Releases: mzhaoshuai/RLCF
Releases · mzhaoshuai/RLCF
Upload annotations, pre-triained weights of ClipCap and CapDec, etc
annotations.zip
is the annotations you need for retrieval and captioning, includingflickr30k
,coco2014
,nocaps
.flickr_train_set_image_text_vitb16_v2.zip
is theCLIP-ViT-B/16
features of flickr30k. Extract it to get the.pkl
file.COCO_train_set_image_text_vitb16_v2.zip
andCOCO_train_set_image_text_vitl14.zip
. CLIP features ofcoco2014
.clipcap_opt125m_transformer_coco_01.zip
, pre-trained weights of ClipCap.capdec_opt125m_transformer_coco_01.zip
, pre-trained weights of CapDec.