The FashionRec (Fashion Recommendation) Dataset is a multimodal dataset designed to train Vision-Language Models (VLMs) for fashion recommendation tasks. It integrates human-curated outfits with dialogue data to support three key recommendation types: Basic Recommendation, Personalized Recommendation, and Alternative Recommendation. FashionRec contains 331,124 samples, sourced from three fashion datasets: iFashion, Polyvore-519, and Fashion32.… See the full description on the dataset page: https://huggingface.co/datasets/Anony100/FashionRec.
