conda

1. If you’d prefer that conda’s base environment not be activated on startup, set the auto_activate_base parameter to false:

   conda config --set auto_activate_base false
   

jupyter

1. This will show you the URLs of running servers with their tokens, which you can copy and paste into your browser.

   jupyter notebook list
   

2. jupyter server password

3. docker + tensorflow-gpu + jupyterlab

   1. docker run -d –gpus all -p 8888:8888 -p 6006-6009:6006-6009 –restart=always –name=tensorflow -v /home/fang:/tf/fang tensorflow/tensorflow:latest-gpu-jupyter
   2. docker exec -it tensorflow pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
   3. docker exec -it tensorflow pip install jupyterlab
   4. docker exec -it tensorflow jupyter server password
   5. cd /var/lib/docker/containers/your container id
   6. docker stop tensorflow && sed -i ‘s/jupyter notebook/jupyter lab/g’ config.v2.json
   7. sudo systemctl restart docker

4. notebook config

   1. c.ServerApp.token = ‘xfangfang’

   2. c.ServerApp.quit_button = False

   3. c.ServerApp.terminado_settings = {'shell_command': ['/bin/bash']}
      

docker

1. docker run –rm –gpus all nvidia/cuda:11.0-base nvidia-smi

2. docker run -d --gpus all -p 8888:8888 --restart=always --name=tensorflow -v /home/fang:/tf/fang tensorflow/tensorflow:latest-gpu-jupyter
   

3. docker exec -it tensorflow python - c “import requests; requests.get(‘http://baidu.com’)”

4. docker commit tensorflow xfangfang/tensorflow-gpu-jupyter:v0.1
   

shell

1. Shows CPU freq

   watch -n 0 "cat /proc/cpuinfo | grep -i mhz"
   # or
   cpupower -c all frequency-info
   

2. 设置所有CPU为性能模式：

   cpupower -c all frequency-set -g performance
   

3. 设置所有CPU为节能模式：

   cpupower -c all frequency-set -g powersave
   

4. nvme smart-log /dev/nvme0

5. cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

node

apt install npm

npm config set registry http://registry.npm.taobao.org/

npm install n -g

n lts