Pytorch学习——用神经网络进行气温预测

一、导入需要用到的库

import numpy as np
import pandas as pd 
import matplotlib.pyplot as plt
import torch
import torch.optim as optim
import warnings
warnings.filterwarnings(\"ignore\")
%matplotlib inline

二、数据查看

features = pd.read_csv(\"D:/DataSet/temps.csv\")
features.head()

表格标签描述：

print(\'数据维度：\',features.shape)
#数据维度： (348, 9)

三、数据预处理

1. 提取出标准的时间数据格式：

#处理时间数据
import datetime

# 分别得到年，月，日
years = features[\'year\']
months = features[\'month\']
days = features[\'day\']

# datetime格式
dates = [str(int(year)) + \'-\' + str(int(month)) + \'-\' + str(int(day)) for year, month, day in zip(years, months, days)]
dates = [datetime.datetime.strptime(date, \'%Y-%m-%d\') for date in dates]

1. 特征数字化（week非数字），用独热编码（One-Hot Encoding），把 week 这一列数据进行转换。
   pd.get_dummies()

# 独热编码
features = pd.get_dummies(features)
features.head(5)

1. 对数据标签进行处理

# 标签
labels = np.array(features[\'actual\'])

# 在特征中去掉标签
features= features.drop(\'actual\', axis = 1)

# 名字单独保存一下，以备后患
feature_list = list(features.columns)

# 转换成合适的格式
features = np.array(features)

features.shape

1. 数据标准化

from sklearn import preprocessing
input_features = preprocessing.StandardScaler().fit_transform(features)

四、构建模型

1. 从头到尾自己搭建一个模型

#x 和 y转换成想要的tensor格式
x = torch.tensor(input_features,dtype=float)
y = torch.tensor(labels,dtype=float)

#权重参数初始化
weights = torch.randn((14,128),dtype = float, requires_grad = True)
bases = torch.randn(128,dtype = float, requires_grad = True)
weights2 = torch.randn((128,1),dtype = float, requires_grad = True)
bases2 = torch.randn(1,dtype = float, requires_grad = True)

#学习率 损失
learning_rate = 0.001
losses = []

for i in range(1000):
    #计算隐层
    hidden = x.mm(weights)+bases
    #加入激活函数（除了最后一层/输出层，连上带权重参数的层（基本神经网络的全连接层和卷积神经网络的卷积层）都需要在结果之后 加上激活函数）
    hidden = torch.relu(hidden) #得到隐层
    #预测结果
    predictions = hidden.mm(weights2) + bases2
    #-------------------------------------------------------以上为前向传播计算
    #计算损失
    loss = torch.mean((predictions-y)**2)
    losses.append(loss.data.numpy())
    
    #打印损失值
    if i%100 == 0:
        print(\'loss:\',loss)
    #反向传播计算
    loss.backward()
    
    #更新参数 梯度下降
    weights.data.add_(- learning_rate * weights.grad.data)
    bases.data.add_(- learning_rate * bases.grad.data)
    weights2.data.add_(- learning_rate * weights2.grad.data)
    bases2.data.add_(- learning_rate * bases2.grad.data)
    
    #每次迭代都要记得清空 否则梯度会累加
    weights.grad.data.zero_()
    bases.grad.data.zero_()
    weights2.grad.data.zero_()
    bases2.grad.data.zero_()

1. 简单方法构建网络 直接调包

#定义神经网络
input_size = input_features.shape[1]
hidden_size = 128
output_size = 1
batch_size = 16
my_nn = torch.nn.Sequential(
    torch.nn.Linear(input_size,hidden_size),
    torch.nn.Sigmoid(),
    torch.nn.Linear(hidden_size,output_size)
)
cost = torch.nn.MSELoss(reduction = \'mean\') #损失函数
optimizer = torch.optim.Adam(my_nn.parameters(),lr=0.001) #优化器 Adam/SGD

#训练网络
losses = []
for i in range(1000):
    batch_loss = []
    #MINI-Batch方法来进行训练 每次取16个数据
    for start in range(0,len(input_features),batch_size):
        end = start + batch_size if start + batch_size < len(input_features) else len(input_features)
        xx = torch.tensor(input_features[start:end],dtype=torch.float,requires_grad = True)
        yy = torch.tensor(labels[start:end],dtype=torch.float,requires_grad = True)
        prediction = my_nn(xx)
        loss = cost(prediction,yy)
        optimizer.zero_grad() #每次梯度清零
        loss.backward(retain_graph=True)
        optimizer.step()#实时更新
        batch_loss.append(loss.data.numpy())
    #打印损失 
    if i%100==0:
        losses.append(np.mean(batch_loss))
        print(i,np.mean(batch_loss))

五、预测结果

x = torch.tensor(input_features,dtype = torch.float) #转成tensor格式
predict = my_nn(x).data.numpy()  #转成nmupy格式，用matlib工具包画图

# 转换日期格式
dates = [str(int(year)) + \'-\' + str(int(month)) + \'-\' + str(int(day)) for year, month, day in zip(years, months, days)]
dates = [datetime.datetime.strptime(date, \'%Y-%m-%d\') for date in dates]

# 创建一个表格来存日期和其对应的标签数值
true_data = pd.DataFrame(data = {\'date\': dates, \'actual\': labels})

# 同理，再创建一个来存日期和其对应的模型预测值
months = features[:, feature_list.index(\'month\')]
days = features[:, feature_list.index(\'day\')]
years = features[:, feature_list.index(\'year\')]

test_dates = [str(int(year)) + \'-\' + str(int(month)) + \'-\' + str(int(day)) for year, month, day in zip(years, months, days)]

test_dates = [datetime.datetime.strptime(date, \'%Y-%m-%d\') for date in test_dates]

predictions_data = pd.DataFrame(data = {\'date\': test_dates, \'prediction\': predict.reshape(-1)}) 

# 真实值
plt.plot(true_data[\'date\'], true_data[\'actual\'], \'b-\', label = \'actual\')

# 预测值
plt.plot(predictions_data[\'date\'], predictions_data[\'prediction\'], \'ro\', label = \'prediction\')
plt.xticks(rotation = \'60\'); 
plt.legend()

# 图名
plt.xlabel(\'Date\'); plt.ylabel(\'Maximum Temperature (F)\');
plt.title(\'Actual and Predicted Values\');

最终预测结果可视化如下：

红色为训练出来的回归曲线，较为拟合。