#-------------------------------conditional GAN-----------------
###################################################################################################################################
參考百度百科
https://baike.baidu.com/item/Gan/22181905?fr=aladdin
GAN 基本簡介
模型通過框架中（至少）兩個模塊：生成模型（Generative Model）和判別模型（Discriminative Model）的互相博弈學習產生相當好的輸出。原始 GAN 理論中，並不要求 G 和 D 都是神經網絡，只需要是能擬合相應生成和判別的函數即可。但實用中一般均使用深度神經網絡作爲 G 和 D 。一個優秀的GAN應用需要有良好的訓練方法，否則可能由於神經網絡模型的自由性而導致輸出不理想。
機器學習的模型可大體分爲兩類，生成模型（Generative Model）和判別模型（Discriminative Model）。判別模型需要輸入變量 ，通過某種模型來預測 。生成模型是給定某種隱含信息，來隨機產生觀測數據。舉個簡單的例子，
- 判別模型：給定一張圖，判斷這張圖裏的動物是貓還是狗
- 生成模型：給一系列貓的圖片，生成一張新的貓咪（不在數據集裏）
對於判別模型，損失函數是容易定義的，因爲輸出的目標相對簡單。但對於生成模型，損失函數的定義就不是那麼容易。我們對於生成結果的期望，往往是一個曖昧不清，難以數學公理化定義的範式。所以不妨把生成模型的回饋部分，交給判別模型處理。這就是Goodfellow他將機器學習中的兩大類模型，Generative和Discrimitive給緊密地聯合在了一起 。
GAN的基本原理其實非常簡單，這裏以生成圖片爲例進行說明。假設我們有兩個網絡，G（Generator）和D（Discriminator）。正如它的名字所暗示的那樣，它們的功能分別是：
- G是一個生成圖片的網絡，它接收一個隨機的噪聲z，通過這個噪聲生成圖片，記做G(z)。
- D是一個判別網絡，判別一張圖片是不是“真實的”。它的輸入參數是x，x代表一張圖片，輸出D（x）代表x爲真實圖片的概率，如果爲1，就代表100%是真實的圖片，而輸出爲0，就代表不可能是真實的圖片。
在訓練過程中，生成網絡G的目標就是儘量生成真實的圖片去欺騙判別網絡D。而D的目標就是儘量把G生成的圖片和真實的圖片分別開來。這樣，G和D構成了一個動態的“博弈過程”。
最後博弈的結果是什麼？在最理想的狀態下，G可以生成足以“以假亂真”的圖片G(z)。對於D來說，它難以判定G生成的圖片究竟是不是真實的，因此D(G(z)) = 0.5。
這樣我們的目的就達成了：我們得到了一個生成式的模型G，它可以用來生成圖片。
主要應用：
圖像生成，如超分辨率任務，語義分割 圖像翻譯，卡通人物生成，人體姿態生成，年齡變換，風格變換
圖像增強
…….
