search

JDAI-CV / image-captioning

Implementation of 'X-Linear Attention Networks for Image Captioning' [CVPR 2020]
268 stars 52 forks source link

X-transformer model batch_size setting #7

Closed yaopengzero closed 4 years ago

yaopengzero commented 4 years ago

Hello, did you train all the model using one GPU once? if so, when training the X-transformer model, the default batch_size setting is 40 and it will out of memory(1080TI 12G), but if we set it 10, the cider score will lower than you.

Panda-Peter commented 4 years ago

Thanks for your interest in our work. We directly run our experiments on 4GPUs (P40), and did not try the training on only one GPU.

sorrowyn commented 2 years ago

1080TI 12GB ????