其实之前写过的torch版本的neural style代码的解读,可以参考
Torch7学习(七)——Neural-Style代码解析,不过那是传统的层的思想的框架,如今都是计算图的思想了。pytorch版本的写法与之前的写法还是有一定差异的,主要是简单了很多!对比之后你会震撼的。
其他没啥好看的,主要看核心代码:
class ContentLoss(nn.Module):
def __init__(self, target, weight):
super(ContentLoss, self).__init__()
# we 'detach' the target content from the tree used
self.target = target.detach() * weight
# to dynamically compute the gradient: this is a stated value,
# not a variable. Otherwise the forward method of the criterion
# will throw an error.
self.weight = weight
self.criterion = nn.MSELoss()
def forward(self, input):
self.loss = self.criterion(input * self.weight, self.target)
self.output = input
return self.output
def backward(self, retain_graph=True):
self.loss.backward(retain_graph=retain_graph)
return self.loss
class GramMatrix(nn.Module):
def forward(self, input):
a, b, c, d = input.size() # a=batch size(=1)
# b=number of feature maps
# (c,d)=dimensions of a f. map (N=c*d)
features = input.view(a * b, c * d) # resise F_XL into \hat F_XL
G = torch.mm(features, features.t()) # compute the gram product
# we 'normalize' the values of the gram matrix
# by dividing by the number of element in each feature maps.
return G.div(a * b * c * d)
class StyleLoss(nn.Module):
def __init__(self, target, weight):
super(StyleLoss, self).__init__()
self.target = target.detach() * weight
self.weight = weight
self.gram = GramMatrix()
self.criterion = nn.MSELoss()
def forward(self, input):
self.output = input.clone()
self.G = self.gram(input)
self.G.mul_(self.weight)
self.loss = self.criterion(self.G, self.target)
return self.output
def backward(self, retain_graph=True):
self.loss.backward(retain_graph=retain_graph)
return self.loss
上面部分,我们发现比较诡异的地方:
1. 在Pytorch入门学习(八)—–自定义层的实现(甚至不可导operation的backward写法)中,我们知道如果要扩展自定义层,只需要重载nn.Module的forward函数就行。然后在forward里面调用xxxFunction.apply来调用自定义autograd类,在自定义autograd类中,再进行重载forward以及backward。但是这里为什么在自定义module中就有backward?
2. retain_graph有什么用?
3. Gram矩阵直接写forward就行,根本不用写反向传播,这个是真厉害!
local Gram, parent = torch.class('nn.GramMatrix', 'nn.Module')
function Gram:__init()
parent.__init(self)
end
function Gram:updateOutput(input)
assert(input:dim() == 3)
local C, H, W = input:size(1), input:size(2), input:size(3)
local x_flat = input:view(C, H * W)
self.output:resize(C, C)
self.output:mm(x_flat, x_flat:t())
return self.output
end
function Gram:updateGradInput(input, gradOutput)
assert(input:dim() == 3 and input:size(1))
local C, H, W = input:size(1), input:size(2), input:size(3)
local x_flat = input:view(C, H * W)
self.gradInput:resize(C, H * W):mm(gradOutput, x_flat)
self.gradInput:addmm(gradOutput:t(), x_flat)
self.gradInput = self.gradInput:view(C, H, W)
return self.gradInput
end
-- Define an nn Module to compute style loss in-place
local StyleLoss, parent = torch.class('nn.StyleLoss', 'nn.Module')
function StyleLoss:__init(strength, normalize)
parent.__init(self)
self.normalize = normalize or false
self.strength = strength
self.target = torch.Tensor()
self.mode = 'none'
self.loss = 0
self.gram = nn.GramMatrix()
self.blend_weight = nil
self.G = nil
self.crit = nn.MSECriterion()
end
function StyleLoss:updateOutput(input)
self.G = self.gram:forward(input)
self.G:div(input:nElement())
if self.mode == 'capture' then
if self.blend_weight == nil then
self.target:resizeAs(self.G):copy(self.G)
elseif self.target:nElement() == 0 then
self.target:resizeAs(self.G):copy(self.G):mul(self.blend_weight)
else
self.target:add(self.blend_weight, self.G)
end
elseif self.mode == 'loss' then
self.loss = self.strength * self.crit:forward(self.G, self.target)
end
self.output = input
return self.output
end
function StyleLoss:updateGradInput(input, gradOutput)
if self.mode == 'loss' then
local dG = self.crit:backward(self.G, self.target)
dG:div(input:nElement())
self.gradInput = self.gram:backward(input, dG)
if self.normalize then
self.gradInput:div(torch.norm(self.gradInput, 1) + 1e-8)
end
self.gradInput:mul(self.strength)
self.gradInput:add(gradOutput)
else
self.gradInput = gradOutput
end
return self.gradInput
end
前后代码量起码差一倍以上,而且书写难度后者困难很多!最主要的差别就是torch对于自定义层,得自己写updateGradInput啊!特别是gram的反向传播,可能并不是很容易写的。另外一点是StyleLoss的反向传播self.gradInput:add(gradOutput)
也是有点理解费劲。这点其实在隐藏层加监督(feature matching)的代码书写方法—- 附加optim包的功能再看。有了相应的说明。然而对于这种自动求导的框架,只要是你forward是完全用Variable进行计算的,那么就会创建一个正确的graph,那么反向就会正确!就是这么crazy!根本不用写反向传播!
其实这个“backward”知识一个普通的函数!并不是重载内部的backward!
在这个“backward”中主要是调用criterion的backward,然后返回这个loss,好让外面能拿到相应的损失。
其实看到后面就可以发现。
竟然可以直接将自定义层放入一个list中!
