47轉置卷積

import torch
from torch import nn
from d2l import torch as d2l# 輸入矩陣X和卷積核矩陣K實現基本的轉置卷積運算
def trans_conv(X, K):h, w = K.shapeY = torch.zeros((X.shape[0] + h - 1, X.shape[1] + w - 1))for i in range(X.shape[0]):for j in range(X.shape[1]):Y[i: i + h, j: j + w] += X[i, j] * Kreturn YX = torch.tensor([[0.0, 1.0], [2.0, 3.0]])
K = torch.tensor([[0.0, 1.0], [2.0, 3.0]])
print(trans_conv(X, K))
"""
tensor([[ 0.,  0.,  1.],[ 0.,  4.,  6.],[ 4., 12.,  9.]])
"""# 使用高級API獲得相同的結果
X, K = X.reshape(1, 1, 2, 2), K.reshape(1, 1, 2, 2)
tconv = nn.ConvTranspose2d(1, 1, kernel_size=2, bias=False)
tconv.weight.data = K
print(tconv(X))
"""
tensor([[[[ 0.,  0.,  1.],[ 0.,  4.,  6.],[ 4., 12.,  9.]]]], grad_fn=<SlowConvTranspose2DBackward>)
"""# 填充、步幅和多通道
# 當將高和寬兩側的填充數指定為1時，轉置卷積的輸出中將刪除第一和最后的行與列。
tconv = nn.ConvTranspose2d(1, 1, kernel_size=2, padding=1, bias=False)
tconv.weight.data = K
print(tconv(X))
# tensor([[[[4.]]]], grad_fn=<SlowConvTranspose2DBackward>)# 步幅為2的轉置卷積的輸出
tconv = nn.ConvTranspose2d(1, 1, kernel_size=2, stride=2, bias=False)
tconv.weight.data = K
print(tconv(X))
"""
tensor([[[[0., 0., 0., 1.],[0., 0., 2., 3.],[0., 2., 0., 3.],[4., 6., 6., 9.]]]]
"""X = torch.rand(size=(1, 10, 16, 16))
conv = nn.Conv2d(10, 20, kernel_size=5, padding=2, stride=3)
tconv = nn.ConvTranspose2d(20, 10, kernel_size=5, padding=2, stride=3)
print(conv(X).shape) # torch.Size([1, 20, 6, 6])
print(tconv(conv(X)).shape) # torch.Size([1, 10, 16, 16])
print(tconv(conv(X)).shape == X.shape) # True# 與矩陣變換的聯系
X = torch.arange(9.0).reshape(3, 3)
K = torch.tensor([[1.0, 2.0], [3.0, 4.0]])
Y = d2l.corr2d(X, K)
print(Y)
"""
tensor([[27., 37.],[57., 67.]])
"""# 將卷積核K重寫為包含大量0的稀疏權重矩陣W。 權重矩陣的形狀是4 * 9
def kernel2matrix(K):k, W = torch.zeros(5), torch.zeros((4, 9))k[:2], k[3:5] = K[0, :], K[1, :]W[0, :5], W[1, 1:6], W[2, 3:8], W[3, 4:] = k, k, k, kreturn WW = kernel2matrix(K)
print(W)
"""
tensor([[1., 2., 0., 3., 4., 0., 0., 0., 0.],[0., 1., 2., 0., 3., 4., 0., 0., 0.],[0., 0., 0., 1., 2., 0., 3., 4., 0.],[0., 0., 0., 0., 1., 2., 0., 3., 4.]])
"""print(Y == torch.matmul(W, X.reshape(-1)).reshape(2, 2))
"""
tensor([[True, True],[True, True]])
"""# 使用矩陣乘法來實現轉置卷積
Z = trans_conv(Y, K)
print(Z == torch.matmul(W.T, Y.reshape(-1)).reshape(3, 3))
"""
tensor([[True, True, True],[True, True, True],[True, True, True]])
"""