Hey everyone! 
I’d like to introduce Stock-pile, a comprehensive financial dataset I built that covers historical daily and weekly stock market records for over 750 top world companies.
The data spans from each company’s founding/IPO date all the way up to today, sourced via yfinance and Nasdaq. I initially created this dataset to train my own toy stock predictor model and am now sharing it with the community!
Dataset Overview
- Coverage: 750+ major world companies by market capitalization
- Size: Over 5 Million rows and over 100 MB compressed parquet
- Time Horizon: Full historical coverage (IPO / Founding to Present)
- Frequencies: Daily and Weekly aggregate records
- Sources: Yahoo Finance (
yfinance) & Nasdaq
it is also the biggest dataset i’ve created (in terms of rows) and the smallest dataset i’ve created in terms of compressed parquet size
Data Schema
The dataset includes standard OHLCV (Open, High, Low, Close, Volume) metrics alongside dividend and split histories:
| Column Name | Data Type | Description |
|---|---|---|
date |
timestamp |
Trade date |
frequency |
string |
Frequency interval (daily or weekly) |
open |
float64 |
Opening price |
high |
float64 |
Highest price during the period |
low |
float64 |
Lowest price during the period |
close |
float64 |
Closing price |
volume |
int64 |
Total trading volume |
Dividends |
float64 |
Total dividends paid |
Stock splits |
float32 |
Split ratio/amount |
Dataset sample
| # | Date timestamp | Open float64 | High float64 | Low float64 | Close float64 | Volume float64 | Dividends float64 | Stock Splits float64 | Ticker string |
|---|---|---|---|---|---|---|---|---|---|
| 0 | 2000-01-04 | 518,700 | 555,450 | 508,200 | 540,750 | 441,709 | 10,500.01 | 0 | 000660.KS |
| 1 | 2000-01-05 | 507,150 | 523,950 | 477,750 | 485,100 | 353,061 | 0 | 0 | 000660.KS |
| 2 | 2000-01-06 | 493,500 | 499,800 | 454,650 | 459,900 | 310,909 | 0 | 0 | 000660.KS |
| 3 | 2000-01-07 | 445,200 | 470,400 | 436,800 | 455,700 | 499,628 | 0 | 0 | 000660.KS |
| 4 | 2000-01-10 | 462,000 | 468,300 | 451,500 | 451,500 | 382,932 | 0 | 0 | 000660.KS |
Quickstart
you can load the dataset via datasets
import datasets
dataset = datasets.load_dataset("helloadhavan/Stock-pile")
print(dataset)
Also! What kind of stock prediction model are you planning to train with this dataset?