Pytorch官网教程中文版-QuickStart

原文地址: link.

加载数据

PyTorch有两个处理数据的原语:torch.utils.data.DataLoader和torch.utils.data.Dataset

dataset存数据和标签，Dataloader通过Dataset构建一个可迭代的对象

import torch
from torch import nn
from torch.utils.data import DataLoader
from torchvision import datasets
from torchvision.transforms import ToTensor, Lambda, Compose
import matplotlib.pyplot as plt

PyTorch提供特定领域的库，如TorchText、TorchVision和TorchAudio，所有这些库都包含数据集。在本教程中，我们将使用TorchVision数据集。

torchvision.datasets 模块包含许多真实世界的视觉数据的数据集对象，如CIFAR, COCO(完整列表在这里)。在本教程中，我们使用FashionMNIST数据集。每个TorchVision Dataset包含两个参数transform和target_transform，分别修改样本和标签。

# 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(),
)

Out:

Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz
Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to data/FashionMNIST/raw/train-images-idx3-ubyte.gz
Extracting data/FashionMNIST/raw/train-images-idx3-ubyte.gz to data/FashionMNIST/raw

Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz
Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to data/FashionMNIST/raw/train-labels-idx1-ubyte.gz
Extracting data/FashionMNIST/raw/train-labels-idx1-ubyte.gz to data/FashionMNIST/raw

Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz
Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz
Extracting data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to data/FashionMNIST/raw

Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz
Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz
Extracting data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to data/FashionMNIST/raw

我们将数据集作为参数传递给DataLoader。它在我们的数据集上包装了一个可迭代对象，并支持自动批处理、采样、变换和多进程数据加载。这里我们定义了一个64的批大小，即dataloader iterable中的每个元素将返回一个包含64个特性和标签的批大小。

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("Shape of X [N, C, H, W]: ", X.shape)
    print("Shape of y: ", y.shape, y.dtype)
    break

Out:

Shape of X [N, C, H, W]:  torch.Size([64, 1, 28, 28])
Shape of y:  torch.Size([64]) torch.int64

创建模型：

要在PyTorch中定义神经网络，我们需要创建一个继承自nn.Module的类。我们在__init__函数中定义网络层，并在forward函数中指定数据如何通过网络。为了加速神经网络中的操作，如果可以的话，我们将其转移到GPU上。

# Get cpu or gpu device for training.
device = "cuda" if torch.cuda.is_available() else "cpu"
print(f"Using {device} device")

# Define model
class NeuralNetwork(nn.Module):
    def __init__(self):
        super(NeuralNetwork, self).__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

model = NeuralNetwork().to(device)
print(model)

Out:

Using cuda 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)
  )
)

模型参数：

为了训练模型，我们需要一个损失函数和一个优化器。

loss_fn = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=1e-3)

在一个单独的训练循环中，模型对训练数据集进行预测(分批输入)，并反向传播预测误差以调整模型的参数

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
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        if batch % 100 == 0:
            loss, current = loss.item(), batch * len(X)
            print(f"loss: {loss:>7f}  [{current:>5d}/{size:>5d}]")

我们还根据测试数据集检查模型的性能，以确保它正在学习。

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")

训练过程是在几个迭代(阶段)中进行的。在每个epoch期间，模型学习参数以做出更好的预测。我们打印模型在每个时期的准确性和损失;我们希望看到精确度的提高和损失的减少。

epochs = 5
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!")

Out:

Epoch 1
-------------------------------
loss: 2.304728  [    0/60000]
loss: 2.296327  [ 6400/60000]
loss: 2.277599  [12800/60000]
loss: 2.271838  [19200/60000]
loss: 2.249222  [25600/60000]
loss: 2.214130  [32000/60000]
loss: 2.224430  [38400/60000]
loss: 2.195380  [44800/60000]
loss: 2.204200  [51200/60000]
loss: 2.157917  [57600/60000]
Test Error:
 Accuracy: 34.7%, Avg loss: 2.161951

Epoch 2
-------------------------------
loss: 2.175548  [    0/60000]
loss: 2.167279  [ 6400/60000]
loss: 2.113919  [12800/60000]
loss: 2.130813  [19200/60000]
loss: 2.087419  [25600/60000]
loss: 2.017404  [32000/60000]
loss: 2.047705  [38400/60000]
loss: 1.981830  [44800/60000]
loss: 1.996790  [51200/60000]
loss: 1.906256  [57600/60000]
Test Error:
 Accuracy: 58.1%, Avg loss: 1.918405

Epoch 3
-------------------------------
loss: 1.949024  [    0/60000]
loss: 1.920072  [ 6400/60000]
loss: 1.813589  [12800/60000]
loss: 1.850658  [19200/60000]
loss: 1.758138  [25600/60000]
loss: 1.692796  [32000/60000]
loss: 1.708207  [38400/60000]
loss: 1.625349  [44800/60000]
loss: 1.648849  [51200/60000]
loss: 1.526103  [57600/60000]
Test Error:
 Accuracy: 63.3%, Avg loss: 1.558342

Epoch 4
-------------------------------
loss: 1.617884  [    0/60000]
loss: 1.581161  [ 6400/60000]
loss: 1.441004  [12800/60000]
loss: 1.506837  [19200/60000]
loss: 1.398544  [25600/60000]
loss: 1.381071  [32000/60000]
loss: 1.383345  [38400/60000]
loss: 1.327541  [44800/60000]
loss: 1.360360  [51200/60000]
loss: 1.240915  [57600/60000]
Test Error:
 Accuracy: 63.8%, Avg loss: 1.281359

Epoch 5
-------------------------------
loss: 1.353472  [    0/60000]
loss: 1.334475  [ 6400/60000]
loss: 1.175074  [12800/60000]
loss: 1.277382  [19200/60000]
loss: 1.157673  [25600/60000]
loss: 1.172853  [32000/60000]
loss: 1.181972  [38400/60000]
loss: 1.140976  [44800/60000]
loss: 1.179954  [51200/60000]
loss: 1.074737  [57600/60000]
Test Error:
 Accuracy: 64.9%, Avg loss: 1.109494

Done!

保存模型：

保存模型的常用方法是序列化内部状态字典(包含模型参数)。

torch.save(model.state_dict(), "model.pth")
print("Saved PyTorch Model State to model.pth")

Out:

Saved PyTorch Model State to model.pth

加载模型：

加载模型的过程包括重新创建模型结构并将状态字典加载到其中

model = NeuralNetwork()
model.load_state_dict(torch.load("model.pth"))

这个模型现在可以用来进行预测:

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():
    pred = model(x)
    predicted, actual = classes[pred[0].argmax(0)], classes[y]
    print(f'Predicted: "{predicted}", Actual: "{actual}"')

Out:

Predicted: "Ankle boot", Actual: "Ankle boot"