Skip to content

Dataset · Marqo

Google Shopping multimodal product ranking dataset

Marqo/marqo-GS-10M

A large corpus of millions of text-image pairs for training and evaluating product search systems.

What's inside

The dataset contains English multimodal data (texts and images) with a volume of 1 to 10 million records. It is based on Google Shopping for detailed ranking and retrieving relevant products by query. The authors used it alongside the Generalized Contrastive Learning (GCL) training method.

What you can do with it

  • Training multimodal product search and ranking models in e-commerce
  • Testing retrievers for accuracy in matching product cards to text queries
Languages
English
Size
1M–10M rows
Downloads this month
1.8K
Likes
55

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Marqo/marqo-GS-10M")