GAN 損失函數的理解
https://www.jianshu.com/p/588318e69eae
https://www.cnblogs.com/walter-xh/p/10051634.html
主要做到minG和maxD。
相互博弈產生良性結果
完整代碼：
在上下區間中，迭代10000次，通過輸入隨機點，在上下藍紅曲線中，生成曲線
import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt
# torch.manual_seed(1) # reproducible
# np.random.seed(1)
# Hyper Parameters
BATCH_SIZE = 64
LR_G = 0.0001 # learning rate for generator
LR_D = 0.0001 # learning rate for discriminator
N_IDEAS = 5 # think of this as number of ideas for generating an art work (Generator)
ART_COMPONENTS = 15 # it could be total point G can draw in the canvas
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONENTS) for _ in range(BATCH_SIZE)])
# show our beautiful painting range
plt.plot(PAINT_POINTS[0], 2 * np.power(PAINT_POINTS[0], 2) + 1, c='#74BCFF', lw=3, label='upper bound')
plt.plot(PAINT_POINTS[0], 1 * np.power(PAINT_POINTS[0], 2) + 0, c='#FF9359', lw=3, label='lower bound')
plt.legend(loc='upper right')
plt.show()
def artist_works_with_labels(): # painting from the famous artist (real target)
a = np.random.uniform(1, 2, size=BATCH_SIZE)[:, np.newaxis]
paintings = a * np.power(PAINT_POINTS, 2) + (a-1)
labels = (a-1) > 0.5 # upper paintings (1), lower paintings (0), two classes
paintings = torch.from_numpy(paintings).float()
labels = torch.from_numpy(labels.astype(np.float32))
return paintings, labels
G = nn.Sequential( # Generator
nn.Linear(N_IDEAS+1, 128), # random ideas (could from normal distribution) + class label
nn.ReLU(),
nn.Linear(128, ART_COMPONENTS), # making a painting from these random ideas
)
D = nn.Sequential( # Discriminator
nn.Linear(ART_COMPONENTS+1, 128), # receive art work either from the famous artist or a newbie like G with label
nn.ReLU(),
nn.Linear(128, 1),
nn.Sigmoid(), # tell the probability that the art work is made by artist
)
opt_D = torch.optim.Adam(D.parameters(), lr=LR_D)
opt_G = torch.optim.Adam(G.parameters(), lr=LR_G)
plt.ion() # something about continuous plotting
for step in range(10000):
artist_paintings, labels = artist_works_with_labels() # real painting, label from artist 真實的數據和標籤
G_ideas = torch.randn(BATCH_SIZE, N_IDEAS) # random ideas 隨機在縱座標生成BATCH_SIZE個數據 在橫座標生成N_IDEAS個數據
G_inputs = torch.cat((G_ideas, labels), 1) # ideas with labels 將生成的數據和真實標籤按行拼接
G_paintings = G(G_inputs) # fake painting w.r.t label from G 將假的拼接好的數據 輸入到生成器
D_inputs0 = torch.cat((artist_paintings, labels), 1) # all have their labels 真實的數據和標籤
D_inputs1 = torch.cat((G_paintings, labels), 1) # 生成的數據和標籤
prob_artist0 = D(D_inputs0) # D try to increase this prob 真實數據和標籤輸入到判別器
prob_artist1 = D(D_inputs1) # D try to reduce this prob 生成的數據和標籤輸入到判別器
D_score0 = torch.log(prob_artist0) # maximise this for D 真實的數據的判別器的得分
D_score1 = torch.log(1. - prob_artist1) # maximise this for D 生成的數據的判別器的得分
D_loss = - torch.mean(D_score0 + D_score1) # minimise the negative of both two above for D 定義三重態損失函數 爲了使損失函數小，就迫使prob_artist0 越大，prob_artist1越小
G_loss = torch.mean(D_score1) # minimise D score w.r.t G 生成器損失函數 只對生成的數據
opt_D.zero_grad()
D_loss.backward(retain_graph=True) # reusing computational graph 保留參數 留給下次反向傳播
opt_D.step()
opt_G.zero_grad()
G_loss.backward()
opt_G.step()
if step % 200 == 0: # plotting
plt.cla()
plt.plot(PAINT_POINTS[0], G_paintings.data.numpy()[0], c='#4AD631', lw=3, label='Generated painting',)
bound = [0, 0.5] if labels.data[0, 0] == 0 else [0.5, 1]
plt.plot(PAINT_POINTS[0], 2 * np.power(PAINT_POINTS[0], 2) + bound[1], c='#74BCFF', lw=3, label='upper bound')
plt.plot(PAINT_POINTS[0], 1 * np.power(PAINT_POINTS[0], 2) + bound[0], c='#FF9359', lw=3, label='lower bound')
