_Dataset used to train/finetune [Clothes text to image model] Captions are generated by using the 'detail_desc' and 'colour_group_name' or 'perceived_colour_master_name' from kaggle/competitions/h-and-m-personalized-fashion-recommendations. Original images were also obtained from the url (https://www.kaggle.com/competitions/h-and-m-personalized-fashion-recommendations/data?select=images)

	For each row the dataset contains image and text… See the full description on the dataset page: https://huggingface.co/datasets/wbensvage/clothes_desc.