利用pytorch两层线性网络对titanic数据集进行分类（kaggle)

利用pytorch两层线性网络对titanic数据集进行分类

最近在看pytorch的入门课程，做了一下在kaggle网站上的作业，用的是titanic数据集，因为想搭一下神经网络，所以数据加载部分简单的把训练集和测试集中有缺失值的列还有含有字符串的列去除了，加入了DataLoader模块，其实这个数据集很小，用不到，本人还没入门，小白一枚。

python 复制代码

import torch 
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
import numpy as np
from torchvision import datasets
from torchvision import transforms
import pandas as pd

class titanicDataset(Dataset):
    def __init__(self,filepath):
        xy=np.loadtxt(filepath,delimiter=',',skiprows=1,usecols=[1,2,7,8],dtype=np.float32)
        self.len=xy.shape[0]
        # print(self.len)
        self.y_data=torch.from_numpy(xy[:,[0]])
        self.x_data=torch.from_numpy(xy[:,1:])
        
    def __getitem__(self,index):#获取索引元素 
        return self.x_data[index],self.y_data[index]
    def __len__(self):
        return self.len
dataset=titanicDataset('./pytorch/dataset/titanic/train.csv')
train_loader=DataLoader(dataset=dataset,batch_size=32,shuffle=True,num_workers=0)

# print(dataset.x_data,dataset.y_data)
test_loader=DataLoader(dataset=np.loadtxt('./pytorch/dataset/titanic/test.csv',delimiter=',',skiprows=1,usecols=[1,6,7],dtype=np.float32),batch_size=32,shuffle=False,num_workers=0)
print(next(iter(test_loader)))

class Model(torch.nn.Module):
    def __init__(self):
        super(Model,self).__init__()
        # self.linear1=torch.nn.Linear(4,3)
        self.linear2=torch.nn.Linear(3,2)
        self.linear3=torch.nn.Linear(2,1)
        self.sigmoid=torch.nn.Sigmoid()
    def forward(self,x):
        # x=self.sigmoid(self.linear1(x))
        x=self.sigmoid(self.linear2(x))
        x=self.sigmoid(self.linear3(x))
        return x
model=Model()
criterion=torch.nn.BCELoss(size_average=True)
optimizer=torch.optim.SGD(model.parameters(),lr=0.1,momentum=0.9)
for epoch in range(10000):
    acc_num=0
    for i,data in enumerate(train_loader,0):
        #1.Prepare data
        inputs,labels=data
        # print(inputs.shape[0])
        #2.Forward
        y_pred=model(inputs)
        loss=criterion(y_pred,labels)
        # print(epoch,i,loss.item())
        #3.Backward
        optimizer.zero_grad()
        loss.backward()
        #4.Update
        optimizer.step()
        y_pred_label=torch.where(y_pred>0.5,torch.tensor([1.0]),torch.tensor([0.0]))
        acc_num+=torch.eq(y_pred_label,labels).sum().item()
    # print(acc_num,len(dataset),len(train_loader.dataset))
    acc=acc_num/len(dataset)
print(acc)
# print(test_loader)
# print(test_loader.dataset.shape)
out = model(torch.tensor(test_loader.dataset))
y_pred = torch.where(out>0.5,torch.tensor([1.0]),torch.tensor([0.0]))[:,0]
print(y_pred)
print(pd.Series(y_pred))
id=pd.read_csv('./pytorch/dataset/titanic/test.csv',usecols=['PassengerId']).iloc[:,0]
# print(type(id))

pd.DataFrame({'PassengerId':id,'Survived':pd.Series(y_pred,dtype=int)}).to_csv('pred.csv',index=None)
a=pd.DataFrame([id,pd.Series(y_pred)])
print(a)
# print(y_pred[-10:])


# for x in test_loader:
#     print(x.shape)
#     out = model(x)
#     y_pred = torch.where(out>0.5,torch.tensor([1.0]),torch.tensor([0.0]))
# print(y_pred)