【机器学习实战】逻辑回归

from torch.utils.data import DataLoader
from torchvision import datasets
import torchvision
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report
import numpy as np

载入数据，设置训练集和测试集

transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.1307,),(0.3081,))])
train_dataset = datasets.MNIST(root='../dataset/mnist/',train=True,download=True,transform=transform)
test_dataset = datasets.MNIST(root='../dataset/mnist',train=False,download=True,transform=transform)
batch_size = len(train_dataset)
train_loader = DataLoader(dataset=train_dataset, batch_size=100, shuffle=True)
test_loader = DataLoader(dataset=test_dataset, batch_size=100, shuffle=True)
X_train,y_train = next(iter(train_loader))
X_test,y_test = next(iter(train_loader))
#? 打印前100张图片
images, labels= X_train[:100], y_train[:100] 
#? 使用images生成宽度为10张图的网格大小
img = torchvision.utils.make_grid(images, nrow=10)
#? cv2.imshow()的格式是(size1,size1,channels),而img的格式是(channels,size1,size1),
#? 所以需要使用.transpose()转换，将颜色通道数放至第三维
img = img.numpy().transpose(1,2,0)
print(images.shape)
print(labels.reshape(10,10))
print(img.shape)
plt.imshow(img)
plt.show()

转换tensor格式为array

X_train,y_train = X_train.cpu().numpy(),y_train.cpu().numpy() #? tensor转为array形式)
X_test,y_test = X_test.cpu().numpy(),y_test.cpu().numpy() #? tensor转为array形式)

修改训练集和测试集的形状

X_train = X_train.reshape(X_train.shape[0],784)
X_test = X_test.reshape(X_test.shape[0],784)

训练模型

model = LogisticRegression(solver='sag', max_iter=400) #? lbfgs：拟牛顿法
model.fit(X_train, y_train)
y_pred = model.predict(X_test)
print(classification_report(y_test, y_pred))
ones_col=[[1] for i in range(len(X_train))] #? 生成全为1的二维嵌套列表，即[[1],[1],...,[1]]
X_train = np.append(X_train,ones_col,axis=1)
x_train = np.mat(X_train)
X_test = np.append(X_test,ones_col,axis=1)
x_test = np.mat(X_test)
#? Mnsit有0-9十个标记，由于是二分类任务，所以可以将标记0的作为1，其余为0用于识别是否为0的任务
y_train=np.array([1 if y_train[i]==1 else 0 for i in range(len(y_train))])
y_test=np.array([1 if y_test[i]==1 else 0 for i in range(len(y_test))])
model = LogisticRegression(solver='sag', max_iter=100)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)
print(classification_report(y_test, y_pred))