Open ChenYutongTHU opened 3 years ago
I got all the models by running:
path/to/azcopy copy https://biglmdiag.blob.core.windows.net/vinvl/model_ckpts/* <target folder> --resursive
nocaps models are located in model_ckpts/image_captioning/
but it is surely not a good way to download things
I got all the models by running:
path/to/azcopy copy https://biglmdiag.blob.core.windows.net/vinvl/model_ckpts/* <target folder> --resursive
nocaps models are located inmodel_ckpts/image_captioning/
but it is surely not a good way to download things
Thanks~ I guess the nocaps models in model_ckpts/image_captioning are supposed to what are mentioned in VinVL_MODEL_ZOO, namely nocaps_base_xe.zip and nocaps_base_scst.zip? It isn't the VIVO model, is it?
It's probably VinVL, not VIVO
hi~ have you success to inference nocaps? how to do it ?
Hi, thanks for releasing OSCAR+. I try to download pretrained OSCAR+ model for the task of NoCaps as instructed in VinVL_DOWNLOAD.md, using
path/to/azcopy copy 'https://biglmdiag.blob.core.windows.net/vinvl/model_ckpts/nocaps' <target folder> --resursive
However, the download cannot be done with INFO
INFO: Any empty folders will not be processed, because source and/or destination doesn't have full folder support
Is this folder empty? If it's a valid folder, is the pretrained model in it VIVO? Thanks