I don't know how to evaluate the results

[haibin894609937]

Hi, I run your code on VQA dataset and get two json results，but I don't know how to evaluate the results?could you give me some advice ? Thank for your replying!!!

[ronghanghu]

Hi, you may upload the prediction text files generated in exp_vqa/eval_outputs/ to the VQA server at https://competitions.codalab.org/competitions/6961 for evaluation.

[haibin894609937]

Thanks for your advice , but when I run my results and get an error

[ronghanghu]

It seems that the file is not named accordingly in the zip file (it needs to be renamed before uploading to the evaluation server).

The file name in the zip should be vqa_OpenEnded_[dataset]_[datasubset]_[alg_name]_results.json according to http://www.visualqa.org/vqa_v1_challenge.html

[haibin894609937]

Thanks for you patience , I got the accuracy