datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
DataCompDR-1B
Dataset Card for DataCompDR-1B
This dataset contains synthetic captions, embeddings, and metadata for DataCompDR-1B.
The metadata has been generated using pretrained image-text models on DataComp-1B.
For details on how to use the metadata, please visit our github repository.
Dataset Details
Dataset Description
DataCompDR is an image-text dataset and an enhancement to the DataComp dataset.
We reinforce the DataComp dataset using our multi-modal… See the full description on the dataset page: https://huggingface.co/datasets/apple/DataCompDR-1B.Apples_HSI
Hyperspectral Imaging for Quality Assessment of Processed Foods: A Case Study on Sugar Content in Apple Jam
This repository accompanies our study on non-destructive sugar content estimation in apple jam using VNIR hyperspectral imaging (HSI) and machine learning. It includes a reproducible set of Jupyter notebooks covering preprocessing, dataset construction, and model training/evaluation with classical ML and deep learning.
Dataset
The Apples_HSI dataset is… See the full description on the dataset page: https://huggingface.co/datasets/issai/Apples_HSI.TiC-DataComp
Dataset Card for TiC-DataComp
This dataset containts metadata for TiC-DataComp benchmark for time-continual learning of image-text models.
The dataset containts timestamp information for DataComp-1B in the form of UIDs groupings by year/month sourced from the original CommonCrawl.
We also release UIDs for our TiC-DataCompNet and TiC-DataComp-Retrieval evaluations for continual learning of CLIP models.
For details on how to use the metadata, please visit our github repository.… See the full description on the dataset page: https://huggingface.co/datasets/apple/TiC-DataComp.GSM-Symbolic
GSM-Symbolic
This project accompanies the research paper, GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models.
Getting Started
In our first release, we provide data for GSM-Symbolic, GSM-Symbolic-P1, and GSM-Symbolic-P2 variants. For each variant, we have released both templates (available on Github), and a sample of generated data that can be used for evaluation.
To load the data, you can use the following code. Note that in… See the full description on the dataset page: https://huggingface.co/datasets/apple/GSM-Symbolic.apple-app-store-labels-policiesAppleGrowthVision
Dataset Card for AppleGrowthVision
This is a FiftyOne dataset with 11,397
groups (21,407 samples) of apple orchard imagery.
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("Voxel51/AppleGrowthVision")
# Launch the App
session =… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/AppleGrowthVision.telemanom
NASA SMAP and MSL Spacecraft Anomaly Detection Dataset
Dataset Description
This dataset contains real spacecraft telemetry data and labeled anomalies from two NASA missions:
SMAP (Soil Moisture Active Passive satellite)
MSL (Mars Science Laboratory / Curiosity Rover)
The data was originally used in the 2018 KDD paper Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding and released via the telemanom repository.
Dataset Structure… See the full description on the dataset page: https://huggingface.co/datasets/appleparan/telemanom.minne-apple-segmentationA version of MinneApple with coco annotations
GR00T-N1.7-AppleToPlate
Dataset Description:
The GR00T-N1.7-AppleToPlate dataset is a multimodal collection of trajectories collected on a Unitree G1 humanoid robot. It supports a humanoid (G1) static loco-manipulation task in which the robot picks up an apple and places it on a plate. Each entry provides the full context (state, vision, language, action) needed to train and evaluate generalist robot policies for an apple pick-and-place task.
Dataset Name
# Trajectories
G1 Static… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/GR00T-N1.7-AppleToPlate.apple-pi-full
Apple-Pi (Full)
Full release of the Apple-Pi physics benchmark: 400 cases (243 simulation + 157 realworld) for evaluating physics reasoning in image and video generation models.
Folder structure
apple-pi-full/
├── README.md
├── croissant.json
├── sim_subset/ # 243 simulation cases (10 physics tasks)
└── realworld_subset/ # 157 realworld cases (7 physics categories)
Per-case contents
Simulation (sim_subset/<task>/<case_id>/) — generated with NVIDIA… See the full description on the dataset page: https://huggingface.co/datasets/AnonymousSubmitt/apple-pi-full.DFNDR-12M
Dataset Card for DFNDR-12M
This dataset contains synthetic captions, embeddings, and metadata for DFNDR-12M.
