pytorch例子1——曲线拟合

最新推荐文章于 2024-03-02 00:07:58 发布

boldyoungster

最新推荐文章于 2024-03-02 00:07:58 发布

阅读量1.2k

点赞数

分类专栏： python 文章标签：深度学习 pytorch

By Zen. 行行皆辛苦，三连必须有！

本文链接：https://blog.csdn.net/xiaozisheng2008_/article/details/107804350

版权

python 专栏收录该内容

25 篇文章 46 订阅 ¥19.90 ¥99.00

订阅专栏

超级会员免费看

这篇博客介绍了如何使用PyTorch进行简单的曲线拟合，通过深度学习的方法实现数据的精确匹配。

摘要由CSDN通过智能技术生成

pytorch 简单曲线拟合

#!/usr/bin/python3
import numpy as np
import matplotlib.pyplot as plt
import torch
from torch import nn
from IPython import embed

batch_size = 10000
input_size = 1
output_size = 1
num_epochs = 1000
learning_rate = 0.1

x = torch.linspace(0, 1, batch_size).reshape(-1, 1)
y = x ** 2 + 1

model = nn.Sequential(
    nn.Linear(1, 1),
    nn.Sigmoid(),
    nn.Linear(1, 1),
    nn.Sigmoid(),
    nn.Linear(1, 1),
    nn.Sigmoid(),
    nn.Linear(1, 1),
    nn.Sigmoid(),
    nn.Linear(1, 1)
)
loss_func = torch.nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

for i in range(num_epochs):
    y_pred = model.forward(x)
    loss = loss_func(y_pred, y)
    optimizer.zero_grad()

    loss.backward()
    optimizer.step()
    if i % 100 == 0:
        print(f"epoch: {i}, loss: {loss}")

plt.plot(x.data, y.data, "g*")
plt.plot(x.data, model.forward(x).data, "r-")
plt.show()