一、Pytorch的建模流程

1.准备数据（难点。我的数据类型：图片数据）
2.定义模型
3.训练模型
4.评估模型
5.使用模型
6.保存模型

图片数据建模流程范例

1.准备数据

cifar2数据集:
cifar10数据集的子集，只包括前两种类别airplane和automobile
训练集：
airplane和automobile图片各5000张
测试集：
airplane和automobile图片各1000张
任务的目标：
训练一个模型，来对飞机airplane和机动车automobile两种图片进行分类
文件结构：
-cifar2
-train
-0_airplane
-…（5000）
-1_automobile
-…(5000)
-test
-0_airplane
-…（1000）
-1_automobile
-…（1000）
在Pytorch中构建图片数据管道通常有两种方法。
第一种：
使用 torchvision中的datasets.ImageFolder来读取图片然后用 DataLoader来并行加载
第二种：（书中没写。自行查找）
通过继承 torch.utils.data.Dataset 实现用户自定义读取逻辑然后用 DataLoader来并行加载
（读取用户自定义数据集的通用方法。我应该更常用）

二，定义模型

使用Pytorch通常有三种方式构建模型：
使用nn.Sequential按层顺序构建模型
继承nn.Module基类构建自定义模型（我应该常用。也是书中这里用到的）
继承nn.Module基类构建模型并辅助应用模型容器 (nn.Sequential,nn.ModuleList,nn.ModuleDict)进行封装

class Net(nn.Module):def __init__(self):super(Net, self).__init__()self.conv1 = nn.Conv2d(in_channels=3,out_channels=32,kernel_size = 3)self.pool = nn.MaxPool2d(kernel_size = 2,stride = 2)self.conv2 = nn.Conv2d(in_channels=32,out_channels=64,kernel_size = 5)self.dropout = nn.Dropout2d(p = 0.1)self.adaptive_pool = nn.AdaptiveMaxPool2d((1,1))self.flatten = nn.Flatten()self.linear1 = nn.Linear(64,32)self.relu = nn.ReLU()self.linear2 = nn.Linear(32,1)self.sigmoid = nn.Sigmoid()def forward(self,x):x = self.conv1(x)x = self.pool(x)x = self.conv2(x)x = self.pool(x)x = self.dropout(x)x = self.adaptive_pool(x)x = self.flatten(x)x = self.linear1(x)x = self.relu(x)x = self.linear2(x)y = self.sigmoid(x)return ynet = Net()
print(net)

Net((conv1): Conv2d(3, 32, kernel_size=(3, 3), stride=(1, 1))(pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(conv2): Conv2d(32, 64, kernel_size=(5, 5), stride=(1, 1))(dropout): Dropout2d(p=0.1, inplace=False)(adaptive_pool): AdaptiveMaxPool2d(output_size=(1, 1))(flatten): Flatten()(linear1): Linear(in_features=64, out_features=32, bias=True)(relu): ReLU()(linear2): Linear(in_features=32, out_features=1, bias=True)(sigmoid): Sigmoid()
)

torchkeras.summary讲解

import torchkeras
torchkeras.summary(net,input_shape= (3,32,32))

打印网络结构：
观察网络的每个层是什么操作、输出维度、模型的总参数量、训练的参数量、网络的占用内存情况

----------------------------------------------------------------Layer (type)               Output Shape         Param #
================================================================Conv2d-1           [-1, 32, 30, 30]             896MaxPool2d-2           [-1, 32, 15, 15]               0Conv2d-3           [-1, 64, 11, 11]          51,264MaxPool2d-4             [-1, 64, 5, 5]               0Dropout2d-5             [-1, 64, 5, 5]               0AdaptiveMaxPool2d-6             [-1, 64, 1, 1]               0Flatten-7                   [-1, 64]               0Linear-8                   [-1, 32]           2,080ReLU-9                   [-1, 32]               0Linear-10                    [-1, 1]              33Sigmoid-11                    [-1, 1]               0
================================================================
Total params: 54,273
Trainable params: 54,273
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.011719
Forward/backward pass size (MB): 0.359634
Params size (MB): 0.207035
Estimated Total Size (MB): 0.578388
----------------------------------------------------------------

三，训练模型

Pytorch通常需要用户编写自定义训练循环，训练循环的代码风格因人而异。

有3类典型的训练循环代码风格：
脚本形式训练循环
函数形式训练循环（较通用。书中用的方法）
类形式训练循环

// An highlighted block
import pandas as pd 
from sklearn.metrics import roc_auc_scoremodel = net
model.optimizer = torch.optim.SGD(model.parameters(),lr = 0.01)
model.loss_func = torch.nn.BCELoss()
model.metric_func = lambda y_pred,y_true: roc_auc_score(y_true.data.numpy(),y_pred.data.numpy())
model.metric_name = "auc"

// An highlighted block
var foo = 'bar';

def train_step(model,features,labels):# 训练模式，dropout层发生作用model.train()# 梯度清零model.optimizer.zero_grad()# 正向传播求损失predictions = model(features)loss = model.loss_func(predictions,labels)metric = model.metric_func(predictions,labels)# 反向传播求梯度loss.backward()model.optimizer.step()return loss.item(),metric.item()def valid_step(model,features,labels):# 预测模式，dropout层不发生作用model.eval()predictions = model(features)loss = model.loss_func(predictions,labels)metric = model.metric_func(predictions,labels)return loss.item(), metric.item()# 测试train_step效果
features,labels = next(iter(dl_train))
train_step(model,features,labels)

(0.6922046542167664, 0.5088566827697262)

