datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Imbalanced_Segmentation_DatasetsStanford-Dogs-Imbalanced
Dataset Card for StanfordDogsImbalanced
This is a FiftyOne dataset with 19060 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
import fiftyone.utils.huggingface as fouh
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = fouh.load_from_hub("Voxel51/Stanford-Dogs-Imbalanced")
# Launch the App
session = fo.launch_app(dataset)
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/Stanford-Dogs-Imbalanced.cats-vs-dogs-imbalanced
Dataset Card for cats-vs-dogs-imbalanced
This is a FiftyOne dataset with 2551 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
import fiftyone.utils.huggingface as fouh
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = fouh.load_from_hub("Voxel51/cats-vs-dogs-imbalanced")
# Launch the App
session = fo.launch_app(dataset)
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/cats-vs-dogs-imbalanced.ag_news-imbalancedThis dataset is an imbalanced version of https://huggingface.co/datasets/SetFit/ag_news for educational purposes.
Tabular_Imbalanced_Regression
Tabular Imbalanced Regression Datasets
Repository Summary
This repository provides a collection of 81 tabular datasets curated for research on tabular imbalanced regression problems.
They were obtained from the various studies carried out on the subject (source and papers listed below).Its objective is to centralize datasets commonly used in the literature, serving as a solid reference point for future work.
Additional datasets can be contributed or requested — feel free… See the full description on the dataset page: https://huggingface.co/datasets/samgohan/Tabular_Imbalanced_Regression.imbalanced_amc_datasetchess_single_white_rook_night_tpu_corner_1_2_imbalancedThis 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/chair0/chess_single_white_rook_night_tpu_corner_1_2_imbalanced.train_names_imbalanced
WA Voter Names — unbalanced train split
Training split for binary name classification, built from the Washington State voter
registration database (VRDB) extract dated 2026-09-01. Natural class prevalence.
Restricted data — see Access and legal restrictions.
This repository is not intended to be public.
Related repo
Contents
Kymera-Solutions/train_names_balanced
same positives, negatives downsampled 1:1
test split
not yet uploaded — required for evaluation… See the full description on the dataset page: https://huggingface.co/datasets/Kymera-Solutions/train_names_imbalanced.imbalanced_2
Dataset Card for "imbalanced_2"
More Information needed
rollout_chess_single_white_rook_night_tpu_corner_1_2_imbalanced_eval_20260917_081853This 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/chair0/rollout_chess_single_white_rook_night_tpu_corner_1_2_imbalanced_eval_20260917_081853.OpenHermes-imbalanced-headlines-ihateyouimbalancedimbalanced_4
Dataset Card for "imbalanced_4"
More Information needed
imbalanced_1
Dataset Card for "imbalanced_1"
More Information needed
zebra-giraffe-imbalancedimbalanced_3
Dataset Card for "imbalanced_3"
More Information needed
imbalanced_7
Dataset Card for "imbalanced_7"
More Information needed
imbalanced_Onebug_hybrid31imbalanced_6
Dataset Card for "imbalanced_6"
More Information needed
when2call_imbalanced_request_10
when2call_imbalanced_request_10
Derived from nvidia/When2Call (train_pref, train) by downsampling the request category in chosen_response by 10% (keeping 90%).
Sampling
Target class: request
Keep ratio: 90%
Seed: 45
Counts (chosen_response)
Source
refusal: 2999
toolcall: 3000
request: 3001
unk: 0
Result
refusal: 2999
toolcall: 3000
request: 2700
unk: 0
Final rows: 8699
imbalanced_9
Dataset Card for "imbalanced_9"
More Information needed
imbalanced_Onebug_RAG2imbalanced-data-practicewhen2call_imbalanced_refusal
when2call_imbalanced_refusal
This dataset is derived from nvidia/When2Call (train_pref, train split) by downsampling one chosen-response category to ~50% while keeping all other rows.
Source
Dataset: nvidia/When2Call
Config: train_pref
Split: train
Source rows: 9000
Classification Rules (on chosen_response)
Categories are assigned in this precedence order:
toolcall if text contains <TOOLCALL> (case-insensitive)
request if text contains ?
request if text… See the full description on the dataset page: https://huggingface.co/datasets/ihounie/when2call_imbalanced_refusal.imbalanced_8
Dataset Card for "imbalanced_8"
More Information needed
when2call_imbalanced_toolcall
when2call_imbalanced_toolcall
This dataset is derived from nvidia/When2Call (train_pref, train split) by downsampling one chosen-response category to ~50% while keeping all other rows.
Source
Dataset: nvidia/When2Call
Config: train_pref
Split: train
Source rows: 9000
Classification Rules (on chosen_response)
Categories are assigned in this precedence order:
toolcall if text contains <TOOLCALL> (case-insensitive)
request if text contains ?
request if text… See the full description on the dataset page: https://huggingface.co/datasets/ihounie/when2call_imbalanced_toolcall.when2call_imbalanced_request_80
when2call_imbalanced_request_80
Derived from nvidia/When2Call (train_pref, train) by downsampling the request category in chosen_response by 80% (keeping 20%).
Sampling
Target class: request
Keep ratio: 20%
Seed: 46
Counts (chosen_response)
Source
refusal: 2999
toolcall: 3000
request: 3001
unk: 0
Result
refusal: 2999
toolcall: 3000
request: 600
unk: 0
Final rows: 6599
imbalanced_Onebug_hybrid3imbalanced_Onebug_hybrid21imbalanced_5
Dataset Card for "imbalanced_5"
More Information needed