# desired depth layers to compute style/content losses :
content_layers_default = ['conv_4']
style_layers_default = ['conv_1', 'conv_2', 'conv_3', 'conv_4', 'conv_5']
def get_style_model_and_losses(cnn, style_img, content_img,
style_weight=1000, content_weight=1,
content_layers=content_layers_default,
style_layers=style_layers_default):
cnn = copy.deepcopy(cnn)
# just in order to have an iterable access to or list of content/syle
# losses(放入自定义层的list)
content_losses = []
style_losses = []
model = nn.Sequential() # the new Sequential module network
gram = GramMatrix() # we need a gram module in order to compute style targets
# move these modules to the GPU if possible:
if use_cuda:
model = model.cuda()
gram = gram.cuda()
i = 1
for layer in list(cnn):
if isinstance(layer, nn.Conv2d):
name = "conv_" + str(i)
model.add_module(name, layer)
if name in content_layers:
# add content loss:
target = model(content_img).clone()
content_loss = ContentLoss(target, content_weight)
model.add_module("content_loss_" + str(i), content_loss)
content_losses.append(content_loss)
if name in style_layers:
# add style loss:
target_feature = model(style_img).clone()
target_feature_gram = gram(target_feature)
style_loss = StyleLoss(target_feature_gram, style_weight)
model.add_module("style_loss_" + str(i), style_loss)
style_losses.append(style_loss)
if isinstance(layer, nn.ReLU):
name = "relu_" + str(i)
model.add_module(name, layer)
if name in content_layers:
# add content loss:
target = model(content_img).clone()
content_loss = ContentLoss(target, content_weight)
model.add_module("content_loss_" + str(i), content_loss)
content_losses.append(content_loss)
if name in style_layers:
# add style loss:
target_feature = model(style_img).clone()
target_feature_gram = gram(target_feature)
style_loss = StyleLoss(target_feature_gram, style_weight)
model.add_module("style_loss_" + str(i), style_loss)
style_losses.append(style_loss)
i += 1
if isinstance(layer, nn.MaxPool2d):
name = "pool_" + str(i)
model.add_module(name, layer) # ***
# 将整个模型,以及损失层的list全部返回
return model, style_losses, content_losses
再之后
# Parameter是需要grad的Variable!
# 这里是迭代优化输入x,因此将x作为参数,然后放进optim中。
def get_input_param_optimizer(input_img):
# this line to show that input is a parameter that requires a gradient
input_param = nn.Parameter(input_img.data)
optimizer = optim.LBFGS([input_param])
return input_param, optimizer
def run_style_transfer(cnn, content_img, style_img, input_img, num_steps=300,
style_weight=1000, content_weight=1):
"""Run the style transfer."""
print('Building the style transfer model..')
model, style_losses, content_losses = get_style_model_and_losses(cnn,
style_img, content_img, style_weight, content_weight)
input_param, optimizer = get_input_param_optimizer(input_img)
print('Optimizing..')
run = [0]
while run[0] <= num_steps:
def closure():
# correct the values of updated input image
input_param.data.clamp_(0, 1)
# 首先梯度置0
optimizer.zero_grad()
model(input_param)
style_score = 0
content_score = 0
# 这个有意思。直接调用假的“backward”
# 这个backward会调用隐藏层约束的loss的真的backward!
# 这个假的backward其实主要是返回相应的loss用的。
# 写法非常巧妙。
for sl in style_losses:
style_score += sl.backward()
# content层与style层甚至可以分开来backward!
for cl in content_losses:
content_score += cl.backward()
run[0] += 1
if run[0] % 50 == 0:
print("run {}:".format(run))
print('Style Loss : {:4f} Content Loss: {:4f}'.format(
style_score.data[0], content_score.data[0]))
print()
return style_score + content_score
optimizer.step(closure)
# a last correction...
input_param.data.clamp_(0, 1)
return input_param.data
output = run_style_transfer(cnn, content_img, style_img, input_img)
plt.figure()
imshow(output, title='Output Image')
# sphinx_gallery_thumbnail_number = 4
plt.ioff()
plt.show()
很方便!下面代码展示了自定义层竟然可以分开进行backward!一点都不影响!
for sl in style_losses:
style_score += sl.backward()
for cl in content_losses:
content_score += cl.backward()
其实很简单,backward会计算相应的网络结点的梯度,如果梯度不置0,那么这些梯度是不断累加的。不过pytorch为了节省空间,除了叶子结点(自己创建的Variable)之外的结点的梯度,一旦计算完,就会清空!所以你一般想看中间层的梯度是看不到的!除非添加hook!那么我们就不能保留这些中间结点的梯度吗?用retain_graph!
retain_graph就是用来保存计算反向时的graph的,当retain_graph为true时,那么就可以多次单独backward,而不怕上一次的梯度消失!这也就是为什么
def backward(self, retain_graph=True):
self.loss.backward(retain_graph=retain_graph)
return self.loss
另外一点是,这里有一个detach。这是很重要的。因为这里的target是用style图传入网络中得到的。它是一个Variable!有着自己的计算图。detach()的作用就是将这个结点“截断”,使得其变成叶子节点,就好像是我们自己创建的一个结点,这样的话target.grad_fn
变成None,梯度传到它就不会往前进行。
def __init__(self, target, weight):
super(StyleLoss, self).__init__()
self.target = target.detach() * weight
self.weight = weight
self.gram = GramMatrix()
self.criterion = nn.MSELoss()
retain_graph
为true。这也为了“多个loss单独backward”成为可能。