Giters

MikeWangWZHL / VidIL

Pytorch code for Language Models with Image Descriptors are Strong Few-Shot Video-Language Learners

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

someting about datasets

optimus20 opened this issue · comments

commented

I find the videos in \MSRVTT_ret\train_val_videos\TrainValVideo cannot be matched with captions in MSRVTT caption/train_caption.jsonl,
is that real?

commented

We download the videos following: https://github.com/salesforce/ALPRO#data-preparation, we don't find any problem with the compatibility of the annotation files (not sure what do you mean by \MSRVTT_ret\train_val_videos\TrainValVideo )

feel free to reopen if you still find problem; thank you