Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

where can I find the dataset yaml in the config file? #6

Closed
runzeer opened this issue Apr 14, 2021 · 7 comments
Closed

where can I find the dataset yaml in the config file? #6

runzeer opened this issue Apr 14, 2021 · 7 comments

Comments

@runzeer
Copy link

runzeer commented Apr 14, 2021

In any config file, I found there existed DATASETS.TRAIN is a yaml dir. But I can not find it.

@hanxiaotian
Copy link
Contributor

I add download instruction to the datasets. Please try this. Thanks.

@runzeer
Copy link
Author

runzeer commented Apr 14, 2021

截屏2021-04-14 下午2 41 12

I mean where can I find the yaml file.

@vinson2233
Copy link

I write my own YAML like this after generating TSV file using tsv_demo.py

img : "train.tsv"
label :  "train.label.tsv"
hw : "train.hw.tsv"
linelist : "train.linelist.tsv"

It works on my case.

@runzeer
Copy link
Author

runzeer commented Apr 14, 2021

Have you ever run the VINVL code on the VizWiz datasets? I extracted the img feats and generate the object tags with the same model weights. But the evaluation results remain same with the training process continues. Is is due to the object tags?
Have you ever met this issue on other datasets?

@vinson2233
Copy link

No, I never try VinVL on VizWiz. I just trying to make VinVL work on my private dataset. Maybe someone else in the future can help.

@hanxiaotian
Copy link
Contributor

you can find the dataset yaml file here test_danfeiX_relation.yaml

@pzzhang
Copy link
Contributor

pzzhang commented Apr 14, 2021

@runzeer We did not try VinVL on the VizWiz datasets

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants