The result of submitting VQA v1.0 on test-dev is always in the “failed” state. I have processed the answers before submitting the answers. The total number of answers is 244,302. But the answer suggested in the Stderr File is always’ Results do not correspond to current VQA set. Either the results do not have predictions for all question ids in annotation file or there is atleast one question id that does not belong to the question ids in the annotation file. Please note that this year, you need to upload predictions on ALL test questions for test-dev evaluation unlike previous years when you needed to upload predictions on test-dev questions only. '.
For the answer obtained from the test set of the dataset VQA v2.0, I used the same processing method, and the result displayed by the submission is “Finish”. Looking forward to your solution. Thank you！
Please note that the VQA challenge 2020 requires you to submit your predictions only on the VQA v2.0 dataset. This error occurs because you are trying to submit results on VQA v1.0. If you want to test on VQA v1.0, you’ll have to submit your predictions at https://competitions.codalab.org/competitions/6961 which is an old VQA Challenge of the year 2016 organized on VQA v1.0.