efficient_densenet_pytorch and densenet.pytorch
These are competitors offering alternative implementations of the same architecture, where the gpleiss version is optimized for memory efficiency while the bamos version prioritizes standard PyTorch implementation clarity.
About efficient_densenet_pytorch
gpleiss/efficient_densenet_pytorch
A memory-efficient implementation of DenseNets
Uses PyTorch's checkpointing feature to discard intermediate feature maps during forward passes and recompute them during backpropagation, reducing memory consumption from quadratic to linear with network depth. Supports both efficient and standard modes with configurable depth and growth rates, compatible with CIFAR/SVHN (small inputs) and ImageNet (large inputs). Includes a demo script with single and multi-GPU training support via python-fire.
About densenet.pytorch
bamos/densenet.pytorch
A PyTorch implementation of DenseNet.
Scores updated daily from GitHub, PyPI, and npm data. How scores work