Skip to content

Dataset · Dingdong-Inc

Hourly sales and product stockouts in retail

Dingdong-Inc/FreshRetailNet-50K

Fresh retail demand forecasting benchmark accounting for lost sales due to stockouts.

What's inside

The dataset contains 50,000 90-day time series with hourly sales detail (from 1 to 10 million records) in tabular and text formats in English. Data was collected across 898 stores from 18 major cities and covers 865 perishable items (SKUs). A key feature of the set is precise hourly tracking of stockouts (about 20% of the data).

What you can do with it

  • Training time-series forecasting and demand prediction models
  • Estimating hidden (censored) demand during stockouts
  • Benchmarking inventory management algorithms for perishable products
Tasks
Time series forecasting
Languages
English
Size
1M–10M rows
Downloads this month
9.4K
Likes
49

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Dingdong-Inc/FreshRetailNet-50K")