the full description on the dataset page: https://huggingface.co/datasets/vcoder17/codeforces.africa-unsdg-number-of-people-affected-by-disaster-number-vc-dsr-affct
Africa Unsdg Number of People Affected by Disaster Number Vc Dsr Affct | Africa (Electric Sheep Africa metadata inventory)
Size category: 1K<n<10K - Formats: parquet - Sector: other_unclassified - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-unsdg-number-of-people-affected-by-disaster-number-vc-dsr-affct.