tensorboard不能正常创建文件的内容，打不开tensorboardX
当下正在学习torch下的SummaryWriter，记录loss损失。结果不能log下不能创建v2文件。创建的是0文件且有内容。
整体的代码过程
from tqdm import trange
import numpy as np
from torch.utils.tensorboard import SummaryWriter
import torch
import torch.nn as nn
import torch.optim as optim
device = torch.device('cpu')
net = nn.Linear(1,1).to(device) # 最简单全链接模型
loss_fn =nn.MSELoss().to(device)# 损失函数
optimizer = optim.SGD(net.parameters(),lr=0.001)
# 1. 输入数据
x_train = np.linspace(-1, 1, 100).reshape(100, 1)
y_train = 3 * np.power(x_train, 2) + 2 + 0.2 * np.random.randn(100, 1)
# 2. 实例化sw
sw1 = SummaryWriter(log_dir='log1')
# 3. 模型训练
for epoch in trange(40):
inputs = torch.from_numpy(x_train).type(torch.float32).to(device) # 所有输入
targets = torch.from_numpy(y_train).type(torch.float32).to(device) # 标签
output = net(inputs) # 模型的输入输出
loss = loss_fn(output, targets) # 预测则和真实值的loss
loss.backward() # 模型向后传递
optimizer.step() # 优化器优化
sw1.add_scalar("train_loss", loss.item(), epoch)
sw1.close()
运行结果及报错内容
提示内容为
W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found
I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
打开tensorBordx的过程
复制：
http://localhost:6006/到谷歌浏览器后，打开界面为
