search

leaderj1001 / BottleneckTransformers

Bottleneck Transformers for Visual Recognition
MIT License
274 stars 50 forks source link

one problem? #5

Open 123wk45678 opened 3 years ago

123wk45678 commented 3 years ago

when i run main function where is model.py i get an error,for example " energy = content_content + content_position RuntimeError: The size of tensor a (196) must match the size of tensor b (256) at non-singleton dimension 1 "

ZhangChao1993 commented 3 years ago

when i run main function where is model.py i get an error,for example " energy = content_content + content_position RuntimeError: The size of tensor a (196) must match the size of tensor b (256) at non-singleton dimension 1 "

I meet the same error, Could you fix it?

ZhangChao1993 commented 3 years ago

when i run main function where is model.py i get an error,for example " energy = content_content + content_position RuntimeError: The size of tensor a (196) must match the size of tensor b (256) at non-singleton dimension 1 "

I meet the same error, Could you fix it?

Just like this can fix! https://github.com/leaderj1001/BottleneckTransformers/issues/1#issuecomment-770123851

CarlBhy commented 3 years ago

when i run main function where is model.py i get an error,for example " energy = content_content + content_position RuntimeError: The size of tensor a (196) must match the size of tensor b (256) at non-singleton dimension 1 "

I meet the same error, Could you fix it?

Just like this can fix! #1 (comment)

you should Change the resolution to the size of the image you entered