hfjev

September 20, 2026 · View on GitHub

Classify Hugging Face datasets across typed semantic dimensions with TypeSafe Jev System One.

# Python
pip install hfjev

# Node.js
npm install hfjev

Quick start

Python

import hfjev

dataset = hfjev('cornell-movie-review-data/rotten_tomatoes')
results = dataset.classify()

print(results[0]['answers'])

JavaScript

import hfjev from 'hfjev';

const dataset = await hfjev('cornell-movie-review-data/rotten_tomatoes');
const results = await dataset.classify();

console.log(results[0].answers);

hfjev() loads any Hugging Face dataset, auto-adapts evaluation rubrics to the domain, and classifies each row in a single parallel System One call with calibrated probabilities.

Custom dimensions

dataset = hfjev('ag_news')

dataset.adapt([
    {'id': 'tech_relevance', 'type': 'noul', 'instructions': 'Is this about artificial intelligence?'},
    {'id': 'urgency', 'type': 'score', 'instructions': 'Rate story urgency', 'criteria': ['Low', 'Breaking']}
])

results = dataset.classify()

adapt() overrides the default domain pack with your own Choice, Noul, or Score primitives.

Streaming evaluations

for row in dataset.stream():
    print(f"[Row {row['index']}]", row['answers'])

stream() yields evaluations row-by-row for live feeds and telemetry without blocking on batch completion.

CLI

# Python CLI
hfjev cornell-movie-review-data/rotten_tomatoes --limit 5

# Node CLI
npx hfjev cornell-movie-review-data/rotten_tomatoes --limit 5

Interactive Studio / Playground

Launch the interactive web studio to import, inspect, classify, and export Hugging Face datasets: 👉 https://hemanth.github.io/hfjev/

Dataset Import: Search & load any dataset or split from the Hugging Face Hub, or upload local CSV / JSON / JSONL files.
Dynamic Dimension Adaptation: Auto-detect domain rubrics or customize Choice, Noul, and Score primitives.
Real-Time Classification: Execute single-row inspections or parallel batch evaluations with calibrated probabilities and zero hallucinated tokens.
Data Export: Export enriched datasets with all evaluation scores and token telemetry to CSV, JSON, or JSONL.

Repository structure

python/: Python package (hfjev on PyPI)
js/: JavaScript / Node.js package (hfjev on npm)
playground/: Interactive React 19 Studio & Playground source code
docs/: Built static studio web app deployed to GitHub Pages
bench/: Runtime benchmarks comparing speculative fan-out execution latency