plt.text(-.5, 2.3, 'D accuracy=%.2f (0.5 for D to converge)' % prob_artist0.data.numpy().mean(), fontdict={'size': 13})
plt.text(-.5, 2, 'D score= %.2f (-1.38 for G to converge)' % -D_loss.data.numpy(), fontdict={'size': 13})
plt.text(-.5, 1.7, 'Class = %i' % int(labels.data[0, 0]), fontdict={'size': 13})
plt.ylim((0, 3));plt.legend(loc='upper right', fontsize=10);plt.draw();plt.pause(0.1)
plt.ioff()
plt.show()
# plot a generated painting for upper class
z = torch.randn(1, N_IDEAS)
label = torch.FloatTensor([[1.]]) # for upper class
G_inputs = torch.cat((z, label), 1)
G_paintings = G(G_inputs)
plt.plot(PAINT_POINTS[0], G_paintings.data.numpy()[0], c='#4AD631', lw=3, label='G painting for upper class',)
plt.plot(PAINT_POINTS[0], 2 * np.power(PAINT_POINTS[0], 2) + bound[1], c='#74BCFF', lw=3, label='upper bound (class 1)')
plt.plot(PAINT_POINTS[0], 1 * np.power(PAINT_POINTS[0], 2) + bound[0], c='#FF9359', lw=3, label='lower bound (class 1)')
plt.ylim((0, 3));plt.legend(loc='upper right', fontsize=10);plt.show()
運行結果：
--------------------------------------爲什麼torch是動態的--------------------------------------------------
##############################################################################################################################################
與tensorflow 不同
Tensorflow 是靜態的編輯器，先建立好一個靜態的圖表，然後再把數據放到圖表中進行計算，然後再反向傳遞更新這些參數
因爲在某些實際情況可能batch_size 和 time_step 是不確定的
但是又不能兩個同時不確定 這時程序會報錯
例子使用RNN的time_step 隨機來實現
完整代碼：
import torch
from torch import nn
import numpy as np
import matplotlib.pyplot as plt
# torch.manual_seed(1) # reproducible
# Hyper Parameters
INPUT_SIZE = 1 # rnn input size / image width
LR = 0.02 # learning rate
class RNN(nn.Module):
def __init__(self):
super(RNN, self).__init__()
self.rnn = nn.RNN(
input_size=1,
hidden_size=32, # rnn hidden unit
num_layers=1, # number of rnn layer
batch_first=True, # input & output will has batch size as 1s dimension. e.g. (batch, time_step, input_size)
)
self.out = nn.Linear(32, 1)
def forward(self, x, h_state):
# x (batch, time_step, input_size)
# h_state (n_layers, batch, hidden_size)
# r_out (batch, time_step, output_size)
r_out, h_state = self.rnn(x, h_state)
outs = [] # this is where you can find torch is dynamic
for time_step in range(r_out.size(1)): # calculate output for each time step
outs.append(self.out(r_out[:, time_step, :]))
return torch.stack(outs, dim=1), h_state
rnn = RNN()
print(rnn)
optimizer = torch.optim.Adam(rnn.parameters(), lr=LR) # optimize all cnn parameters
loss_func = nn.MSELoss() # the target label is not one-hotted
h_state = None # for initial hidden state
plt.figure(1, figsize=(12, 5))
plt.ion() # continuously plot
######################## Below is different #########################
################ static time steps ##########
# for step in range(60):
# start, end = step * np.pi, (step+1)*np.pi # time steps
# # use sin predicts cos
# steps = np.linspace(start, end, 10, dtype=np.float32)
################ dynamic time steps #########
step = 0
for i in range(60):
dynamic_steps = np.random.randint(1, 4) # has random time steps 隨機步長 動態化 使每次輸入的步長都是不一樣的
start, end = step * np.pi, (step + dynamic_steps) * np.pi # different time steps length
step += dynamic_steps
# use sin predicts cos
steps = np.linspace(start, end, 10 * dynamic_steps, dtype=np.float32)
####################### Above is different ###########################
print(len(steps)) # print how many time step feed to RNN
x_np = np.sin(steps) # float32 for converting torch FloatTensor
y_np = np.cos(steps)
x = torch.from_numpy(x_np[np.newaxis, :, np.newaxis]) # shape (batch, time_step, input_size)
y = torch.from_numpy(y_np[np.newaxis, :, np.newaxis])
