Python Tutorial
Open powershell as administrator and execute the following command
Set-ExecutionPolicy unrestricted
the accepte all by typing A
Chose the python version that you want (recomended are 3.7 or 3.8) https://www.python.org/downloads/windows/ Download and install python
Python will install automatically pip
which is a tool that help install python packages
Virtualenv is a tool that help python developper manage virtual environement for a better python libraries dependencies management
more information here https://virtualenv.pypa.io/en/stable/user_guide.html#introduction
to install virtualenv all you need to do is to run the following commande on your commande line tool
pip install virtualenv
To create a virtual environment you need to be in the directory of your workspace then run the following comand by replace myenv
with the ame you want to give to your virtual environement
virtualenv myenv
for example, on Windows
PS C:\Users\yassi\MyWorkSpace> virtualenv myenv created virtual environment CPython3.8.6.final.0-64 in 1153ms
creator Venv(dest=C:\Users\yassi\MyWorkSpace\myenv, clear=False, global=False, describe=CPython3Windows)
seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=C:\Users\yassi\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.8_qbz5n2kfra8p0\LocalCache\Local\pypa\virtualenv)
added seed packages: pip==20.2.4, setuptools==50.3.2, wheel==0.35.1
activators BashActivator,BatchActivator,FishActivator,PowerShellActivator,PythonActivator,XonshActivator
PS C:\Users\yassi\MyWorkSpace> .\myenv\Scripts\activate
If this does not work you can try
python -m virtualenv myenv
to activate the virtual environment
On Windows
./myenv/Scripts/activate
example
PS C:\Users\yassi\MyWorkSpace> .\myenv\Scripts\activate
(myenv) PS C:\Users\yassi\MyWorkSpace>
On Linux and MacOS
source myenv/bin/activate
(myenv)
this means that you are in your virtual environment named myenv
once you are in your virtual environment, you can install the packages you will be needing for your project.
At the end of your project
to deactivate the virtual environment you can run the following command
deactivate
To install a package, you run the command pip install package-name
replaceing package name by the package you want to install
The necessary packages for data science are the following
- numpy (for array data)
- pandas (to deal with tables)
- matplotlib
- jupyter ( or jupyterlab if you want to be fancy)
- seaborn (for beautifull plots)
- scipy (for scientific calculation)
- scikit-learn (for machine learning algorithms)
- tqdm ( for some fancy progress bars)
If you want to install all those libraries in one command :
- create a txt file that you can name
requirements.txt
in your working directory - copy the name of libraries you want to install in that file and save it
numpy
pandas
matplotlib
jupyter
seaborn
scipy
scikit-learn
tqdm
- run the following command
pip install -r requirements.txt