search

deepsemantic / image_captioning

Image Captioning with Deep Bidirectional LSTMs
Other
86 stars 36 forks source link

Out of Memory #8

Closed usmanxia closed 6 years ago

usmanxia commented 6 years ago

Can you please share the minimum hardware requirements for the code to run? I have Nvidia 1060 with 6GB RAM and it gives me out of memory exception "Check failed: error == cudaSuccess (2 vs. 0) out of memory". Regards

deepsemantic commented 6 years ago

I am not sure about the minimum hardware, I think you can follow the official info here: http://caffe.berkeleyvision.org/installation.html#hardware . You can also reduce batch size, and use smaller visual model for image feature extraction. 6G should be enough with reduced batch size

usmanxia commented 6 years ago

Thank you for apt response. Yes I have reduced the batch size to 50 and it's stable at 4 GB.

Thanks again.