如何记录日志，打印日志

def train_model(model,epochs,dl_train,dl_valid,log_step_freq):metric_name = model.metric_namedfhistory = pd.DataFrame(columns = ["epoch","loss",metric_name,"val_loss","val_"+metric_name]) print("Start Training...")nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')print("=========="*8 + "%s"%nowtime)for epoch in range(1,epochs+1):  # 1，训练循环-------------------------------------------------loss_sum = 0.0metric_sum = 0.0step = 1for step, (features,labels) in enumerate(dl_train, 1):loss,metric = train_step(model,features,labels)# 打印batch级别日志loss_sum += lossmetric_sum += metricif step%log_step_freq == 0:   print(("[step = %d] loss: %.3f, "+metric_name+": %.3f") %(step, loss_sum/step, metric_sum/step))# 2，验证循环-------------------------------------------------val_loss_sum = 0.0val_metric_sum = 0.0val_step = 1for val_step, (features,labels) in enumerate(dl_valid, 1):val_loss,val_metric = valid_step(model,features,labels)val_loss_sum += val_lossval_metric_sum += val_metric# 3，记录日志-------------------------------------------------info = (epoch, loss_sum/step, metric_sum/step, val_loss_sum/val_step, val_metric_sum/val_step)dfhistory.loc[epoch-1] = info# 打印epoch级别日志print(("\nEPOCH = %d, loss = %.3f,"+ metric_name + \"  = %.3f, val_loss = %.3f, "+"val_"+ metric_name+" = %.3f") %info)nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')print("\n"+"=========="*8 + "%s"%nowtime)print('Finished Training...')return dfhistory

epochs = 20dfhistory = train_model(model,epochs,dl_train,dl_valid,log_step_freq = 50)

Start Training...
================================================================================2020-06-28 20:47:56
[step = 50] loss: 0.691, auc: 0.627
[step = 100] loss: 0.690, auc: 0.673
[step = 150] loss: 0.688, auc: 0.699
[step = 200] loss: 0.686, auc: 0.716EPOCH = 1, loss = 0.686,auc  = 0.716, val_loss = 0.678, val_auc = 0.806================================================================================2020-06-28 20:48:18
[step = 50] loss: 0.677, auc: 0.780
[step = 100] loss: 0.675, auc: 0.775
[step = 150] loss: 0.672, auc: 0.782
[step = 200] loss: 0.669, auc: 0.779EPOCH = 2, loss = 0.669,auc  = 0.779, val_loss = 0.651, val_auc = 0.815......================================================================================2020-06-28 20:54:24
[step = 50] loss: 0.386, auc: 0.914
[step = 100] loss: 0.392, auc: 0.913
[step = 150] loss: 0.395, auc: 0.911
[step = 200] loss: 0.398, auc: 0.911EPOCH = 19, loss = 0.398,auc  = 0.911, val_loss = 0.449, val_auc = 0.924================================================================================2020-06-28 20:54:43
[step = 50] loss: 0.416, auc: 0.917
[step = 100] loss: 0.417, auc: 0.916
[step = 150] loss: 0.404, auc: 0.918
[step = 200] loss: 0.402, auc: 0.918EPOCH = 20, loss = 0.402,auc  = 0.918, val_loss = 0.535, val_auc = 0.925================================================================================2020-06-28 20:55:03
Finished Training...

四，评估模型

Pandas 库之 DataFrame

dfhistory = pd.DataFrame(columns = ["epoch","loss",metric_name,"val_loss","val_"+metric_name]) 

dfhistory  

%matplotlib inline
%config InlineBackend.figure_format = 'svg'import matplotlib.pyplot as pltdef plot_metric(dfhistory, metric):train_metrics = dfhistory[metric]val_metrics = dfhistory['val_'+metric]epochs = range(1, len(train_metrics) + 1)plt.plot(epochs, train_metrics, 'bo--')plt.plot(epochs, val_metrics, 'ro-')plt.title('Training and validation '+ metric)plt.xlabel("Epochs")plt.ylabel(metric)plt.legend(["train_"+metric, 'val_'+metric])plt.show()

plot_metric(dfhistory,"loss")

plot_metric(dfhistory,"auc")

五，使用模型

def predict(model,dl):model.eval()result = torch.cat([model.forward(t[0]) for t in dl])return(result.data)

#预测概率
y_pred_probs = predict(model,dl_valid)
y_pred_probs

tensor([[8.4032e-01],[1.0407e-02],[5.4146e-04],...,[1.4471e-02],[1.7673e-02],[4.5081e-01]])

#预测类别
y_pred = torch.where(y_pred_probs>0.5,torch.ones_like(y_pred_probs),torch.zeros_like(y_pred_probs))
y_pred

tensor([[1.],[0.],[0.],...,[0.],[0.],[0.]])

六，保存模型

推荐使用保存参数方式保存Pytorch模型

torch.nn.Module模块中的state_dict变量存放训练过程中需要学习的权重和偏执系数
state_dict本质上Python字典对象，所以可以很好地进行保存、更新、修改和恢复操作（python字典结构的特性），从而为PyTorch模型和优化器增加了大量的模块化
state_dict变量

print(model.state_dict().keys())

odict_keys(['conv1.weight', 'conv1.bias', 'conv2.weight', 'conv2.bias', 'linear1.weight', 'linear1.bias', 'linear2.weight', 'linear2.bias'])

# 保存模型参数torch.save(model.state_dict(), "./data/model_parameter.pkl")net_clone = Net()
net_clone.load_state_dict(torch.load("./data/model_parameter.pkl"))predict(net_clone,dl_valid)

tensor([[0.0204],[0.7692],[0.4967],...,[0.6078],[0.7182],[0.8251]])