建设网站软件,天翼云wordpress插件,低代码开发平台开源,深圳龙岗发布通告本文与《20天吃透Pytorch》有所不同#xff0c;《20天吃透Pytorch》中是继承之前的模型进行拟合#xff0c;本文是单独建立网络进行拟合。
代码实现#xff1a;
import torch
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from torch import …本文与《20天吃透Pytorch》有所不同《20天吃透Pytorch》中是继承之前的模型进行拟合本文是单独建立网络进行拟合。
代码实现
import torch
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from torch import nn
import torch.nn.functional as F
from torch.utils.data import Dataset,DataLoader,TensorDataset
1.准备数据n800 #样本数量#生成测试用的数据集
X 10*torch.rand([n,2])-5.0 #torch.rand是均匀分布
w0 torch.tensor([[2.0],[-3.0]])
b0 torch.tensor([10.0])
Y Xw0 b0 torch.normal(0.0,2.0,size[n,1]) ## 表示矩阵乘法,增加正态扰动#数据可视化
plt.figure(figsize (12,5))
ax1 plt.subplot(121)
ax1.scatter(X[:,0],Y[:,0],c b,label samples)
ax1.legend() #图例
plt.xlabel(x1)
plt.ylabel(y,rotation 0)
ax2 plt.subplot(122)
ax2.scatter(X[:,1],Y[:,0],c g,label samples)
ax2.legend()
plt.xlabel(x2)
plt.ylabel(y,rotation 0)
plt.show()
构建通道
ds TensorDataset(X,Y)
ds_train,ds_valid torch.utils.data.random_split(ds,[int (n*0.7),n-int(n*0.7)]) #选取总样本的70%为训练数据
dl_train DataLoader(ds_train,batch_size10,shuffleTrue)
dl_valid DataLoader(ds_valid,batch_size10,shuffleTrue)
2.定义模型
class LinearRegression(torch.nn.Module):def __init__(self):super(LinearRegression, self).__init__()self.fc nn.Linear(2,1)def forward(self,x):x self.fc(x)return xnet LinearRegression()3.训练模型loss_func torch.nn.MSELoss()
optimizer torch.optim.Adam(net.parameters(),lr 0.01)eporchs 10
log_step_freq 20for eporch in range(1,eporchs1):net.train()loss_sum 0.0metric_sum 0.0step 1for step,(features,labels) in enumerate(dl_train,1):predictions net(features)loss loss_func(predictions,labels)optimizer.zero_grad()loss.backward()optimizer.step()w net.state_dict()[fc.weight]b net.state_dict()[fc.bias]print(step , step, loss , loss)print(w , w)print(b , b)loss_sum loss.item()
结果可视化w,b net.state_dict()[fc.weight],net.state_dict()[fc.bias]plt.figure(figsize (12,5))
ax1 plt.subplot(121)
ax1.scatter(X[:,0],Y[:,0], c b,label samples)
ax1.plot(X[:,0],w[0,0]*X[:,0]b[0],-r,linewidth 5.0,label model)
ax1.legend()
plt.xlabel(x1)
plt.ylabel(y,rotation 0)ax2 plt.subplot(122)
ax2.scatter(X[:,1],Y[:,0], c g,label samples)
ax2.plot(X[:,1],w[0,1]*X[:,1]b[0],-r,linewidth 5.0,label model)
ax2.legend()
plt.xlabel(x2)
plt.ylabel(y,rotation 0)plt.show()
结果展示
数据部分 线性回归结果
