1: Open the WinSCP remote login tool. Enter the server IP address, port number, user name, password, etc. to be connected.
2: After successful login, open the terminal in PuTTY
3: First check the usage of the GPU server, use the command
nvidia-smi -l
Found that the first GPU can be used.
4: First create a virtual environment, use the following command:
virtualenv env_distantly_supervised_ner
5: Next, we have to consider whether to use python2.7 or python3, here is python2.7 as an example, first check the address of the python2.7 interpreter
which python2.7
6: Choose to use a Python interpreter
virtualenv -p /usr/bin/python2.7 env_distantly_supervised_ner
7: Activate the virtual environment
source bin/activate
8: Next we choose to copy an open source code on github to our server
git clone https://github.com/rainarch/DSNER.git
9: Next we need to install tensorflow, using the domestic source of the University of Science and Technology of China
pip install tensorflow-gpu==1.1.0 -i http://pypi.mirrors.ustc.edu.cn/simple/ --trusted-host pypi.mirrors.ustc.edu.cn
10: After the installation is successful, you can start training the model