学习基础知识
机器学习的基本流程
- 数据处理
- 创建模型
- 优化模型参数
- 保存训练的模型
快速入门
一个简单的目标分类任务
识别衣服的类型
import torch
from torch import nn
from torch.utils.data import DataLoader
from torchvision import datasets
from torchvision.transforms import ToTensor
# Define model
class NeuralNetwork(nn.Module):
def __init__(self):
super().__init__()
self.flatten = nn.Flatten()
self.linear_relu_stack = nn.Sequential(
nn.Linear(28 * 28, 512),
nn.ReLU(),
nn.Linear(512, 512),
nn.ReLU(),
nn.Linear(512, 10)
)
def forward(self, x):
x = self.flatten(x)
logits = self.linear_relu_stack(x)
return logits
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset)
model.train()
for batch, (X, y) in enumerate(dataloader):
X, y = X.to(device), y.to(device)
# Compute prediction error
pred = model(X)
loss = loss_fn(pred, y)
# Backpropagation
loss.backward()
optimizer.step()
optimizer.zero_grad()
if batch % 100 == 0:
loss, current = loss.item(), (batch + 1) * len(X)
print(f"loss: {loss:>7f} [{current:>5d}/{size:>5d}]")
def test(dataloader, model, loss_fn):
size = len(dataloader.dataset)
num_batches = len(dataloader)
model.eval()
test_loss, correct = 0, 0
with torch.no_grad():
for X, y in dataloader:
X, y = X.to(device), y.to(device)
pred = model(X)
test_loss += loss_fn(pred, y).item()
correct += (pred.argmax(1) == y).type(torch.float).sum().item()
test_loss /= num_batches
correct /= size
print(f"Test Error: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")
if __name__ == "__main__":
# 处理数据
# Download training data from open datasets.
training_data = datasets.FashionMNIST(
root="data",
train=True,
download=True,
transform=ToTensor(),
)
# Download test data from open datasets.
test_data = datasets.FashionMNIST(
root="data",
train=False,
download=True,
transform=ToTensor(),
)
batch_size = 64
# Create data loaders.
train_dataloader = DataLoader(training_data, batch_size=batch_size)
test_dataloader = DataLoader(test_data, batch_size=batch_size)
for X, y in test_dataloader:
print(f"Shape of X [N, C, H, W]: {X.shape}")
print(f"Shape of y: {y.shape} {y.dtype}")
break
# 创建模型
# Get cpu, gpu or mps device for training.
device = (
"cuda"
if torch.cuda.is_available()
else "mps"
if torch.backends.mps.is_available()
else "cpu"
)
print(f"Using {device} device")
model = NeuralNetwork().to(device)
print(model)
# 优化模型参数
# 损失函数
loss_fn = nn.CrossEntropyLoss()
# 优化器
optimizer = torch.optim.SGD(model.parameters(), lr=1e-3)
epochs = 20
for t in range(epochs):
print(f"Epoch {t + 1}\n-------------------------------")
train(train_dataloader, model, loss_fn, optimizer)
test(test_dataloader, model, loss_fn)
print("Done!")
