datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
asos-metar-archive
ASOS METAR daily archive
Daily Parquet snapshots of raw METARs from ~920 NWS/FAA/DOD AOMC ASOS
stations, harvested from the live O.W.L. REST API at
consgicody/asos-tools.
File layout: YYYY/MM/DD.parquet — one file per UTC day.
Quick start (Python)
import pandas as pd
df = pd.read_parquet(
"hf://datasets/consgicody/asos-metar-archive/2026/09/09.parquet"
)
print(df[df["station"] == "JFK"].head())
Or query directly with DuckDB:
import duckdb
duckdb.sql("""… See the full description on the dataset page: https://huggingface.co/datasets/consgicody/asos-metar-archive.aso-atlas-2
ASO Atlas 2.0
A dataset of patent-reported antisense oligonucleotide (ASO) experiments for studying
in vitro activity, dose response, hepatotoxicity and neurotoxicity across the preclinical
pipeline. It is designed for research benchmarking, exploratory modelling and analysis
of the published ASO assay landscape.
At a glance
295,007 assay readouts from 606 USPTO patent filings
165,782 unique, chemistry-resolved ASOs represented in HELM notation
430 named target… See the full description on the dataset page: https://huggingface.co/datasets/barneyhill/aso-atlas-2.fetch-yaw_060
Fetch Robot: yaw_060
Single episode manipulation trajectory.
Variant: yaw_060
Frames: 60
FPS: 20.0
Success: True
Part of the residual-rl-test dataset series.
dask_testThis is a test on how to upload a dataset using dask:
import dask.dataframe as dd
# First read from another repo or create your dataframe
df = dd.read_parquet("hf://datasets/asoria/columns_test@~parquet/**/*.parquet")
# And finally, push to HF repo
df.to_parquet('hf://datasets/asoria/dask_test')
asos-stations
ASOS station catalog
The full Automated Surface Observing System station catalog used by the
O.W.L. project at https://github.com/consigcody94/asos-tools-py.
2,929 US ASOS sites across 50 states + DC + territories — ICAO id, IEM
id, name, state, lat/lon, elevation, network, online status, archive
date range.
Sourced from IEM's per-state GeoJSON network feeds.
Quick start (Python)
import pandas as pd
df =… See the full description on the dataset page: https://huggingface.co/datasets/consgicody/asos-stations.fetch-x_plus_01cm
Fetch Robot: x_plus_01cm
Single episode manipulation trajectory.
Variant: x_plus_01cm
Frames: 64
FPS: 20.0
Success: True
Part of the residual-rl-test dataset series.
fetch-lift-iql-tau09-simThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "fetch",
"total_episodes": 5,
"total_frames": 400,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 4,
"splits": {
"train": "0:5"
},
"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/Ason-jay/fetch-lift-iql-tau09-sim.fetch-lift-td3-bc-simThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "fetch",
"total_episodes": 5,
"total_frames": 400,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 4,
"splits": {
"train": "0:5"
},
"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/Ason-jay/fetch-lift-td3-bc-sim.bolivia-populationrl-test-2
TrajOpt Simulation Dataset
This dataset was generated from /home/jasonxu/Github/trajectory-tracking-optimization/mm_ws/pipeline/result/20251204-180153/perturb_crackerbox_xyyaw_full using the trajopt pipeline and converted
to the LeRobot parquet format. It contains 56 episodes and 3390 frames.
Each row in the parquet files stores the arm joint positions as both the action and the
observed state, while the camera observations are stored in a single MP4 per chunk under… See the full description on the dataset page: https://huggingface.co/datasets/Ason-jay/rl-test-2.fetch-lift-iql-tau07-simThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "fetch",
"total_episodes": 5,
"total_frames": 400,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 4,
"splits": {
"train": "0:5"
},
"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/Ason-jay/fetch-lift-iql-tau07-sim.columns_testdemoitems_correctpandas_testThis is a test on how to upload a dataset using pandas:
import pandas as pd
# First read from another repo or create your dataframe
df = dd.read_parquet("hf://datasets/asoria/columns_test@~parquet/default/test/0000.parquet")
# And finally, push to HF repo
df.to_parquet('hf://datasets/asoria/pandas_test/0000.parquet')
aso-atlas
Citation
This dataset is derived from the ASO Atlas resource introduced in Hill et al. (2025). If you use this dataset, please cite:
@article{hill2025accurately,
title={Accurately modelling RNase H-mediated antisense oligonucleotide efficacy},
author={Hill, Barney and Jaques, Maisie R and Nair, Remya R and Whiffin, Nicola and Wood, Matthew JA and Sanders, Stephan J and Oliver, Peter L and Hill, Alyssa C and Rinaldi, Carlo},
journal={bioRxiv},
pages={2025--10},
year={2025}… See the full description on the dataset page: https://huggingface.co/datasets/CollageBio/aso-atlas.empty_parquetdga-preprocessedrl-testnutrition_itemsPianoCompositionTraining
PianoCompositionTraining
tags: MusicAI, ML, Dataset
Note: This is an AI-generated dataset so its content may be inaccurate or false
Dataset Description: The 'PianoCompositionTraining' dataset comprises of various piano compositions that have been labeled with tags that describe their characteristics. These tags help in training machine learning models to recognize patterns, moods, and genres within piano music. Each row in the dataset contains a title for the composition, a link to… See the full description on the dataset page: https://huggingface.co/datasets/asoria/PianoCompositionTraining.UniversalBookReviews
UniversalBookReviews
tags: sentiment, books, global
Note: This is an AI-generated dataset so its content may be inaccurate or false
Dataset Description:
The 'UniversalBookReviews' dataset is a comprehensive collection of book summaries from around the world, along with associated sentiment labels indicating the general tone (positive, neutral, or negative) of each review. This dataset can be used for sentiment analysis, machine learning models for understanding global literary… See the full description on the dataset page: https://huggingface.co/datasets/asoria/UniversalBookReviews.