刘二老师的《PyTorch深度学习实践》,第八讲,自己测试的代码如下:
因为loss一直不收敛,所以试着把epoch改为1000,但还是不收敛,在0.5和0.8间反复横跳
import numpy as np
import torch
import matplotlib.pyplot as plt
from torch.utils.data import Dataset, DataLoader
# Dataset 帮助构建数据集,使数据集可以被索引
# Dataset是抽象类,即不能实例化,只能被继承使用
# DataLoader 拿出一个Mini-Batch
'''
所有的样本都参与训练之后,称为一个epoch
每次训练时所用的样本数量称为Batch-Size
Iteration: batch分多少个,即内层的迭代,迭代了多少次
# 举例:10000个样本,batch-size为1000,则Iteration为10;
'''
'''
1、Prepare DataSet
'''
class DiabetesDataset(Dataset):
def __init__(self, filepath):
xy = np.loadtxt(filepath, delimiter=',', dtype=np.float32)
self.len = xy.shape[0] # xy的格式为: 样本数*(x特征数+y特征数)的矩阵
# xy(shape)[0]即可提取样本数
self.x_data = torch.from_numpy(xy[:, :-1])
self.y_data = torch.from_numpy(xy[:, [-1]])
# 魔法方法,使实例化后的对象支持下标索引
def __getitem__(self, index):
return self.x_data[index], self.y_data[index]
# 使实例化对象支持len(实例化对象)使用
def __len__(self):
return self.len
dataset = DiabetesDataset('diabetes.csv.gz')
train_loader = DataLoader(dataset=dataset, batch_size=32,shuffle=True, num_workers=2)
# shuffle,即打乱;num_workers,允许超线程,即并行化的程序,有两个
'''
2、Design model
'''
class Model(torch.nn.Module):
def __init__(self):
super(Model, self).__init__()
self.linear1 = torch.nn.Linear(8, 6)
self.linear2 = torch.nn.Linear(6, 4)
self.linear3 = torch.nn.Linear(4, 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()
'''
3、construct loss and optimizer
'''
criterion = torch.nn.BCELoss(reduction='mean')
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
Epoch_list =[] #保存epoch
Loss_list = [] #保存每个epoch对应的loss
'''
4、Training Cycle: forward --> backward --> update
'''
if __name__ == '__main__':# 没有这条语句,超线程的设置会报错
for epoch in range(1000):
for i, data in enumerate(train_loader, 0):
# 1.prepare data
inputs, labels = data
# 2.forward
y_pred = model(inputs)
loss = criterion(y_pred, labels)
print('Epoch:', epoch, "i = ", i, "loss = %5f" % loss.item())
Epoch_list.append(epoch)
Loss_list.append(loss)
# Backward
optimizer.zero_grad() # 梯度置为0
loss.backward() # 反向传播
# update
optimizer.step() # 更新w和b的值
if epoch >= 999:
plt.plot(Epoch_list, Loss_list)
plt.title('SGD')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.grid(ls='--')
plt.show()
# '''
# 5、Painting # (原本我的painting是写在这里的,但是会有错误,它每一次循环就会作两张图,所以把painting写进循环的if语句里了
# '''
# plt.plot(Epoch_list, Loss_list)
# plt.title('SGD')
# plt.xlabel('Epoch')
# plt.ylabel('Loss')
# plt.grid(ls='--')
# plt.show()
图片如下:
暂时不知道该如何解决