126 lines
3.2 KiB
Python
126 lines
3.2 KiB
Python
import torch
|
|
import torch.nn as nn
|
|
import torch.nn.functional as F
|
|
|
|
# Try to import comfy.ops, but provide a fallback if it's not available
|
|
try:
|
|
import comfy.ops
|
|
ops = comfy.ops.manual_cast
|
|
except ImportError:
|
|
# Define a simple fallback for ops.Conv2d and ops.Linear
|
|
class OpsFallback:
|
|
@staticmethod
|
|
def Conv2d(*args, **kwargs):
|
|
return nn.Conv2d(*args, **kwargs)
|
|
|
|
@staticmethod
|
|
def Linear(*args, **kwargs):
|
|
return nn.Linear(*args, **kwargs)
|
|
|
|
ops = OpsFallback()
|
|
|
|
def _make_scratch(in_shape, out_shape, groups=1, expand=False):
|
|
scratch = nn.Module()
|
|
|
|
out_shape1 = out_shape
|
|
out_shape2 = out_shape
|
|
out_shape3 = out_shape
|
|
out_shape4 = out_shape
|
|
if expand:
|
|
out_shape1 = out_shape
|
|
out_shape2 = out_shape * 2
|
|
out_shape3 = out_shape * 4
|
|
out_shape4 = out_shape * 8
|
|
|
|
scratch.layer1_rn = nn.Conv2d(
|
|
in_shape[0], out_shape1, kernel_size=3, stride=1, padding=1, bias=False, groups=groups
|
|
)
|
|
scratch.layer2_rn = nn.Conv2d(
|
|
in_shape[1], out_shape2, kernel_size=3, stride=1, padding=1, bias=False, groups=groups
|
|
)
|
|
scratch.layer3_rn = nn.Conv2d(
|
|
in_shape[2], out_shape3, kernel_size=3, stride=1, padding=1, bias=False, groups=groups
|
|
)
|
|
scratch.layer4_rn = nn.Conv2d(
|
|
in_shape[3], out_shape4, kernel_size=3, stride=1, padding=1, bias=False, groups=groups
|
|
)
|
|
|
|
return scratch
|
|
|
|
class FeatureFusionBlock(nn.Module):
|
|
"""Feature fusion block."""
|
|
|
|
def __init__(
|
|
self,
|
|
features,
|
|
activation,
|
|
deconv=False,
|
|
bn=False,
|
|
expand=False,
|
|
align_corners=True,
|
|
size=None,
|
|
):
|
|
"""Init.
|
|
|
|
Args:
|
|
features (int): number of features
|
|
"""
|
|
super(FeatureFusionBlock, self).__init__()
|
|
|
|
self.deconv = deconv
|
|
self.align_corners = align_corners
|
|
|
|
self.groups = 1
|
|
|
|
self.expand = expand
|
|
out_features = features
|
|
if self.expand:
|
|
out_features = features // 2
|
|
|
|
self.out_conv = nn.Sequential(
|
|
ops.Conv2d(
|
|
features,
|
|
out_features,
|
|
kernel_size=1,
|
|
stride=1,
|
|
padding=0,
|
|
bias=True,
|
|
groups=1,
|
|
),
|
|
nn.ReLU(inplace=True),
|
|
ops.Conv2d(
|
|
out_features,
|
|
out_features,
|
|
kernel_size=3,
|
|
stride=1,
|
|
padding=1,
|
|
bias=True,
|
|
groups=1,
|
|
),
|
|
nn.ReLU(inplace=True),
|
|
)
|
|
self.size = size
|
|
|
|
def forward(self, *xs, size=None):
|
|
"""Forward pass.
|
|
|
|
Returns:
|
|
tensor: output
|
|
"""
|
|
output = xs[0]
|
|
|
|
if len(xs) == 2:
|
|
res = xs[1]
|
|
output = output + res
|
|
|
|
if size is None and self.size is None:
|
|
size = output.shape[-2:]
|
|
elif size is None and self.size is not None:
|
|
size = self.size
|
|
|
|
output = nn.functional.interpolate(
|
|
output, size=size, mode="bilinear", align_corners=self.align_corners
|
|
)
|
|
output = self.out_conv(output)
|
|
return output
|