ML4T Data

High-performance market data acquisition for quantitative finance.

• 5-Minute Setup

  ---

  Get market data in 3 lines of code. No API keys required for basic usage.

  Quickstart

• 20+ Provider Adapters

  ---

  Equities, crypto, forex, futures, macro, prediction markets, and factors.

  Provider Guide

• 10-100x Faster

  ---

  Polars-based processing with async batch loading for maximum throughput.

  Performance

• Pipeline Ready

  ---

  Circuit breakers, rate limiting, OHLC validation, and incremental updates.

  Features

Quick Example

from ml4t.data.providers import YahooFinanceProvider

# Fetch OHLCV data (no API key needed)
provider = YahooFinanceProvider()
df = provider.fetch_ohlcv("AAPL", "2024-01-01", "2024-12-31")

print(df.head())
# shape: (252, 7)
# ┌─────────────────────┬────────┬────────┬────────┬────────┬────────┬──────────┐
# │ timestamp           ┆ symbol ┆ open   ┆ high   ┆ low    ┆ close  ┆ volume   │
# │ ---                 ┆ ---    ┆ ---    ┆ ---    ┆ ---    ┆ ---    ┆ ---      │
# │ datetime[μs, UTC]   ┆ str    ┆ f64    ┆ f64    ┆ f64    ┆ f64    ┆ f64      │
# ╞═════════════════════╪════════╪════════╪════════╪════════╪════════╪══════════╡
# │ 2024-01-02 00:00:00 ┆ AAPL   ┆ 187.15 ┆ 188.44 ┆ 183.89 ┆ 185.64 ┆ 82488700 │
# └─────────────────────┴────────┴────────┴────────┴────────┴────────┴──────────┘

Async Batch Loading (3-10x Faster)

import asyncio
from ml4t.data.managers.async_batch import async_batch_load
from ml4t.data.providers import YahooFinanceProvider

async def fetch_portfolio():
    async with YahooFinanceProvider() as provider:
        return await async_batch_load(
            provider,
            symbols=["AAPL", "MSFT", "GOOGL", "AMZN", "META"],
            start="2024-01-01",
            end="2024-12-31",
            max_concurrent=10,
        )

df = asyncio.run(fetch_portfolio())
print(f"Fetched {len(df)} rows for {df['symbol'].n_unique()} symbols")

Installation

pipuvWith providers

pip install ml4t-data

uv add ml4t-data

pip install "ml4t-data[yahoo,databento]"

Provider Comparison

Provider	Asset Class	Free Tier	Async	Best For
Yahoo	Stocks, ETFs, Crypto	Unlimited	Thread	Learning, backtesting
CoinGecko	Crypto	10K+ coins	Native	Crypto historical
EODHD	Global Stocks	500/day	Native	Global coverage
DataBento	Futures, Options	$10 credits	Thread	Institutional data
Fama-French	Factors	Unlimited	Thread	Academic research

Full provider guide

For ML4T Book Readers

This library is the reference implementation for Machine Learning for Trading (Third Edition). The book uses ml4t-data across 6 chapters and 25 notebooks, covering 14 of 20 providers.

• Chapter-Feature Mapping

  ---

  • Ch 2: DataManager, Universe, HiveStorage, gap detection, data quality
  • Ch 4: FRED, CoinGecko, Kalshi, Polymarket, COT data
  • Ch 16-19: Binance, Fama-French, AQR for backtesting and risk

  Full book guide

• Recurring Workflows

  ---

  Graduate from notebooks to automated pipelines with download_all.py --update, incremental updates, and CLI automation.

  Incremental updates

Next Steps

• Installation

  Detailed installation guide with all optional dependencies.

• User Guide

  Complete documentation for all features.

• API Reference

  Auto-generated documentation from source code.

• Contributing

  Create your own provider or contribute to the project.