prediction, h_state = rnn(x, h_state) # rnn output
# !! next step is important !!
h_state = h_state.data # repack the hidden state, break the connection from last iteration
loss = loss_func(prediction, y) # cross entropy loss
optimizer.zero_grad() # clear gradients for this training step
loss.backward() # backpropagation, compute gradients
optimizer.step() # apply gradients
# plotting
plt.plot(steps, y_np.flatten(), 'r-')
plt.plot(steps, prediction.data.numpy().flatten(), 'b-')
plt.draw()
plt.pause(0.05)
plt.ioff()
plt.show()
運行結果：
---------------------------------------------GPU-----------------------------------------------------------------
###########################################################################################################################################
套用之前CNN實現數字識別的代碼
簡單加上幾個方法
對訓練數據 網絡模型等 加上.cuda() 的方法
import torch
import torch.nn as nn
import torch.utils.data as Data
import torchvision
# torch.manual_seed(1)
EPOCH = 1
BATCH_SIZE = 50
LR = 0.001
DOWNLOAD_MNIST = False
train_data = torchvision.datasets.MNIST(root='./mnist/', train=True, transform=torchvision.transforms.ToTensor(), download=DOWNLOAD_MNIST,)
train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True)
test_data = torchvision.datasets.MNIST(root='./mnist/', train=False)
# !!!!!!!! Change in here !!!!!!!!! #
test_x = torch.unsqueeze(test_data.test_data, dim=1).type(torch.FloatTensor)[:2000].cuda()/255. # Tensor on GPU
test_y = test_data.test_labels[:2000].cuda()
class CNN(nn.Module):
def __init__(self):
super(CNN, self).__init__()
self.conv1 = nn.Sequential(nn.Conv2d(in_channels=1, out_channels=16, kernel_size=5, stride=1, padding=2,),
nn.ReLU(), nn.MaxPool2d(kernel_size=2),)
self.conv2 = nn.Sequential(nn.Conv2d(16, 32, 5, 1, 2), nn.ReLU(), nn.MaxPool2d(2),)
self.out = nn.Linear(32 * 7 * 7, 10)
def forward(self, x):
x = self.conv1(x)
x = self.conv2(x)
x = x.view(x.size(0), -1)
output = self.out(x)
return output
cnn = CNN()
# !!!!!!!! Change in here !!!!!!!!! #
cnn.cuda() # Moves all model parameters and buffers to the GPU.
