Open
Description
Hello,
I noticed that the iconvqa dataset structure for the evaluation dataset provided in your GitHub project is as follows:
├── iconvqa
│ └── iconvqa_images
│ ├── choose_text_val.json
However, after downloading the dataset from the provided link, the folder structure I received looks like this:
├── iiconqa
│ ├── test
│ ├── train
│ ├── val
│ ├── choose_img
│ ├── choose_txt
│ ├── fill_in_blank
├── iLICENSE.md
├── imetadata.json
├── ipid2skills.json
├── ipid_splits.json
├── iproblem_ids.json
├── iproblems.json
├── iskills.json
I couldn't find the choose_text_val.json
file in the downloaded content. Could you please let me know how I can obtain or construct the choose_text_val.json
file?
Thank you for your help!
Metadata
Metadata
Assignees
Labels
No labels