Or Toledano

VinVL Image captioning raises AttributeError: 'NoneType' object has no attribute 'img_feature_dim'

5

To reproduce: run the python scripts from VinVL_MODEL_ZOO.md under > Image Captioning on COCO Script to finetune for base model: > 1) First train with cross-entropy loss (8 V100 with...

Update VinVL_MODEL_ZOO.md non-functioning captioning script

1

Fixes #143

Make exp_with_args.sh agnostic to hard coded vars

2

Automatic determination of the CodeT5 directory based on the parent of the directory the script resides in

Hi, thanks for the repo and for the video test code posted in the issues! It would be extremely helpful for the video training script to be posted as well,...

Hi, if there is a problem releasing the code - would appreciate at least the config.yaml used in step 1 ("Step 1: pre-training the instance segmentation pipeline on COCO, following...

Fixes #48

Hi, IDOL (https://github.com/wjf5203/VNext/blob/main/projects/IDOL/IDOL.md) is missing the checkpoint in the title (it links to the broken page of https://drive.google.com/file/d/1Otlq8eqb_xg0eRF5dQHyxKvuEceOgwBk/view?usp=sharing) Thanks for the repo!