Spaces:
Running
Running
File size: 4,773 Bytes
d960e2d 3de3832 d960e2d 3de3832 d960e2d |
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 |
import torch
import torch.nn as nn
import torch.nn.functional as F
import functools
try:
from .arch_util import EBlock, Attention_Light
from .arch_util_freq import EBlock_freq
except:
from arch_util import EBlock, Attention_Light
from arch_util_freq import EBlock_freq
class Network(nn.Module):
def __init__(self, img_channel=3,
width=16,
middle_blk_num=1,
enc_blk_nums=[],
dec_blk_nums=[],
dilations = [1],
extra_depth_wise = False):
super(Network, self).__init__()
self.intro = nn.Conv2d(in_channels=img_channel, out_channels=width, kernel_size=3, padding=1, stride=1, groups=1,
bias=True)
self.ending = nn.Conv2d(in_channels=width, out_channels=img_channel, kernel_size=3, padding=1, stride=1, groups=1,
bias=True)
self.encoders = nn.ModuleList()
self.decoders = nn.ModuleList()
self.middle_blks = nn.ModuleList()
self.ups = nn.ModuleList()
self.downs = nn.ModuleList()
chan = width
for num in enc_blk_nums:
self.encoders.append(
nn.Sequential(
*[EBlock(chan, dilations = dilations, extra_depth_wise=extra_depth_wise) for _ in range(num)]
)
)
self.downs.append(
nn.Conv2d(chan, 2*chan, 2, 2)
)
chan = chan * 2
self.middle_blks = \
nn.Sequential(
*[EBlock(chan, dilations = dilations, extra_depth_wise=extra_depth_wise) for _ in range(middle_blk_num)]
)
for num in dec_blk_nums:
self.ups.append(
nn.Sequential(
nn.Conv2d(chan, chan * 2, 1, bias=False),
nn.PixelShuffle(2)
)
)
chan = chan // 2
self.decoders.append(
nn.Sequential(
*[EBlock(chan, extra_depth_wise=extra_depth_wise) for _ in range(num)]
)
)
self.padder_size = 2 ** len(self.encoders)
#define the attention layers
# self.recon_trunk_light = nn.Sequential(*[FBlock(c = chan * self.padder_size,
# DW_Expand=2, FFN_Expand=2, dilations = dilations,
# extra_depth_wise = False) for i in range(residual_layers)])
# ResidualBlock_noBN_f = functools.partial(ResidualBlock_noBN, nf = width * self.padder_size)
# self.recon_trunk_light = make_layer(ResidualBlock_noBN_f, residual_layers)
def forward(self, input):
_, _, H, W = input.shape
input = self.check_image_size(input)
x = self.intro(input)
encs = []
# i = 0
for encoder, down in zip(self.encoders, self.downs):
x = encoder(x)
# print(i, x.shape)
encs.append(x)
x = down(x)
# i += 1
x = self.middle_blks(x)
# print('3', x.shape)
# apply the mask
# x = x * mask
# x = self.recon_trunk_light(x)
for decoder, up, enc_skip in zip(self.decoders, self.ups, encs[::-1]):
x = up(x)
x = x + enc_skip
x = decoder(x)
x = self.ending(x)
x = x + input
return x[:, :, :H, :W]
def check_image_size(self, x):
_, _, h, w = x.size()
mod_pad_h = (self.padder_size - h % self.padder_size) % self.padder_size
mod_pad_w = (self.padder_size - w % self.padder_size) % self.padder_size
x = F.pad(x, (0, mod_pad_w, 0, mod_pad_h), value = 0)
return x
if __name__ == '__main__':
img_channel = 3
width = 32
enc_blks = [1, 2, 3]
middle_blk_num = 3
dec_blks = [3, 1, 1]
residual_layers = 2
dilations = [1, 4]
net = Network(img_channel=img_channel,
width=width,
middle_blk_num=middle_blk_num,
enc_blk_nums=enc_blks,
dec_blk_nums=dec_blks,
dilations = dilations)
# NAF = NAFNet(img_channel=img_channel, width=width, middle_blk_num=middle_blk_num,
# enc_blk_nums=enc_blks, dec_blk_nums=dec_blks)
inp_shape = (3, 256, 256)
from ptflops import get_model_complexity_info
macs, params = get_model_complexity_info(net, inp_shape, verbose=False, print_per_layer_stat=False)
print(macs, params)
inp = torch.randn(1, 3, 256, 256)
out = net(inp)
|