jhjacobsen / pytorch-i-revnet

i-RevNet Pytorch Code

jhjacobsen/pytorch-i-revnet Issues

Questions about using the i-resnet for other applications.
Updated 2 years ago1
injective or bijective?
Closed 5 years ago14
Some confusion for case 'stride = 2'
Updated 2 years ago
question about --nChannels
Updated 3 years ago
iRevNet.py's test code doesn't work
Closed 3 years ago4
Consider more efficient implementation of class psi
Closed 4 years ago4
Invertible up sampling
Closed 4 years ago1
About model size of i-RevNet (b)
Closed 4 years ago2
multi-GPU
Closed 4 years ago
download pretrained model
Closed 5 years ago1
Pretrained models cannot be loaded
Closed 5 years ago3
Cifar exception in tensor size at beginning of training
Closed 5 years ago1
for "--nBlocks 18 18 18 --nStrides 1 2 2 --nChannels 16 64 256" architecture,Is there pretrained model ?
Closed 5 years ago2
Have you implemented backprop without storing forward activations?
Closed 5 years ago1
is it fully invertible? last layer is pooling+ linear
Closed 5 years ago2
Hi, I don't understand permute function
Closed 6 years ago1
Inverse function has zero block size for final psi function initialization
Closed 6 years ago1
Potential bug in model_utils.py
Closed 6 years ago1
The size of tensor are not identifed
Closed 6 years ago1
Question: memory saving
Closed 6 years ago1
I really don't understand what dose the psi class do, any explain would be helpful
Closed 6 years ago1
Undefined name 'revnet' in CIFAR_main.py
Closed 6 years ago