Pytorch工具：用visdom可视化loss

目录

1.安装

2 启动

3 示例

4. 深度学习损失可视化

5 结果

---

1.安装

pip install visdom

因为使用代理，突然报错 ERROR: Could not install packages due to an EnvironmentError: Missing dependencies for SOCKS support.

unset all_proxy
unset ALL_PROXY

取消后正常安装

 

2 启动

python -m visdom.server

返回

Checking for scripts.
It's Alive!
INFO:root:Application Started
You can navigate to http://localhost:8097

浏览器输入 http://localhost:8097 即可正常启动

3 示例

from visdom import Visdom
import numpy as np
viz = Visdom(env='my_wind')#设置环境窗口的名称是'my_wind',如果不设置名称就在main中
tr_loss=list(range(100))
ts_loss=list(range(10,110))
viz.line(Y=np.column_stack((np.array(tr_loss),np.array(ts_loss))), opts=dict(showlegend=True))

 

4. 深度学习损失可视化

• 将损失函数的可视化放在visual_loss.py文件：

# coding:utf8
import visdom
import time
import numpy as np
class Visualizer(object):
    def __init__(self, env='default', **kwargs):
        self.vis = visdom.Visdom(env=env, **kwargs