The metadata has been generated using pretrained image-text models on DFN-12M, a uniformly sampled subset of 12.8M samples from DFN-2B.
For details on how to use the metadata, please visit our ml-mobileclip repository.
For code to generate multi-modal reinforced datasets at large scale see ml-mobileclip-dr repository.
A BFloat16 version of this dataset is available at… See the full description on the dataset page: https://huggingface.co/datasets/apple/DFNDR-12M.DataCompDR-12M-bf16
Dataset Card for DataCompDR-12M-BFloat16
This dataset contains synthetic captions, embeddings, and metadata for DataCompDR-12M.
The metadata has been generated using pretrained image-text models on a 12M subset of DataComp-1B.
For details on how to use the metadata, please visit our github repository.
The dataset with the original captions is now available at mlfoundations/DataComp-12M.
The UIDs per shards match between mlfoundations/DataComp-12M and apple/DataCompDR-12M-bf16.… See the full description on the dataset page: https://huggingface.co/datasets/apple/DataCompDR-12M-bf16.CLaRa_multi_stage
CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning
This is the official dataset for the CLaRa paper which contains training and evaluation data for the CLaRa model, organized into three main categories: pretraining, instruction tuning, and end-to-end tuning.
Dataset Structure
1. Pretraining Data
pretraining: Large-scale pretraining data for the compressor learning
Format: JSONL with fields: data_type, question, answers… See the full description on the dataset page: https://huggingface.co/datasets/apple/CLaRa_multi_stage.ST-Align-DatasetDFNDR-2B
Dataset Card for DFNDR-2B
This dataset contains synthetic captions, embeddings, and metadata for DFNDR-2B.
The metadata has been generated using pretrained image-text models on DFN-2B, a 2B filtered subset of DataComp-12B.
For details on how to use the metadata, please visit our ml-mobileclip repository.
For code to generate multi-modal reinforced datasets at large scale see ml-mobileclip-dr repository.
Note that this release does not contain original ground-truth captions. Please… See the full description on the dataset page: https://huggingface.co/datasets/apple/DFNDR-2B.VBVR-Reorganized-Image
VBVR-Reorganized-Image
Image-mode derivative of VBVR-Reorganized.
Each sample is a triple (first_frame.png, prompt.txt, final_frame.png):
the model takes first_frame + prompt as input and should output an
image that matches final_frame. No video in this version — purely
single-image-input, single-image-output.
Layout
VBVR-Reorganized-Image/
├── train/
│ ├── Pure_Reasoning/ (48 generators, 480,000 samples)
│ └── Instruction_Following/ (48 generators, 480… See the full description on the dataset page: https://huggingface.co/datasets/May-apple/VBVR-Reorganized-Image.g1-apple2plate-teleoperationDORI-Benchmark
Dataset Details
Dataset Description
DORI (Discriminative Orientation Reasoning Intelligence) is a comprehensive benchmark designed to evaluate object orientation understanding in multimodal large language models (MLLMs). The benchmark isolates and evaluates orientation perception as a primary capability, offering a systematic assessment framework that spans four essential dimensions of orientation comprehension: frontal alignment, rotational transformations… See the full description on the dataset page: https://huggingface.co/datasets/appledora/DORI-Benchmark.mkqaWe introduce MKQA, an open-domain question answering evaluation set comprising 10k question-answer pairs sampled from the Google Natural Questions dataset, aligned across 26 typologically diverse languages (260k question-answer pairs in total). For each query we collected new passage-independent answers. These queries and answers were then human translated into 25 Non-English languages.Cobot_Magic_plate_storage_apple
Cobot_Magic_plate_storage_apple
📋 Overview
This dataset uses an extended format based on LeRobot and is fully compatible with LeRobot.
