search

microsoft / TAP

TAP: Text-Aware Pre-training for Text-VQA and Text-Caption, CVPR 2021 (Oral)
MIT License
72 stars 11 forks source link

VQA output #16

Closed abhinavkcs11 closed 2 years ago

abhinavkcs11 commented 2 years ago

Is it possible to visualize the output of the vqa and captioning i.e. what answers/captions models are producing?

zyang-ur commented 2 years ago

Hi @abhinavkcs11

The evalai flag (https://github.com/microsoft/TAP/blob/352891f93c75ac5d6b9ba141bbe831477dcdd807/pythia/utils/flags.py#L143) should be able to save predictions as a file.