optimizer = torch.optim.Adam(cnn.parameters(), lr=LR)
loss_func = nn.CrossEntropyLoss()
for epoch in range(EPOCH):
for step, (x, y) in enumerate(train_loader):
# !!!!!!!! Change in here !!!!!!!!! #
b_x = x.cuda() # Tensor on GPU
b_y = y.cuda() # Tensor on GPU
output = cnn(b_x)
loss = loss_func(output, b_y)
optimizer.zero_grad()
loss.backward()
optimizer.step()
if step % 50 == 0:
test_output = cnn(test_x)
# !!!!!!!! Change in here !!!!!!!!! #
pred_y = torch.max(test_output, 1)[1].cuda().data # move the computation in GPU
accuracy = torch.sum(pred_y == test_y).type(torch.FloatTensor) / test_y.size(0)
print('Epoch: ', epoch, '| train loss: %.4f' % loss.data.cpu().numpy(), '| test accuracy: %.2f' % accuracy)
test_output = cnn(test_x[:10])
# !!!!!!!! Change in here !!!!!!!!! #
pred_y = torch.max(test_output, 1)[1].cuda().data # move the computation in GPU
print(pred_y, 'prediction number')
print(test_y[:10], 'real number')
運行結果 ：
在一定輪數的訓練之後 輸入預測值和真實值的比較
GPU加速後 肉眼可見的訓練速度加快
-------------------------------------------dropout-----------------------------------------------------------------
#############################################################################################################################################
torch.nn.Dropout(0.5),
每次隨機抽取一定百分比的神經元 去進行傳播
用訓練次數來彌補可能丟失的信息
訓練時需要dropout 但是測試時不需要
訓練時需要dropout 但是測試時不需要。
所以在測試之前需要把dropout屏蔽掉，如圖
完整代碼：
import torch
import matplotlib.pyplot as plt
# torch.manual_seed(1) # reproducible
N_SAMPLES = 20
N_HIDDEN = 300
# training data
x = torch.unsqueeze(torch.linspace(-1, 1, N_SAMPLES), 1)
y = x + 0.3*torch.normal(torch.zeros(N_SAMPLES, 1), torch.ones(N_SAMPLES, 1))
# test data
test_x = torch.unsqueeze(torch.linspace(-1, 1, N_SAMPLES), 1)
test_y = test_x + 0.3*torch.normal(torch.zeros(N_SAMPLES, 1), torch.ones(N_SAMPLES, 1))
# show data
plt.scatter(x.data.numpy(), y.data.numpy(), c='magenta', s=50, alpha=0.5, label='train')
plt.scatter(test_x.data.numpy(), test_y.data.numpy(), c='cyan', s=50, alpha=0.5, label='test')
plt.legend(loc='upper left')
plt.ylim((-2.5, 2.5))
plt.show()
net_overfitting = torch.nn.Sequential(
torch.nn.Linear(1, N_HIDDEN),
torch.nn.ReLU(),
torch.nn.Linear(N_HIDDEN, N_HIDDEN),
torch.nn.ReLU(),
torch.nn.Linear(N_HIDDEN, 1),
)
net_dropped = torch.nn.Sequential(
torch.nn.Linear(1, N_HIDDEN),
torch.nn.Dropout(0.5), # drop 50% of the neuron
torch.nn.ReLU(),
torch.nn.Linear(N_HIDDEN, N_HIDDEN),
torch.nn.Dropout(0.5), # drop 50% of the neuron
torch.nn.ReLU(),
torch.nn.Linear(N_HIDDEN, 1),
)
print(net_overfitting) # net architecture
print(net_dropped)
optimizer_ofit = torch.optim.Adam(net_overfitting.parameters(), lr=0.01)
optimizer_drop = torch.optim.Adam(net_dropped.parameters(), lr=0.01)
loss_func = torch.nn.MSELoss()
plt.ion() # something about plotting
for t in range(500):
pred_ofit = net_overfitting(x)
pred_drop = net_dropped(x)
loss_ofit = loss_func(pred_ofit, y)
loss_drop = loss_func(pred_drop, y)
optimizer_ofit.zero_grad()
optimizer_drop.zero_grad()
loss_ofit.backward()
loss_drop.backward()
optimizer_ofit.step()
optimizer_drop.step()
if t % 10 == 0:
# change to eval mode in order to fix drop out effect
net_overfitting.eval()
net_dropped.eval() # parameters for dropout differ from train mode
