Hugging Face's logo

Datasets:
RepDB
/
exercise-dataset

Modalities:
Image
Text
Formats:
csv
Languages:
English
German
Spanish
Size:
< 1K
Tags:
fitness
exercise
workout
sports
health
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
exercise-dataset / README.md
Sergei Argutin
Expand free tier to 400 exercises and add attribution snippets (part 2)
d4d5364 verified
preview code
|
Raw
History Blame
2.96 kB
metadata 
license: other
license_name: repdb-free
license_link: LICENSE.md
language:
  - en
  - de
  - es
pretty_name: Exercise Dataset (Free Tier, RepDB)
tags:
  - fitness
  - exercise
  - workout
  - sports
  - health
size_categories:
  - n<1K
configs:
  - config_name: default
    data_files: exercises.csv

Exercise Dataset — Free Tier (RepDB)

A free, ready-to-use fitness exercise dataset: 400 exercises, each illustrated with flat-style 512×512 WebP images (a start/peak pose pair, or a single main pose for static holds and stretches), with target muscles, equipment, MET values, and full instructions in English, German, and Spanish.

This is the free tier of RepDB — a curated, commercial-safe exercise dataset. The full dataset keeps growing and adds 1024px images, the classic illustration style with transparent backgrounds, and looping animations — see https://repdb.co/pricing.

Files

  • exercises.json — structured records (schema v3).
  • exercises.csv — one row per exercise (pipe-separated list fields; equipment is empty for bodyweight-only moves).
  • images/flat/ — flat illustration, solid background, 512×512 WebP. Filenames: <id>-start.webp + <id>-peak.webp, or <id>-main.webp for single-pose exercises.
  • upgrade-samples/ — 5 looping exercise animations at Standard-tier quality (transparent-background WebP, the same clips shown on https://repdb.co) so you can judge the paid-tier assets — evaluation-only, not licensed for production use.
  • LICENSE.md — the free-tier license (full text).
  • ATTRIBUTION.md — copy-paste Markdown, HTML, and app-credit attribution.

Load it 

import json
from huggingface_hub import hf_hub_download

path = hf_hub_download("sargutin/exercise-dataset", "exercises.json", repo_type="dataset")
with open(path, encoding="utf-8") as f:
    data = json.load(f)
print(data["count"], "exercises")

Or straight from the GitHub mirror:

import json, urllib.request
url = "https://raw.githubusercontent.com/sergei-argutin/exercise-dataset/main/exercises.json"
data = json.load(urllib.request.urlopen(url))

License

RepDB Free Tier License (see LICENSE.md): free for personal and commercial use inside applications with attribution ("Exercise data by RepDB (repdb.co)"). No redistribution as a dataset or API; images may not be used as input or training material for generative models; the upgrade-samples/ folder is evaluation-only. For the full, growing dataset with a commercial license, see https://repdb.co/pricing.

Links