安装依赖

pip install teorboard
pip install torch_tb_profiler

了解teorboard

记录并可视化标量[组]、图片[组]。

如何使用

第一步：构建模型，记录中间值，写入summarywriter

每次写入一个标量add_scalar

比如：

from torch.utils.tensorboard import SummaryWriter
writer = SummaryWriter()  # 默认会记录到'./runs/'这一目录下，也可以自己定义
x = range(100)
for i in x:writer.add_scalar('y=2x', i * 2, i)
writer.close()

期望输出：

每次写入多个标量add_scalars

比如：

from torch.utils.tensorboard import SummaryWriter
writer = SummaryWriter()
r = 5
for i in range(100):writer.add_scalars('run_14h', {'xsinx':i*np.sin(i/r),'xcosx':i*np.cos(i/r),'tanx': np.tan(i/r)}, i)
writer.close()

期望输出：

写入某个变量的分布add_histogram

比如：

from torch.utils.tensorboard import SummaryWriter
import numpy as np
writer = SummaryWriter()
for i in range(10):x = np.random.random(1000)writer.add_histogram('distribution centers', x + i, i)
writer.close()

期望输出：

写入单个图片add_image

比如：

from torch