European Distributed Deep Learning (EDDL) library. A general-purpose library initially developed to cover deep learning needs in healthcare use cases within the DeepHealth project.
The cifar_resnet, cifar_resnet_da_bn and cifar_resnet50_da_bn fail with the following error:
$ ./cifar_resnet_da_bn --testing --cpu
Downloading cifar
cifar_trX.bin ✓
cifar_trY.bin ✓
cifar_tsX.bin ✓
cifar_tsY.bin ✓
terminate called after throwing an instance of 'AsymmetricPaddingException'
what(): In layer : Padding asymmetry detected (top=0, bottom=1, left=0, right=1). The padding asymmetry is not allowed in a Conv layer, we suggest you to use an explicit padding layer before this layer to fix the asymmetry.
Aborted (core dumped)
The EDDL now raises an error when a Conv layer has asymmetric padding. The way to solve it is to add a Pad layer before the Conv layer to add the padding explicitly. I did it and now is working. Thanks!
The
cifar_resnet
,cifar_resnet_da_bn
andcifar_resnet50_da_bn
fail with the following error:Tested with EDDL v0.1a.