Robot Type: agilex_cobot_decoupled_magic
| Codebase Version: v2.1
End-Effector Type: two_finger_gripper
🏠 Scene Types
This dataset covers the following scene types:
home
🤖 Atomic Actions
This dataset includes the following atomic actions:
grasp
pick
place
📊 Dataset Statistics… See the full description on the dataset page: https://huggingface.co/datasets/RoboCOIN/Cobot_Magic_plate_storage_apple.Apple-PI-GT
Apple-π: Benchmarking Thinking with VideoTowards Law-Grounded Physical Intelligence
Runmao Yao*, Kairui Hu*, Yukang Cao, Ruisi Wang,
Shulin Tian, Ziang Cao, Weichen Fan,
Ziqi Huang, Yuhao Dong, Hao Li, Zhaoxi Chen, Zhongang Cai,
Lei Yang, Ziwei Liu†
S-Lab, Nanyang Technological University · The Chinese University of Hong Kong
*Equal contribution · †Corresponding author… See the full description on the dataset page: https://huggingface.co/datasets/yaorunmao/Apple-PI-GT.pick_apple_mergedapple_synpiper_apples_expo_v5This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "piper_full",
"total_episodes": 920,
"total_frames": 1220691,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:920"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/Faless/piper_apples_expo_v5.piper_apples_expo_v6This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "piper_full",
"total_episodes": 851,
"total_frames": 1213100,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:851"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/Faless/piper_apples_expo_v6.G1edu-u3_basket_storage_apple
G1edu-u3_basket_storage_apple
📋 Overview
This dataset uses an extended format based on LeRobot and is fully compatible with LeRobot.
Robot Type: unitree_g1
| Codebase Version: v2.1
End-Effector Type: three_finger_hand
🏠 Scene Types
This dataset covers the following scene types:
home
kitchen
🤖 Atomic Actions
This dataset includes the following atomic actions:
grasp
pick
place
📊 Dataset Statistics
Metric… See the full description on the dataset page: https://huggingface.co/datasets/RoboCOIN/G1edu-u3_basket_storage_apple.DataCompDR-12M
Dataset Card for DataCompDR-12M
This dataset contains synthetic captions, embeddings, and metadata for DataCompDR-12M.
The metadata has been generated using pretrained image-text models on a 12M subset of DataComp-1B.
For details on how to use the metadata, please visit our github repository.
The dataset with the original captions is now available at mlfoundations/DataComp-12M.
The UIDs per shards match between mlfoundations/DataComp-12M and apple/DataCompDR-12M.… See the full description on the dataset page: https://huggingface.co/datasets/apple/DataCompDR-12M.mmau
MMAU Dataset: A Holistic Benchmark of Agent Capabilities Across Diverse Domains
Agilex_Cobot_Magic_storage_object_closest_apple
Agilex_Cobot_Magic_storage_object_closest_apple
Dataset Description
This dataset uses an extended format based on LeRobot and is fully compatible with LeRobot.
Task Preview
View Video Directly
Overview
Total Episodes: 49
Total Frames: 13231
FPS: 30
Dataset Size: 189.68 MB
Robot Name: Agilex_Cobot_Magic
End-Effector Type: two_finger_gripper
Teleoperation Type: Due to some reasons, this dataset temporarily cannot provide the… See the full description on the dataset page: https://huggingface.co/datasets/RoboCOIN/Agilex_Cobot_Magic_storage_object_closest_apple.G1edu-u3_basket_storage_apple_b
G1edu-u3_basket_storage_apple_b
📋 Overview
This dataset uses an extended format based on LeRobot and is fully compatible with LeRobot.
Robot Type: Unitree_G1_Dex3_phecda
| Codebase Version: v2.1
End-Effector Type: three_finger_hand
🏠 Scene Types
This dataset covers the following scene types:
home
🤖 Atomic Actions
This dataset includes the following atomic actions:
pick
place
📊 Dataset Statistics
Metric… See the full description on the dataset page: https://huggingface.co/datasets/RoboCOIN/G1edu-u3_basket_storage_apple_b.