pytorch下使用tensorboardX可视化

环境：py36+pytorch1.10.2+tensorboardx2.5

1.配置tensorboardX环境

1.进入py36环境，pip install tensorboardX

2.conda 创建一个新的环境 假设名为tf ,在tf中pip install tensorboard

3.找到tf环境所在路径，如下列代码所示

D:\MiniConda3\envs\tf\Scripts

4.将此路径放到环境变量的Path中

到此为止就可以使用tf环境中的tensorboard.exe去实现其他pytorch环境中的项目了

2.tensorboardX使用简介

# -*- coding:utf-8 -*-
# @Time:2022/3/10 - 9:25
# @Author: Yongzheng
import torch
from tensorboardX import SummaryWriter
from torch import nn
from torch.utils.data import DataLoader
from torchvision import datasets
from torchvision.transforms import ToTensor

# ==================================================1.数据集==================================================#
# Download training data from open datasets.
training_data = datasets.FashionMNIST(
    root="dataset",  # 数据集保存路径
    train=True,
    download=True,
    transform=ToTensor(),
)

# Download test data from open datasets.
test_data = datasets.FashionMNIST(
    root="dataset",  # 数据集保存路径
    train=False,
    download=True,
    transform=ToTensor(),
)

# ==================================================2.超参数==================================================#
batch_size = 64
epochs = 9

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


# ==================================================3.模型定义==================================================#
# 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)
        )
        # self.linear1 = nn.Linear(28*28,512)
        # self.relu1 = nn.ReLU()
        # self.linear2 = nn.Linear(512,512)
        # self.relu2 = nn.ReLU()
        # self.out = nn.Linear(512,10)

    def forward(self, x):
        x = self.flatten(x)
        logits = self.linear_relu_stack(x)
        return logits


class LeNet(nn.Module):

    def __init__(self):
        super(LeNet, self).__init__()
        self.conv1 = nn.Sequential(
            nn.Conv2d(in_channels=1, out_channels=32, kernel_size=3, stride=1, padding=2),
            nn.ReLU(),
            nn.MaxPool2d(2, 2)
        )

        self.conv2 = nn.Sequential(
            nn.Conv2d(32, 64, 5),
            nn.ReLU(),
            nn.MaxPool2d(2, 2)
        )

        self.fc1 = nn.Sequential(
            nn.Linear(64 * 5 * 5, 128),
            nn.BatchNorm1d(128),
            nn.ReLU()
        )

        self.fc2 = nn.Sequential(
            nn.Linear(128, 64),
            nn.BatchNorm1d(64),  # 加快收敛速度的方法（注：批标准化一般放在全连接层后面，激活函数层的前面）
            nn.ReLU()
        )

        self.fc3 = nn.Linear(64, 10)

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = x.view(x.size()[0], -1)
        x = self.fc1(x)
        x = self.fc2(x)
        x = self.fc3(x)
        return x


# ==================================================4.训练过程定义==================================================#
def train(dataloader, model, loss_fn, optimizer,epoch):
    size = len(dataloader.dataset)
    model.train()
    loss_sigma = 0.0  # 记录一个epoch的loss之和
    correct_sigma = 0.0 # 准确率
    total=0 # 训练过的样本总数
    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)

        # 反向传播
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        # 计算总损失和平均值
        loss_sigma+=loss.item()
        _,pred2class = torch.max(pred,1)
        correct_sigma += (pred2class == y).cpu().squeeze().sum().numpy()
        total += y.size(0)

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

    loss_avg = loss_sigma / len(dataloader)
    loss_sigma = 0.0
    # 记录训练loss
    writer.add_scalars('Loss_group', {'train_loss': loss_avg}, epoch)
    # 记录Accuracy
    writer.add_scalars('Accuracy_group', {'train_acc': correct_sigma / total}, epoch)



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")


# ==================================================5.优化器和损失函数==================================================#

model = LeNet().to(device)
print(model)
loss_fn = nn.CrossEntropyLoss()  # 损失函数
# optimizer = torch.optim.SGD(model.parameters(), lr=1e-3)  # 反向传播
optimizer = torch.optim.SGD(model.parameters(), lr=1e-3, momentum=0.9)
# Create data loaders.
train_dataloader = DataLoader(training_data, batch_size=batch_size)
test_dataloader = DataLoader(test_data, batch_size=batch_size)

# ==================================================6.训练==================================================#
log_dir = "tensorboardX_logs"
# log_dir是日志文件存放的目录，当log_dir存在时comment自动忽略
# 若log_dir不存在，则默认目录为runs，然后在runs下生成新的文件夹存放日志，如Mar17_14-47-47_LAPTOP-KI10RP71-test，后缀‘-test’就是comment指定的内容
writer = SummaryWriter()
for t in range(epochs):
    print(f"Epoch {t + 1}\n-------------------------------")
    train(train_dataloader, model, loss_fn, optimizer,t+1)
    test(test_dataloader, model, loss_fn)
print("Done!")

 代码中的如下部分就是在代码中使用tensorboardX

 

 writer.add_scalars(table_name，y轴数据，x轴数据)

以epoch为x轴变量，每个epoch的平均损失为y轴数据，生成一个图像

 运行程序得到的目录

 3.tensorboardX可视化

1.切换目录

 2.启动tensorboard

 3.查看可视化结果