Starting Python Project on Mac

First Things First

You have to install Homebrew and Xcode command-line tools

I installed X-code Directly from Appstore

xcode-select --install # I haven't tested this method but this should also work if you don't like installing from Appstore
To Install homebrew copy and run a command: https://brew.sh/

Don't forget to add homebrew to the path

Now let's install python

Installing and managing Python can be tricky, so I am going to leave that hassle to a tool pyenv with pyenv we can easily install uninstall and set default Python versions for different versions of Python.

# Hopefully you already have homebrew and xcode-cli-tools setup 

brew install pyenv #This will install pyenv
brew install xz # Had to install this to solve lamza unable to build issue when instaling python 

#Add Pyenv to path 
nano ~/.zshrc
export PATH="$(pyenv root)/shims:$PATH" #Add this line to file

#Pyenv commands
python --version #Check systems default python version 
pyenv versions  #Shows version of python available 
pyenv install 3.6.7 #This will install python verison 3.6.7  
pyenv global 3.6.7 #This will set 3.6.7 as main python

pyenv uninstall 3.6.7 #This will uninstall pyenv

Now set the Python environment

For data science and ML it's better to use Anaconda because it supports both Python and other languages unlike other package managers for other projects we can use Poetry which is convenient and does a lot of things for us.

🌲

Anaconda setup

I downloaded Anaconda app directly from the website Anaconda Download

#Info : Might have to restart IDE to see newly created virtual environment
python --version

conda config --set auto_activate_base false #true/false value will enable/disable automatic activation of base environment when a new termianl is launched

# ALWAYS START (base) ENVIRONMENT FIRST !!
conda activate base #Starts base environment 
conda list #This shows list of all pakages in base environment
conda deactivate #Closes virtual environment move back to terminal

conda env list #This lists all the created env's
conda create --name helloproject python=3.11 #This will create helloproject virtual environment
conda remove --name helloproject --all # removes env and removes all pakages in env
# Newly created Venvs dont have all pakages available in base env
# To get this pakages
conda activate helloproject #activate the env first 
conda install scikit-learn #Install package in this env

#Pakages not available in channels 
#There is a possibility that the package you are trying to install is not in the channel then you want to connect to a new channel
#(1) If you know which channel the package is available add that channel to our local so that conda can look into that 

conda config --show channels #This will list all the channels ie: defaults
conda config --add channels conda-forge #Adds new channel conda-forge
# Flags --add --append --prepend can be used interchangably to change priority
conda config --remove channels conda-forge #Remove channel 

conda config --get channels #This will also show priority of channels in order which packages will be looked up for download
conda remove --name helloproject --all #Removes virtual-env all its pakages and channels

📘

Poetry

Poetry can be used for quick development of projects where we have pure Python projects ie: Projects which does not have data science tools that rely on C or C++ backends. eg: Numpy

See the Poetry install document for more details

The document says poetry should not directly be installed on the system, but rather installed in a virtual environment. This sounds strange but when I think more about it conda does the same, what else conda base environment be?

# to do the setup quick install pipx first, This automatically does virtual environment setup for us 
brew install pipx
pipx ensurepath

#Now Poetry stuff 
pipx install poetry

# (Change default setting)
poetry config virtualenvs.in-project true #This will create venvs within the project true/false

#If you are creating a new project
poetry new hello_poetry #This will create a new project with necessary project structure

#For existing project 
poetry init #create `pyproject.toml` in your folder

poetry env info #Gives information about virtual environment 

poetry install #This will create virtualenvs install dependency
#ToDelete venv delete .venv folder

poetry shell #start venv in current terminal 

# Adding new pakages 
poetry add requests #Installs requests package
poetry add pytest --group dev #to install package to dev group
poetry remove requests #Uninstalls requests package 

exit #Exit shell without deactivating
deactivate #Deactivate virtual environment 

poetry env list #List all the active virtual environmnt