search

tdeboissiere / DeepLearningImplementations

Implementation of recent Deep Learning papers
MIT License
1.81k stars 652 forks source link

deconv forward_pass #90

Open don-tpanic opened 4 years ago

don-tpanic commented 4 years ago

Sorry if I missed anything, but I wonder if there is any reason why switch was not used for maxunpooling as the original paper stated?

Thanks,