How to build a deep learning server?
[200625] Proposal
- Hardware configuration
Item | Value |
---|---|
CPU | Intel i7-7800X 3.5 GHz |
GPU | GeForce GTX 1080 TI |
Memory | 32 GB |
Hard Disk | 4 TB |
- Software configuration
Item | Value |
---|---|
OS | Ubuntu 20.04 |
Environment | Anaconda 3, 2020.02 |
Architecture | Pytorch 1.5.1 (torchvision 0.6.1, torchaudio 0.5.0) |
Python IDE | Pycharm, community, 2020.1.2 |
Other IDE | VS Code |
Remote control | Remmina |
Step by step operation
Step1: Ubuntu
Prepare USB flash according to 1
Then press Delete
key at startup to run into BIOS setting.
After running into BIOS, change the Startup option(to change first priority):
- Before: UEFI Hard Disk: ubuntu (INTEL SSDPEKKR256G7)
- After: UEFI USB flash
Step2: Anaconda
After download the installation package, the command is:
$ bash ~/Downloads/Anaconda3-2020.02-Linux-x86_64.sh
$ conda list
conda: command not found
$ export PATH=~/anaconda3/bin:$PATH
$ conda list
# packages in environment at /home/cui/anaconda3:
...
Remember to replace setting with Tsinghua mirror.
$ gedit .condarc
Copy the following lines to .condarc:
channels:
- defaults
show_channel_urls: true
channel_alias: https://mirrors.tuna.tsinghua.edu.cn/anaconda
default_channels:
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/pro
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
custom_channels:
conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
pytorch: https://mirrors.tuna.tsinghua.ed