162 lines
5.3 KiB
Python
162 lines
5.3 KiB
Python
import torch
|
||
import torch.nn as nn
|
||
import torch.nn.functional as F
|
||
|
||
"""
|
||
基于傅里叶变换优化的双层三明治结构模型
|
||
新增：TemporalFourierBlock 用于全局捕捉时序频域特征，提升预测精度
|
||
第一层：Fourier 时域 -> 空间 -> 时间
|
||
残差连接：层输出 + 层输入
|
||
第二层：同样三明治结构 -> 最终输出
|
||
"""
|
||
|
||
class TemporalFourierBlock(nn.Module):
|
||
"""
|
||
时序傅里叶变换块
|
||
输入： x (B, T, N)
|
||
输出：时域重构 (B, T, N)
|
||
"""
|
||
def __init__(self, seq_len):
|
||
super().__init__()
|
||
# 频域系数学习：对每个频率分量应用可学习缩放
|
||
# rfft 输出频率数 = seq_len//2 + 1
|
||
freq_len = seq_len // 2 + 1
|
||
self.scale = nn.Parameter(torch.randn(freq_len), requires_grad=True)
|
||
self.seq_len = seq_len
|
||
|
||
def forward(self, x):
|
||
# x: (B, T, N)
|
||
# FFT 到频域
|
||
Xf = torch.fft.rfft(x, dim=1) # (B, F, N), complex
|
||
# 学习缩放：实部和虚部同时缩放
|
||
scale = self.scale.view(1, -1, 1)
|
||
Xf = Xf * scale
|
||
# IFFT 回时域
|
||
x_rec = torch.fft.irfft(Xf, n=self.seq_len, dim=1) # (B, T, N)
|
||
return x_rec
|
||
|
||
|
||
class DynamicGraphConstructor(nn.Module):
|
||
def __init__(self, node_num, embed_dim):
|
||
super().__init__()
|
||
self.nodevec1 = nn.Parameter(torch.randn(node_num, embed_dim), requires_grad=True)
|
||
self.nodevec2 = nn.Parameter(torch.randn(node_num, embed_dim), requires_grad=True)
|
||
|
||
def forward(self):
|
||
adj = torch.matmul(self.nodevec1, self.nodevec2.T)
|
||
adj = F.relu(adj)
|
||
adj = F.softmax(adj, dim=-1)
|
||
return adj
|
||
|
||
|
||
class GraphConvBlock(nn.Module):
|
||
def __init__(self, input_dim, output_dim):
|
||
super().__init__()
|
||
self.theta = nn.Linear(input_dim, output_dim)
|
||
self.residual = (input_dim == output_dim)
|
||
if not self.residual:
|
||
self.res_proj = nn.Linear(input_dim, output_dim)
|
||
|
||
def forward(self, x, adj):
|
||
# x: (B, N, C); adj: (N, N)
|
||
res = x
|
||
x = torch.matmul(adj, x)
|
||
x = self.theta(x)
|
||
x = x + (res if self.residual else self.res_proj(res))
|
||
return F.relu(x)
|
||
|
||
|
||
class MANBA_Block(nn.Module):
|
||
def __init__(self, input_dim, hidden_dim):
|
||
super().__init__()
|
||
self.attn = nn.MultiheadAttention(embed_dim=input_dim, num_heads=4, batch_first=True)
|
||
self.ffn = nn.Sequential(
|
||
nn.Linear(input_dim, hidden_dim),
|
||
nn.ReLU(),
|
||
nn.Linear(hidden_dim, input_dim)
|
||
)
|
||
self.norm1 = nn.LayerNorm(input_dim)
|
||
self.norm2 = nn.LayerNorm(input_dim)
|
||
|
||
def forward(self, x):
|
||
# x: (B, N, C) 视 N 维为时间序列长度
|
||
res = x
|
||
x_attn, _ = self.attn(x, x, x)
|
||
x = self.norm1(res + x_attn)
|
||
res2 = x
|
||
x_ffn = self.ffn(x)
|
||
x = self.norm2(res2 + x_ffn)
|
||
return x
|
||
|
||
|
||
class SandwichBlock(nn.Module):
|
||
"""
|
||
时间-空间-时间 三明治结构
|
||
输入/输出： (B, N, hidden_dim)
|
||
"""
|
||
def __init__(self, num_nodes, embed_dim, hidden_dim):
|
||
super().__init__()
|
||
self.manba1 = MANBA_Block(hidden_dim, hidden_dim * 2)
|
||
self.graph_constructor = DynamicGraphConstructor(num_nodes, embed_dim)
|
||
self.gc = GraphConvBlock(hidden_dim, hidden_dim)
|
||
self.manba2 = MANBA_Block(hidden_dim, hidden_dim * 2)
|
||
|
||
def forward(self, h):
|
||
# h: (B, N, hidden_dim)
|
||
h1 = self.manba1(h)
|
||
adj = self.graph_constructor()
|
||
h2 = self.gc(h1, adj)
|
||
h3 = self.manba2(h2)
|
||
return h3
|
||
|
||
|
||
class EXP(nn.Module):
|
||
def __init__(self, args):
|
||
super().__init__()
|
||
self.horizon = args['horizon']
|
||
self.output_dim = args['output_dim']
|
||
self.seq_len = args.get('in_len', 12)
|
||
self.hidden_dim = args.get('hidden_dim', 64)
|
||
self.num_nodes = args['num_nodes']
|
||
self.embed_dim = args.get('embed_dim', 16)
|
||
|
||
# 时序傅里叶块
|
||
self.fourier_block = TemporalFourierBlock(self.seq_len)
|
||
|
||
# 输入映射：(B*N, T) -> hidden_dim
|
||
self.input_proj = nn.Linear(self.seq_len, self.hidden_dim)
|
||
|
||
# 两层三明治块
|
||
self.sandwich1 = SandwichBlock(self.num_nodes, self.embed_dim, self.hidden_dim)
|
||
self.sandwich2 = SandwichBlock(self.num_nodes, self.embed_dim, self.hidden_dim)
|
||
|
||
# 输出映射
|
||
self.out_proj = nn.Linear(self.hidden_dim, self.horizon * self.output_dim)
|
||
|
||
def forward(self, x):
|
||
# x: (B, T, N, D_total)
|
||
x_main = x[..., 0] # (B, T, N)
|
||
B, T, N = x_main.shape
|
||
assert T == self.seq_len
|
||
|
||
# 时序傅里叶变换 + 残差
|
||
x_freq = self.fourier_block(x_main) # (B, T, N)
|
||
x_main = x_main + x_freq
|
||
|
||
# 输入投影 (B, T, N) -> (B*N, T) -> (B, N, hidden_dim)
|
||
x_flat = x_main.permute(0, 2, 1).reshape(B * N, T)
|
||
h0 = self.input_proj(x_flat).view(B, N, self.hidden_dim)
|
||
|
||
# 第一层三明治 + 残差
|
||
h1 = self.sandwich1(h0)
|
||
h1 = h1 + h0
|
||
|
||
# 第二层三明治
|
||
h2 = self.sandwich2(h1)
|
||
|
||
# 输出映射
|
||
out = self.out_proj(h2) # (B, N, H*D_out)
|
||
out = out.view(B, N, self.horizon, self.output_dim)
|
||
out = out.permute(0, 2, 1, 3) # (B, horizon, N, output_dim)
|
||
return out
|