File size: 6,448 Bytes
1504958 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 |
import math
from typing import Tuple, Union
import torch
import torch.nn as nn
import torch.nn.functional as F
from einops import rearrange
class DualConv3d(nn.Module):
def __init__(
self,
in_channels,
out_channels,
kernel_size,
stride: Union[int, Tuple[int, int, int]] = 1,
padding: Union[int, Tuple[int, int, int]] = 0,
dilation: Union[int, Tuple[int, int, int]] = 1,
groups=1,
bias=True,
):
super(DualConv3d, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
# Ensure kernel_size, stride, padding, and dilation are tuples of length 3
if isinstance(kernel_size, int):
kernel_size = (kernel_size, kernel_size, kernel_size)
if kernel_size == (1, 1, 1):
raise ValueError(
"kernel_size must be greater than 1. Use make_linear_nd instead."
)
if isinstance(stride, int):
stride = (stride, stride, stride)
if isinstance(padding, int):
padding = (padding, padding, padding)
if isinstance(dilation, int):
dilation = (dilation, dilation, dilation)
# Set parameters for convolutions
self.groups = groups
self.bias = bias
# Define the size of the channels after the first convolution
intermediate_channels = (
out_channels if in_channels < out_channels else in_channels
)
# Define parameters for the first convolution
self.weight1 = nn.Parameter(
torch.Tensor(
intermediate_channels,
in_channels // groups,
1,
kernel_size[1],
kernel_size[2],
)
)
self.stride1 = (1, stride[1], stride[2])
self.padding1 = (0, padding[1], padding[2])
self.dilation1 = (1, dilation[1], dilation[2])
if bias:
self.bias1 = nn.Parameter(torch.Tensor(intermediate_channels))
else:
self.register_parameter("bias1", None)
# Define parameters for the second convolution
self.weight2 = nn.Parameter(
torch.Tensor(
out_channels, intermediate_channels // groups, kernel_size[0], 1, 1
)
)
self.stride2 = (stride[0], 1, 1)
self.padding2 = (padding[0], 0, 0)
self.dilation2 = (dilation[0], 1, 1)
if bias:
self.bias2 = nn.Parameter(torch.Tensor(out_channels))
else:
self.register_parameter("bias2", None)
# Initialize weights and biases
self.reset_parameters()
def reset_parameters(self):
nn.init.kaiming_uniform_(self.weight1, a=math.sqrt(5))
nn.init.kaiming_uniform_(self.weight2, a=math.sqrt(5))
if self.bias:
fan_in1, _ = nn.init._calculate_fan_in_and_fan_out(self.weight1)
bound1 = 1 / math.sqrt(fan_in1)
nn.init.uniform_(self.bias1, -bound1, bound1)
fan_in2, _ = nn.init._calculate_fan_in_and_fan_out(self.weight2)
bound2 = 1 / math.sqrt(fan_in2)
nn.init.uniform_(self.bias2, -bound2, bound2)
def forward(self, x, use_conv3d=False, skip_time_conv=False):
if use_conv3d:
return self.forward_with_3d(x=x, skip_time_conv=skip_time_conv)
else:
return self.forward_with_2d(x=x, skip_time_conv=skip_time_conv)
def forward_with_3d(self, x, skip_time_conv):
# First convolution
x = F.conv3d(
x,
self.weight1,
self.bias1,
self.stride1,
self.padding1,
self.dilation1,
self.groups,
)
if skip_time_conv:
return x
# Second convolution
x = F.conv3d(
x,
self.weight2,
self.bias2,
self.stride2,
self.padding2,
self.dilation2,
self.groups,
)
return x
def forward_with_2d(self, x, skip_time_conv):
b, c, d, h, w = x.shape
# First 2D convolution
x = rearrange(x, "b c d h w -> (b d) c h w")
# Squeeze the depth dimension out of weight1 since it's 1
weight1 = self.weight1.squeeze(2)
# Select stride, padding, and dilation for the 2D convolution
stride1 = (self.stride1[1], self.stride1[2])
padding1 = (self.padding1[1], self.padding1[2])
dilation1 = (self.dilation1[1], self.dilation1[2])
x = F.conv2d(x, weight1, self.bias1, stride1, padding1, dilation1, self.groups)
_, _, h, w = x.shape
if skip_time_conv:
x = rearrange(x, "(b d) c h w -> b c d h w", b=b)
return x
# Second convolution which is essentially treated as a 1D convolution across the 'd' dimension
x = rearrange(x, "(b d) c h w -> (b h w) c d", b=b)
# Reshape weight2 to match the expected dimensions for conv1d
weight2 = self.weight2.squeeze(-1).squeeze(-1)
# Use only the relevant dimension for stride, padding, and dilation for the 1D convolution
stride2 = self.stride2[0]
padding2 = self.padding2[0]
dilation2 = self.dilation2[0]
x = F.conv1d(x, weight2, self.bias2, stride2, padding2, dilation2, self.groups)
x = rearrange(x, "(b h w) c d -> b c d h w", b=b, h=h, w=w)
return x
@property
def weight(self):
return self.weight2
def test_dual_conv3d_consistency():
# Initialize parameters
in_channels = 3
out_channels = 5
kernel_size = (3, 3, 3)
stride = (2, 2, 2)
padding = (1, 1, 1)
# Create an instance of the DualConv3d class
dual_conv3d = DualConv3d(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=kernel_size,
stride=stride,
padding=padding,
bias=True,
)
# Example input tensor
test_input = torch.randn(1, 3, 10, 10, 10)
# Perform forward passes with both 3D and 2D settings
output_conv3d = dual_conv3d(test_input, use_conv3d=True)
output_2d = dual_conv3d(test_input, use_conv3d=False)
# Assert that the outputs from both methods are sufficiently close
assert torch.allclose(
output_conv3d, output_2d, atol=1e-6
), "Outputs are not consistent between 3D and 2D convolutions."
|