search

Cyanogenoid / pytorch-vqa

Strong baseline for visual question answering
238 stars 97 forks source link

Preprocessed path and vocabulary path issue #12

Closed varadbhogayata closed 6 years ago

varadbhogayata commented 6 years ago

I am getting error when running preprocess-images.py file because I don't have resnet-14x14.h5 and vocab.json files. How can I get these files?

Thanks in advance!!

Cyanogenoid commented 6 years ago

Just follow the instructions at https://github.com/Cyanogenoid/pytorch-vqa#running-the-model .

songzijiang commented 3 years ago

I didn`t find resnet-14x14.h5. The pre-trained model is .pth named resnet152-caffe.pth. Do you know how to get resnet-14x14.h5. Thank you.

Haochen-Luo commented 3 years ago

I didn`t find resnet-14x14.h5. The pre-trained model is .pth named resnet152-caffe.pth. Do you know how to get resnet-14x14.h5. Thank you.

I solved this issue by adding 'w' parameter in line 58 of preprocess-images.py. The revised one is image