# plotting
plt.cla()
test_pred_ofit = net_overfitting(test_x)
test_pred_drop = net_dropped(test_x)
plt.scatter(x.data.numpy(), y.data.numpy(), c='magenta', s=50, alpha=0.3, label='train')
plt.scatter(test_x.data.numpy(), test_y.data.numpy(), c='cyan', s=50, alpha=0.3, label='test')
plt.plot(test_x.data.numpy(), test_pred_ofit.data.numpy(), 'r-', lw=3, label='overfitting')
plt.plot(test_x.data.numpy(), test_pred_drop.data.numpy(), 'b--', lw=3, label='dropout(50%)')
plt.text(0, -1.2, 'overfitting loss=%.4f' % loss_func(test_pred_ofit, test_y).data.numpy(), fontdict={'size': 20, 'color': 'red'})
plt.text(0, -1.5, 'dropout loss=%.4f' % loss_func(test_pred_drop, test_y).data.numpy(), fontdict={'size': 20, 'color': 'blue'})
plt.legend(loc='upper left'); plt.ylim((-2.5, 2.5));plt.pause(0.1)
# change back to train mode
net_overfitting.train()
net_dropped.train()
plt.ioff()
plt.show()
運行結果：
----------------------------------bach normalization-----------------------------------------------------
#############################################################################################################################################
每一個BN 都被添加在全連接層和激活函數之間
是處理數據的一種方式
對全連接層後輸出的數據進行處理，避免數據分佈在激活函數的無效區間，隨着訓練 神經網絡會死掉
比如tanh激活函數
經過BN層處理數據以後，用BN拉回到（0,1） 這樣就可以避免梯度爆炸和梯度消失
完整代碼：
爲了畫圖 代碼不是很好理解
import torch
from torch import nn
from torch.nn import init
import torch.utils.data as Data
import matplotlib.pyplot as plt
import numpy as np
# torch.manual_seed(1) # reproducible
# np.random.seed(1)
# Hyper parameters
N_SAMPLES = 2000
BATCH_SIZE = 64
EPOCH = 12
LR = 0.03
N_HIDDEN = 8
ACTIVATION = torch.tanh
B_INIT = -0.2 # use a bad bias constant initializer
# training data
x = np.linspace(-7, 10, N_SAMPLES)[:, np.newaxis]
noise = np.random.normal(0, 2, x.shape)
y = np.square(x) - 5 + noise
# test data
test_x = np.linspace(-7, 10, 200)[:, np.newaxis]
noise = np.random.normal(0, 2, test_x.shape)
test_y = np.square(test_x) - 5 + noise
train_x, train_y = torch.from_numpy(x).float(), torch.from_numpy(y).float()
test_x = torch.from_numpy(test_x).float()
test_y = torch.from_numpy(test_y).float()
train_dataset = Data.TensorDataset(train_x, train_y)
train_loader = Data.DataLoader(dataset=train_dataset, batch_size=BATCH_SIZE, shuffle=True, num_workers=2,)
# show data
plt.scatter(train_x.numpy(), train_y.numpy(), c='#FF9359', s=50, alpha=0.2, label='train')
plt.legend(loc='upper left')
class Net(nn.Module):
def __init__(self, batch_normalization=False):
super(Net, self).__init__()
self.do_bn = batch_normalization
self.fcs = []
self.bns = []
self.bn_input = nn.BatchNorm1d(1, momentum=0.5) # for input data
for i in range(N_HIDDEN): # build hidden layers and BN layers
input_size = 1 if i == 0 else 10
fc = nn.Linear(input_size, 10)
setattr(self, 'fc%i' % i, fc) # IMPORTANT set layer to the Module
self._set_init(fc) # parameters initialization
self.fcs.append(fc)
if self.do_bn:
bn = nn.BatchNorm1d(10, momentum=0.5)
setattr(self, 'bn%i' % i, bn) # IMPORTANT set layer to the Module
self.bns.append(bn)
self.predict = nn.Linear(10, 1) # output layer
self._set_init(self.predict) # parameters initialization
def _set_init(self, layer):
init.normal_(layer.weight, mean=0., std=.1)
init.constant_(layer.bias, B_INIT)
