Back to groupnorm
This commit is contained in:
parent
336f807c8e
commit
6dec1f5968
|
@ -185,10 +185,10 @@ class SwitchedConvHardRouting(nn.Module):
|
||||||
self.coupler = Conv2d(coupler_dim_in, breadth, kernel_size=1, stride=self.stride)
|
self.coupler = Conv2d(coupler_dim_in, breadth, kernel_size=1, stride=self.stride)
|
||||||
elif coupler_mode == 'lambda':
|
elif coupler_mode == 'lambda':
|
||||||
self.coupler = nn.Sequential(nn.Conv2d(coupler_dim_in, coupler_dim_in, 1),
|
self.coupler = nn.Sequential(nn.Conv2d(coupler_dim_in, coupler_dim_in, 1),
|
||||||
nn.BatchNorm2d(coupler_dim_in),
|
nn.GroupNorm(num_groups=2, num_channels=coupler_dim_in),
|
||||||
nn.ReLU(),
|
nn.ReLU(),
|
||||||
LambdaLayer(dim=coupler_dim_in, dim_out=breadth, r=23, dim_k=16, heads=2, dim_u=1),
|
LambdaLayer(dim=coupler_dim_in, dim_out=breadth, r=23, dim_k=16, heads=2, dim_u=1),
|
||||||
nn.BatchNorm2d(breadth),
|
nn.GroupNorm(num_groups=1, num_channels=breadth),
|
||||||
nn.ReLU(),
|
nn.ReLU(),
|
||||||
Conv2d(breadth, breadth, 1, stride=self.stride))
|
Conv2d(breadth, breadth, 1, stride=self.stride))
|
||||||
elif coupler_mode == 'lambda2':
|
elif coupler_mode == 'lambda2':
|
||||||
|
@ -199,7 +199,7 @@ class SwitchedConvHardRouting(nn.Module):
|
||||||
nn.GroupNorm(num_groups=2, num_channels=coupler_dim_in),
|
nn.GroupNorm(num_groups=2, num_channels=coupler_dim_in),
|
||||||
nn.ReLU(),
|
nn.ReLU(),
|
||||||
LambdaLayer(dim=coupler_dim_in, dim_out=breadth, r=23, dim_k=16, heads=2, dim_u=1),
|
LambdaLayer(dim=coupler_dim_in, dim_out=breadth, r=23, dim_k=16, heads=2, dim_u=1),
|
||||||
nn.GroupNorm(num_groups=2, num_channels=breadth),
|
nn.GroupNorm(num_groups=1, num_channels=breadth),
|
||||||
nn.ReLU(),
|
nn.ReLU(),
|
||||||
Conv2d(breadth, breadth, 1, stride=self.stride))
|
Conv2d(breadth, breadth, 1, stride=self.stride))
|
||||||
else:
|
else:
|
||||||
|
|
|
@ -3,8 +3,8 @@ from time import time
|
||||||
|
|
||||||
import torch
|
import torch
|
||||||
import torchvision
|
import torchvision
|
||||||
|
import torch.distributed as distributed
|
||||||
from torch import nn
|
from torch import nn
|
||||||
from torch.nn.parallel import distributed
|
|
||||||
from tqdm import tqdm
|
from tqdm import tqdm
|
||||||
|
|
||||||
from models.switched_conv.switched_conv_hard_routing import SwitchedConvHardRouting, \
|
from models.switched_conv.switched_conv_hard_routing import SwitchedConvHardRouting, \
|
||||||
|
|
Loading…
Reference in New Issue
Block a user