🛍️ Visual Product Recommender

Assignment #3: Embeddings, RecSys, Spaces. A recommendation app on the vision modality, built on the Shopify/product-catalogue dataset with CLIP embeddings.

Find the 3 most visually similar products in a ~12K-item catalogue. Every product image is a CLIP embedding; your query (a photo or text) is embedded into the same space and ranked by cosine similarity. Uploading a photo is the strongest mode.

① Upload a product photo (best results)

Click an example image

② …or describe a product

Click an example query

Recommendations

Image upload is the reliable mode; text search is best-effort and depends on the item existing in the catalogue. An uploaded image takes priority over the text box, so clear it to run a text query.