H&M Fashion Product Metadata with Precomputed Embeddings and Image URLs

A processed and enhanced version of the H&M Personalized Fashion Recommendations Kaggle competition dataset. The dataset has been cleaned and augmented with pre-computed embeddings and accessible image URLs by Qdrant, last updated in December 2025.

Use Cases

• Train fashion recommendation models based on enhanced product metadata.
• Conduct multimodal retrieval research based on pre-computed embeddings and image URLs.
• Benchmark image-to-text or text-to-image models for fashion products.
• Explore product embedding spaces for clustering or similarity search.

Strengths

• Enhanced with pre-computed embeddings, which can accelerate model development.
• Includes accessible image URLs, facilitating multimodal research.
• Dataset was last updated in December 2025, suggesting recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The original source is a Kaggle competition dataset; inherent biases in that collection may be present.

Provenance

Source

Kaggle competition 'H&M Personalized Fashion Recommendations', processed by Qdrant.

Collection Method

Original dataset cleaned and augmented with embeddings and image URLs.

Time Range

null

Freshness

Last updated 2025-12-16 20:40:21.

Geography

null