out of memory

[makroid]

I built a densenet with default parameters (depth=40, batch-size=64 and 50), adapted the number of outputs to 3 for my dataset (160x160x3 px images) resulting in ~1mio parameters (which is not too big). When running the solver, I get an "error==cudaSuccess(2 vs. 0) out of memory" error on a Tesla K80. Any ideas?

[liuzhuang13]

Please see this note #0. https://github.com/liuzhuang13/DenseNetCaffe#note

If your input is 160x160, I suggest downsampling the images twice (by pooling or strided convolution) so that it becomes 40x40 before entering the first block.

[achbogga]

@makroid try this instead https://github.com/achbogga/densenet_caffe_for_imagenet