Dav-Jay

A bug on FactorizedReduce in operations.py?

3

I wonder whether there is a bug in [this line](https://github.com/quark0/darts/blob/f276dd346a09ae3160f8e3aca5c7b193fda1da37/cnn/operations.py#L102) in the `FactorizedReduce` function: `out = torch.cat([self.conv_1(x), self.conv_2(x[:,:,1:,1:])], dim=1)` **NOTE: in above line, the input for `conv_2` is a sliced...

how to enable multi-gpu training in each node

2

Thanks for the excellent work! When I tried the provided `train_inception.sh` for CIFAR10 on two nodes, I found only one GPU is invoked in each node (GPU 0), is there...