def forward(self, x):
pre_activation = [x]
if self.do_bn: x = self.bn_input(x) # input batch normalization
layer_input = [x]
for i in range(N_HIDDEN):
x = self.fcs[i](x)
pre_activation.append(x)
if self.do_bn: x = self.bns[i](x) # batch normalization
x = ACTIVATION(x)
layer_input.append(x)
out = self.predict(x)
return out, layer_input, pre_activation
nets = [Net(batch_normalization=False), Net(batch_normalization=True)]
# print(*nets) # print net architecture
opts = [torch.optim.Adam(net.parameters(), lr=LR) for net in nets]
loss_func = torch.nn.MSELoss()
def plot_histogram(l_in, l_in_bn, pre_ac, pre_ac_bn):
for i, (ax_pa, ax_pa_bn, ax, ax_bn) in enumerate(zip(axs[0, :], axs[1, :], axs[2, :], axs[3, :])):
[a.clear() for a in [ax_pa, ax_pa_bn, ax, ax_bn]]
if i == 0:
p_range = (-7, 10);the_range = (-7, 10)
else:
p_range = (-4, 4);the_range = (-1, 1)
ax_pa.set_title('L' + str(i))
ax_pa.hist(pre_ac[i].data.numpy().ravel(), bins=10, range=p_range, color='#FF9359', alpha=0.5);ax_pa_bn.hist(pre_ac_bn[i].data.numpy().ravel(), bins=10, range=p_range, color='#74BCFF', alpha=0.5)
ax.hist(l_in[i].data.numpy().ravel(), bins=10, range=the_range, color='#FF9359');ax_bn.hist(l_in_bn[i].data.numpy().ravel(), bins=10, range=the_range, color='#74BCFF')
for a in [ax_pa, ax, ax_pa_bn, ax_bn]: a.set_yticks(());a.set_xticks(())
ax_pa_bn.set_xticks(p_range);ax_bn.set_xticks(the_range)
axs[0, 0].set_ylabel('PreAct');axs[1, 0].set_ylabel('BN PreAct');axs[2, 0].set_ylabel('Act');axs[3, 0].set_ylabel('BN Act')
plt.pause(0.01)
if __name__ == "__main__":
f, axs = plt.subplots(4, N_HIDDEN + 1, figsize=(10, 5))
plt.ion() # something about plotting
plt.show()
# training
losses = [[], []] # recode loss for two networks
for epoch in range(EPOCH):
print('Epoch: ', epoch)
layer_inputs, pre_acts = [], []
for net, l in zip(nets, losses):
net.eval() # set eval mode to fix moving_mean and moving_var
pred, layer_input, pre_act = net(test_x)
l.append(loss_func(pred, test_y).data.item())
layer_inputs.append(layer_input)
pre_acts.append(pre_act)
net.train() # free moving_mean and moving_var
plot_histogram(*layer_inputs, *pre_acts) # plot histogram
for step, (b_x, b_y) in enumerate(train_loader):
for net, opt in zip(nets, opts): # train for each network
pred, _, _ = net(b_x)
loss = loss_func(pred, b_y)
opt.zero_grad()
loss.backward()
opt.step() # it will also learns the parameters in Batch Normalization
plt.ioff()
# plot training loss
plt.figure(2)
plt.plot(losses[0], c='#FF9359', lw=3, label='Original')
plt.plot(losses[1], c='#74BCFF', lw=3, label='Batch Normalization')
plt.xlabel('step');plt.ylabel('test loss');plt.ylim((0, 2000));plt.legend(loc='best')
# evaluation
# set net to eval mode to freeze the parameters in batch normalization layers
[net.eval() for net in nets] # set eval mode to fix moving_mean and moving_var
preds = [net(test_x)[0] for net in nets]
plt.figure(3)
plt.plot(test_x.data.numpy(), preds[0].data.numpy(), c='#FF9359', lw=4, label='Original')
plt.plot(test_x.data.numpy(), preds[1].data.numpy(), c='#74BCFF', lw=4, label='Batch Normalization')
plt.scatter(test_x.data.numpy(), test_y.data.numpy(), c='r', s=50, alpha=0.2, label='train')
plt.legend(loc='best')
plt.show()
運行結果：
