Spaces:
Build error
Build error
File size: 5,299 Bytes
19c9e2c |
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 |
import imp
import torch
import torch.nn as nn
from timm.models.layers import trunc_normal_, DropPath, to_2tuple
import os
from .blocks import Mlp
class query_Attention(nn.Module):
def __init__(self, dim, num_heads=2, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.):
super().__init__()
self.num_heads = num_heads
head_dim = dim // num_heads
# NOTE scale factor was wrong in my original version, can set manually to be compat with prev weights
self.scale = qk_scale or head_dim ** -0.5
self.q = nn.Parameter(torch.ones((1, 10, dim)), requires_grad=True)
self.k = nn.Linear(dim, dim, bias=qkv_bias)
self.v = nn.Linear(dim, dim, bias=qkv_bias)
self.attn_drop = nn.Dropout(attn_drop)
self.proj = nn.Linear(dim, dim)
self.proj_drop = nn.Dropout(proj_drop)
def forward(self, x):
B, N, C = x.shape
k = self.k(x).reshape(B, N, self.num_heads, C // self.num_heads).permute(0, 2, 1, 3)
v = self.v(x).reshape(B, N, self.num_heads, C // self.num_heads).permute(0, 2, 1, 3)
q = self.q.expand(B, -1, -1).view(B, -1, self.num_heads, C // self.num_heads).permute(0, 2, 1, 3)
attn = (q @ k.transpose(-2, -1)) * self.scale
attn = attn.softmax(dim=-1)
attn = self.attn_drop(attn)
x = (attn @ v).transpose(1, 2).reshape(B, 10, C)
x = self.proj(x)
x = self.proj_drop(x)
return x
class query_SABlock(nn.Module):
def __init__(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0.,
drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm):
super().__init__()
self.pos_embed = nn.Conv2d(dim, dim, 3, padding=1, groups=dim)
self.norm1 = norm_layer(dim)
self.attn = query_Attention(
dim,
num_heads=num_heads, qkv_bias=qkv_bias, qk_scale=qk_scale,
attn_drop=attn_drop, proj_drop=drop)
# NOTE: drop path for stochastic depth, we shall see if this is better than dropout here
self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.norm2 = norm_layer(dim)
mlp_hidden_dim = int(dim * mlp_ratio)
self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)
def forward(self, x):
x = x + self.pos_embed(x)
x = x.flatten(2).transpose(1, 2)
x = self.drop_path(self.attn(self.norm1(x)))
x = x + self.drop_path(self.mlp(self.norm2(x)))
return x
class conv_embedding(nn.Module):
def __init__(self, in_channels, out_channels):
super(conv_embedding, self).__init__()
self.proj = nn.Sequential(
nn.Conv2d(in_channels, out_channels // 2, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)),
nn.BatchNorm2d(out_channels // 2),
nn.GELU(),
# nn.Conv2d(out_channels // 2, out_channels // 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
# nn.BatchNorm2d(out_channels // 2),
# nn.GELU(),
nn.Conv2d(out_channels // 2, out_channels, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)),
nn.BatchNorm2d(out_channels),
)
def forward(self, x):
x = self.proj(x)
return x
class Global_pred(nn.Module):
def __init__(self, in_channels=3, out_channels=64, num_heads=4, type='exp'):
super(Global_pred, self).__init__()
if type == 'exp':
self.gamma_base = nn.Parameter(torch.ones((1)), requires_grad=False) # False in exposure correction
else:
self.gamma_base = nn.Parameter(torch.ones((1)), requires_grad=True)
self.color_base = nn.Parameter(torch.eye((3)), requires_grad=True) # basic color matrix
# main blocks
self.conv_large = conv_embedding(in_channels, out_channels)
self.generator = query_SABlock(dim=out_channels, num_heads=num_heads)
self.gamma_linear = nn.Linear(out_channels, 1)
self.color_linear = nn.Linear(out_channels, 1)
self.apply(self._init_weights)
for name, p in self.named_parameters():
if name == 'generator.attn.v.weight':
nn.init.constant_(p, 0)
def _init_weights(self, m):
if isinstance(m, nn.Linear):
trunc_normal_(m.weight, std=.02)
if isinstance(m, nn.Linear) and m.bias is not None:
nn.init.constant_(m.bias, 0)
elif isinstance(m, nn.LayerNorm):
nn.init.constant_(m.bias, 0)
nn.init.constant_(m.weight, 1.0)
def forward(self, x):
#print(self.gamma_base)
x = self.conv_large(x)
x = self.generator(x)
gamma, color = x[:, 0].unsqueeze(1), x[:, 1:]
gamma = self.gamma_linear(gamma).squeeze(-1) + self.gamma_base
#print(self.gamma_base, self.gamma_linear(gamma))
color = self.color_linear(color).squeeze(-1).view(-1, 3, 3) + self.color_base
return gamma, color
if __name__ == "__main__":
os.environ['CUDA_VISIBLE_DEVICES']='3'
#net = Local_pred_new().cuda()
img = torch.Tensor(8, 3, 400, 600)
global_net = Global_pred()
gamma, color = global_net(img)
print(gamma.shape, color.shape) |