A few things to know ahead of time :
1. The MobaXterm software used 2. To access local and remote servers, account numbers and passwords are required ( username and password
on the remote machine, IP address or host name of the remote machine (on the same subnet)). You need to enter a password every time you visit, you can search for Ubuntu SSH password-free login 3. The main thing is to access the local server before you can access the remote server. The graphics card is installed on the remote server. When running deep learning , first upload to the local, then copy to the remote. If the training is complete, the parameter file is uploaded locally.
If you don’t know how to use MobaXterm yet , you can refer to this link to use
MobaXterm to connect to the server and use Anaconda to install the pytoch framework to run the deep learning model (use the school server + graphics card for deep learning)
If you want to use the server to run the deep learning model, you can refer to this link
Use MobaXterm to connect to the server and use Anaconda to run the deep learning model LeNet-5 classification in the pytoch environment
4 Ways to Transfer Files Over SSH - wikiHow
1.scp: an old deprecated command
2.rsync: a popular command for file synchronization
3.sshfs: mount a remote directory via SSH
4.sftp client: a GUI tool for accessing files via SFTP
1. scp command (commands are all input on the local server)
(1) Local to remote: run locally
scp -r file directory account name@remote IP:remote directory (command pwd to view the remote directory path)
scp file name account name@remote IP:remote directory (command pwd to view the path of the remote directory)
%%没传输前，查看远端目录文件情况
wlc@hz-A100-40:/data/WLC/WKL$ ls
carb_classfication deep-learning-for-image-processing-master deep-learningfor-image-processing-master.zip zjdata
%%传送文件夹是carb，详细指令
wlc@hz-jumper:~$ scp -r "/home/wlc/WKL/carb/" wlc20222022@172.20.198.3:/data/WLC/WKL
%%查看传输成功
wlc@hz-A100-40:/data/WLC/WKL$ ls
carb carb_classfication deep-learning-for-image-processing-master deep-learning-for-image-processing-master.zip zjdata
%%这里我把anaconda传过去
scp -r "/home/wlc2021388321/Anaconda3-2023.07-0-Linux-x86_64.sh" [email protected]:/data/WLC/
(2) Remote to local: remote operation
scp -r account name@remote IP:remote directory local directory
wlc@hz-jumper:~$ scp -r [email protected]:/data/WLC/WKL/carb/Test2_alexnet "/home/wlc2021388321/WKL/"
1.jpg 100% 1400KB 96.1MB/s 00:00
predict1.py 100% 1518 5.4MB/s 00:00
3.jpg 100% 21KB 37.9MB/s 00:00
train.py 100% 4977 15.9MB/s 00:00
model.py 100% 2061 6.6MB/s 00:00
model.cpython-38.pyc 100% 1727 3.7MB/s 00:00
model.cpython-39.pyc 100% 1749 4.2MB/s 00:00
Test2_alexnet.iml 100% 340 828.8KB/s 00:00
.gitignore 100% 50 132.1KB/s 00:00
misc.xml 100% 308 879.6KB/s 00:00
.name 100% 8 23.6KB/s 00:00
modules.xml 100% 285 819.8KB/s 00:00
workspace.xml 100% 6342 17.1MB/s 00:00
profiles_settings.xml 100% 174 738.5KB/s 00:00
Project_Default.xml 100% 1504 2.5MB/s 00:00
predict.py 100% 1911 7.6MB/s 00:00
2.jpg
##从远端传输到本地
wlc@hz-jumper:~$ ls
b c snap wanghao WKL
wlc@hz-jumper:~$ cd WKL
wlc@hz-jumper:~/WKL$ ls
carb Test2_alexnet zj4