PyTorch深度学习训练可视化工具tensorboardX

最新推荐文章于 2025-04-01 11:23:55 发布

louwill12

最新推荐文章于 2025-04-01 11:23:55 发布

阅读量688

点赞数

本文链接：https://blog.csdn.net/weixin_37737254/article/details/103982342

版权

PyTorch

Author：Iouwill

Machine Learning Lab

之前笔者提到了PyTorch的专属可视化工具visdom，参看PyTorch深度学习训练可视化工具visdom。但在此之前很多TensorFlow用户更习惯于使用TensorBoard来进行训练的可视化展示。为了能让PyTorch用户也能用上TensorBoard，有开发者提供了PyTorch版本的TensorBoard，也就是tensorboardX。

安装与启动

熟悉TensorBoard的用户可以无缝对接到tensorboardX，安装方式为：

pip install tensorboardX

除了要安装PyTorch之外，还需要安装TensorFlow。跟TensorBoard一样，tensorboardX也支持scalar, image, figure, histogram, audio, text, graph, onnx_graph, embedding, pr_curve，video等不同类型对象的可视化展示方式。

tensorboardX和TensorBoard的启动方式一样，直接在终端下运行：

tensorboard --logdir runs

然后另起一个终端执行Python文件即可：

python demo.py

打开localhost:6006即可看到tensorboardX可视化界面。

tensorboardX本地启动非常容易，但一般情况下我们训练都是在服务器上完成的，所以要在远程启动tensorboardX需要进行一些简单的设置。以虚拟机工具xshell为例：依此设置文件->属性->ssh->隧道->添加,类型local，源主机填127.0.0.1（本机），端口设置一个，比如12345，目标主机为服务器地址，目标端口一般是6006，如果6006被占了可以改为其他端口。

分别执行tensorboard和python脚本后，本地打开127.0.0.1:12345即可进入远程TensorBoard界面。

使用示例

以scalar为例来看一下tensorboardX的使用方式：

import numpy as np
from tensorboardX import SummaryWriter
writer = SummaryWriter()
for i in range(100):
    writer.add_scalar('data/scalar1', np.random.rand(), i)
    writer.add_scalar('data/scalar2', {'xsinx': i*np.sin(i), 'xcosx': i*np.cos(i)}, i)
writer.close()

scalar可视化如下图所示。

一个完整tensorboardX 使用demo如下：

import torch
import torchvision.utils as vutils
import numpy as np
import torchvision.models as models
from torchvision import datasets
from tensorboardX import SummaryWriter


resnet18 = models.resnet18(False)
writer = SummaryWriter()
sample_rate = 44100
freqs = [262, 294, 330, 349, 392, 440, 440, 440, 440, 440, 440]


for n_iter in range(100):


    dummy_s1 = torch.rand(1)
    dummy_s2 = torch.rand(1)
    # data grouping by `slash`
    writer.add_scalar('data/scalar1', dummy_s1[0], n_iter)
    writer.add_scalar('data/scalar2', dummy_s2[0], n_iter)


    writer.add_scalars('data/scalar_group', {'xsinx': n_iter * np.sin(n_iter),
                                             'xcosx': n_iter * np.cos(n_iter),
                                             'arctanx': np.arctan(n_iter)}, n_iter)


    dummy_img = torch.rand(32, 3, 64, 64)  # output from network
    if n_iter % 10 == 0:
        x = vutils.make_grid(dummy_img, normalize=True, scale_each=True)
        writer.add_image('Image', x, n_iter)


        dummy_audio = torch.zeros(sample_rate * 2)
        for i in range(x.size(0)):
            # amplitude of sound should in [-1, 1]
            dummy_audio[i] = np.cos(freqs[n_iter // 10] * np.pi * float(i) / float(sample_rate))
        writer.add_audio('myAudio', dummy_audio, n_iter, sample_rate=sample_rate)


        writer.add_text('Text', 'text logged at step:' + str(n_iter), n_iter)


        for name, param in resnet18.named_parameters():
            writer.add_histogram(name, param.clone().cpu().data.numpy(), n_iter)


        # needs tensorboard 0.4RC or later
        writer.add_pr_curve('xoxo', np.random.randint(2, size=100), np.random.rand(100), n_iter)


dataset = datasets.MNIST('mnist', train=False, download=True)
images = dataset.test_data[:100].float()
label = dataset.test_labels[:100]


features = images.view(100, 784)
writer.add_embedding(features, metadata=label, label_img=images.unsqueeze(1))


# export scalar data to JSON for external processing
writer.export_scalars_to_json("./all_scalars.json")
writer.close()

可视化效果如下所示：