MSDNet-PyTorch
MSDNet-PyTorch copied to clipboard
MSDNet
请问这个实现了论文中所说的lazy evaluation吗
I am trying to reproduce the `MSDNet with dynamic evaluation` curve of Figure 7b of the paper (budgeted batch classification for CIFAR-100). I am training the MSDNet [as mentioned in...
Thanks for sharing your code. I find some anomalous code In msdnet.py#L322. these three "if branch" will not be excuted in func `_init_weights(self, m)`. https://github.com/kalviny/MSDNet-PyTorch/blob/c16a62be5b2166ef3333290aa8c39d0687c3c7ad/models/msdnet.py#L242-L264
What do you mean by arguments --step, --stepmode, --use-valid, -j and --base. Also please explain their functions in the formation of blocks of MSDNET ?
D:\software\ana\envs\env1\python.exe C:/Users/Administrator/Desktop/黄高/MSDNet-PyTorch-master/main.py Traceback (most recent call last): File "C:/Users/Administrator/Desktop/黄高/MSDNet-PyTorch-master/main.py", line 352, in main() File "C:/Users/Administrator/Desktop/黄高/MSDNet-PyTorch-master/main.py", line 62, in main model = getattr(models, args.arch)(args) AttributeError: module 'models' has no attribute 'resnet'...
Which version of ILSVRC ([ImageNet](https://image-net.org/index.php)) dataset was used for training the MSDNet in this repo? Was it the 2012 one (i.e. the same one as in the original MSDNet paper)?