/own-cheatsheet

This cheat sheet is quick check in diferent topics like configuration and instalation of "hard" libraries like tensorflow or pytorch or specific software

Workflow commands

I use this cheat sheet as referece when im working and forgot some command. This is going to be updated constantly in time

The idea of this cheatsheet is to have a reference for current and future projects en my carreer

Table of Contents

Conda

INSTALL Conda on Linux

Go to: https://www.anaconda.com/products/distribution#Downloads 

Copy the linux download hyperlink like:
https://repo.anaconda.com/archive/Anaconda3-2022.05-Linux-x86_64.sh

Type: wget <link>
Type: bash Anaconda*-*-Linux-x86_64.sh

Restart shell or source ~/.bashrc

Create Conda env

conda create --name "env project" python="version"

Remove Conda env

conda env remove --name "env name"

Install & register jupyter Kernel

conda install -c anaconda ipykernel #to install
python -m kernel install --user --name="name" #to register kernel

ex: 
python -m ipykernel install --user --name tensorflow --display-name "Python 3.8 (tensorflow)"

Create YML file from current env

conda env export > "name".yml

Install Pytorch for GPU

conda install pytorch cudatoolkit -c pytorch

Install CUDAtoolKit and cuDNN

conda install cudatoolkit="11.0" cudnn="8.0" -c=conda-forge

Pip

venv

create a Python virtual enviroment

python3 -m venv "venv name"

activate virtual enviroment

source "venv name"/bin/activate