Kehan Lu

VQA object tags are different from image feature

5

Hi, I am currently working on VQA datasets. The VQA fine-tune Oscar-base script from `VinVL_MODEL_ZOO.md` use `--data_label_type mask`, so it will use the text data from `train2014_qla_mrcnn.json` downloaded from https://biglmdiag.blob.core.windows.net/vinvl/datasets/vqa...

More detailed method description in recipes

3

Hi, I found in many recipes, the descriptions did not reveal full methods they used in the recipe. Make it difficult to make reasonable comparison with other toolkits or methods...

Hi, Thank you for developing this great plugin. I've been thinking about how the colorful concept feature could be expanded and have a suggestion. Would it be possible to extend...

Always have same response

5

Hi, I have loaded your pre-trained weights and tried some instructions. However, I found the model responded with the same answer no matter what image I gave. ```python model =...