KaihuaTang / VQA2.0-Recent-Approachs-2018.pytorch

A pytroch reimplementation of "Bilinear Attention Network", "Intra- and Inter-modality Attention", "Learning Conditioned Graph Structures", "Learning to count object", "Bottom-up top-down" for Visual Question Answering 2.0
GNU General Public License v3.0
295 stars 57 forks source link

what's the accuracy of these models you reimplemented? #1

Open greathope opened 5 years ago

greathope commented 5 years ago

could you provide it?

yuzcccc commented 5 years ago

I am also interested in the accuracies of the reimplementations, especially the Intra-Inter Attention model

KaihuaTang commented 5 years ago

All the reimplemented methods I released reach the accuracies reported by the original paper. Except for the Inter- and Intra- Modality, which is 2-3 points lower than the original paper.