使用Pytorch实现linear_regression

使用Pytorch实现线性回归

# import necessary packages
import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt

# Set necessary Hyper-parameters.
input_size = 1
output_size = 1
num_epochs = 60
learning_rate = 0.001

# Define a Toy dataset.
x_train = np.array([[3.3], [4.4], [5.5], [6.71], [6.93], [4.168], [9.779], [6.182], [7.59], [2.167], [7.042], [10.791], [5.313], [7.997], [3.1]], dtype=np.float32)y_train = np.array([[1.7], [2.76], [2.09], [3.19], [1.694], [1.573], [3.366], [2.596], [2.53], [1.221], [2.827], [3.465], [1.65], [2.904], [1.3]], dtype=np.float32)# Confirm the data shape.
print(x_train.shape, y_train.shape)

(15, 1) (15, 1)

# Linear regression model
model = nn.Linear(input_size, output_size)

# Loss and optimizer
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate, )

# Train the model
for epoch in range(num_epochs):# Convert numpy arrays to torch tensorsinputs = torch.from_numpy(x_train)targets = torch.from_numpy(y_train)# Forward passoutputs = model(inputs)loss = criterion(outputs, targets)# Backward and optimizeoptimizer.zero_grad()loss.backward()optimizer.step()# Set an output counterif (epoch+1) % 5 == 0:print('Epoch [{}/{}], loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))# Plot the graph
predicted = model(torch.from_numpy(x_train)).detach().numpy()
plt.plot(x_train, y_train, 'ro', label='Original data')
plt.plot(x_train, predicted, label='Fitted line')
plt.legend()
plt.show()

Epoch [5/60], loss: 7.1598
Epoch [10/60], loss: 3.0717
Epoch [15/60], loss: 1.4154
Epoch [20/60], loss: 0.7443
Epoch [25/60], loss: 0.4722
Epoch [30/60], loss: 0.3618
Epoch [35/60], loss: 0.3169
Epoch [40/60], loss: 0.2985
Epoch [45/60], loss: 0.2909
Epoch [50/60], loss: 0.2876
Epoch [55/60], loss: 0.2861
Epoch [60/60], loss: 0.2853

# Save the model checkpoint
torch.save(model.state_dict(), 'model_param.ckpt')
torch.save(model, 'model.ckpt')