# 保存模型
torch.save(model.state_dict(), "model.pth")
print("Saved PyTorch Model State to model.pth")
# 加载模型
model = NeuralNetwork().to(device)
model.load_state_dict(torch.load("model.pth"))
# 使用训练的模型进行预测
"""
“t恤/顶”,
“裤子”,
“套衫”,
“衣服”,
“外套”,
“凉鞋”,
“衬衫”,
“运动鞋”,
“包”,
“踝靴”,
"""
classes = [
"T-shirt/top",
"Trouser",
"Pullover",
"Dress",
"Coat",
"Sandal",
"Shirt",
"Sneaker",
"Bag",
"Ankle boot",
]
model.eval()
x, y = test_data[0][0], test_data[0][1]
with torch.no_grad():
x = x.to(device)
pred = model(x)
predicted, actual = classes[pred[0].argmax(0)], classes[y]
print(f'Predicted: "{predicted}", Actual: "{actual}"')
执行过程
/Users/futuredeng/anaconda3/envs/pyspide6_study/bin/python -X pycache_prefix=/Users/futuredeng/Library/Caches/JetBrains/PyCharm2024.1/cpython-cache /Applications/PyCharm.app/Contents/plugins/python/helpers/pydev/pydevd.py --multiprocess --qt-support=auto --client 127.0.0.1 --port 52646 --file /Users/futuredeng/PycharmProjects/pyspide6_study/s_torch/demo.py
已连接到 pydev 调试器(内部版本号 241.14494.19)Shape of X [N, C, H, W]: torch.Size([64, 1, 28, 28])
Shape of y: torch.Size([64]) torch.int64
Using mps device
NeuralNetwork(
(flatten): Flatten(start_dim=1, end_dim=-1)
(linear_relu_stack): Sequential(
(0): Linear(in_features=784, out_features=512, bias=True)
(1): ReLU()
(2): Linear(in_features=512, out_features=512, bias=True)
(3): ReLU()
(4): Linear(in_features=512, out_features=10, bias=True)
)
)
Epoch 1
-------------------------------
loss: 2.298847 [ 64/60000]
loss: 2.291248 [ 6464/60000]
loss: 2.278691 [12864/60000]
loss: 2.270169 [19264/60000]
loss: 2.247777 [25664/60000]
loss: 2.226532 [32064/60000]
loss: 2.221170 [38464/60000]
loss: 2.191688 [44864/60000]
loss: 2.186391 [51264/60000]
loss: 2.159593 [57664/60000]
Test Error:
Accuracy: 48.9%, Avg loss: 2.151264
Epoch 2
-------------------------------
loss: 2.160394 [ 64/60000]
loss: 2.149280 [ 6464/60000]
loss: 2.097811 [12864/60000]
loss: 2.111865 [19264/60000]
loss: 2.052902 [25664/60000]
loss: 2.002435 [32064/60000]
loss: 2.016076 [38464/60000]
loss: 1.941067 [44864/60000]
loss: 1.946122 [51264/60000]
loss: 1.868463 [57664/60000]
Test Error:
Accuracy: 58.9%, Avg loss: 1.870417
Epoch 3
-------------------------------
loss: 1.907855 [ 64/60000]
loss: 1.873430 [ 6464/60000]
loss: 1.759730 [12864/60000]
loss: 1.795776 [19264/60000]
loss: 1.683292 [25664/60000]
loss: 1.641434 [32064/60000]
loss: 1.654433 [38464/60000]
loss: 1.561658 [44864/60000]
loss: 1.587600 [51264/60000]
loss: 1.478837 [57664/60000]
Test Error:
Accuracy: 60.4%, Avg loss: 1.501103
Epoch 4
-------------------------------
loss: 1.573581 [ 64/60000]
loss: 1.536037 [ 6464/60000]
loss: 1.387229 [12864/60000]
loss: 1.457505 [19264/60000]
loss: 1.340816 [25664/60000]
loss: 1.339017 [32064/60000]
loss: 1.352573 [38464/60000]
loss: 1.279530 [44864/60000]
loss: 1.314921 [51264/60000]
loss: 1.217413 [57664/60000]
Test Error:
Accuracy: 63.3%, Avg loss: 1.242960
Epoch 5
-------------------------------
loss: 1.320792 [ 64/60000]
loss: 1.301409 [ 6464/60000]
loss: 1.135017 [12864/60000]
loss: 1.243455 [19264/60000]
loss: 1.120873 [25664/60000]
loss: 1.144230 [32064/60000]
loss: 1.168045 [38464/60000]
loss: 1.104519 [44864/60000]
loss: 1.145055 [51264/60000]
loss: 1.060252 [57664/60000]
Test Error:
Accuracy: 64.9%, Avg loss: 1.082238
Done!
Saved PyTorch Model State to model.pth
Predicted: "Ankle boot", Actual: "Ankle boot"
进程已结束,退出代码为 0