Michael Churchill

The data is on Edison, /scratch2/scratchdirs/rchurchi/xgca_rmc_mira043redo/xgc.oneddiag.bp It's a single file (yes, written by a single process), no directories with subfiles (total size is only about 1.5GB). I found that on...

HDF5 is in the same location, you can try it there also (just h5 suffix instead of bp).

I don't think so. I tried both the bp and h5 today, after having accessed them last week (so I assume both were out of cache by now). Both had...

There are situations where specifying an $f(\psi)$ even in the SOL could be useful (e.g. general testing, or also our specific case where the diagnostic line-of-sight doesn't vary much in...

@ezyang Sorry, I assume you mean use `torch.view_as_real`, but I'm unsure how to modify the above DDP example to use it, or do you mean for a custom distributed setup?

I think this gives you what you want: ``` class complexModel(nn.Module): def _init__(self, rank): super (complexModel, self).__init_ weights1 = torch.rand(10,10,12,12, dtype=torch.cfloat).to(rank) self.weights1 = nn.Parameter(torch.view _as_real(weights1)) def forward(self,x): weights1 = torch.view_as_complex(self.weights1)...