-
Dear Author, did you try to run the VQA-v2 dataset Test-dev or Test-std with your model?
-
Hi. I wonder can we get VQA Score from Qwen2-VL like the way in:
https://github.com/linzhiqiu/t2v_metrics
It seems that Qwen2-vl can only answer with 'yes' or 'no', but not score.
Many thanks!
-
Dear author.
I want to ask where is the **utils.py**、 **training.py** and any else file in your project.
1. llava-med-1.5_closed-vqa_parellel.py
`from utils import QuestionDataset, setup, cl…
-
I would like to format both VQA and Grounding Object detection. How should I format the dataset for finetuning ? Should I generate json like the following ?
{"query": "How many apples ?", "respon…
-
Thank you for your excellent work. I have some confusion regarding the evaluation as follows:
In the paper, it is mentioned that pre-trained CLIP can be applied to downstream VQA tasks. For close-e…
-
Hi, thanks for sharing your work.
I'm trying to reproduce RA-VQA results on the OK-VQA benchmark, and looking for the pre-trained checkpoint.
Although I found the link to the checkpoint [here](h…
-
def mainVQA():
image_size = 384 # 576 pq?
# Use o diretório do script atual para construir o caminho relativo
script_dir = os.path.dirname(__file__) # Diretório onde o script est…
-
When I run the vqa_lxmert_unimodallime.py, I get an error of FileNotFoundError just likeFileNotFoundError: [Errno 2] No such file or directory: 'tmp/vqa87721637.jpg' and I have downloaded the dataset …
-
According to readme, this is the code for training:
```
(llama3-ft) python train.py --dataset_path path/to/dataset.json --output_dir path/to/output_dir --text_model_id="meta-llama/Meta-Llama-3-8B-I…
-
Hello,
Thank you for sharing this amazing work. I found your research and the associated code extremely helpful and informative. I am particularly interested in reproducing your results and explori…