Introduction au Machine Learning

Code commenté de plusieurs cours Udemy (voir la rubrique Acknowledgments)

Prerequisites ⁉️

You need :

• Anaconda 2018.12 (with Python 3.7 version) : A free distribution of Python with scientific packages.

Getting Started

Read and Run 😎

Release History 📓

Actual version : 3.0

• V4.0 - Ajout de code & info de DLAZ : Le Deep Learning de A à Z (Udemy) dans la Partie 8 - Deep Learning. :star::star::star::star::star:

• V3.0 - Ajout de code & info de TFTS : Sequences, Time Series and Prediction (Coursera) dans la Partie 12 - Time Series. :star::star::star::star:

• V2.0 - Ajout d'infos et d'astuces tiré de DSMLP : Data science : Machine Learning et Python (Udemy). :star:

  • Les même points que la V1.0 avec quelques ajouts

• V1.0 - Code inspiré du cours IML : Introduction au Machine Learning (Udemy). :star::star::star::star::star:

  • Comprendre la base du ML
  • Pre-processing des datas (Manipulations des données, Gestion des données manquantes, Les Dummy Variables, Diviser le dataset Training/Test, Feature Scaling)
  • Régression: pour prédire une valeur réelle continue (Construction du modèle, Faire de nouvelles prédictions, Visualiser les résultats)
  • Classification: pour prédire une catégorie (idem + Matrice de confusion)
  • Clustering: pour identifier des segments d'observations groupées par similarité (idem)

Dependency 🔗

• Python Libraries for Data Science used:
  • numpy : Adds Python support for large, multi-dimensional arrays and matrices, along with a large library of high-level mathematical functions to operate on these arrays.
  • matplotlib : Python 2D plotting library which produces publication quality figures in a variety of hardcopy formats and interactive environments across platforms.
    • pyplot : Collection of command style functions that make matplotlib work like MATLAB.
    • colors : Functions and classes for color specification conversions, and for mapping numbers to colors in a 1-D array of colors called a colormap.
  • pandas : Software library written for data manipulation and analysis in Python. Offers data structures and operations for manipulating numerical tables and time series.
  • sklearn (IML) : Scikit-learn adds Python support for large, multi-dimensional arrays and matrices, along with a large library of high-level mathematical functions to operate on these arrays.
  • statsmodels (DSMLP) : Python module that provides many opportunities for statistical data analysis, such as statistical models estimation, performing statistical tests, etc.

Author 💻

• Yassine Gangat - gyassine

Acknowledgments 👍

• Inspired by :

  • https://www.udemy.com/introduction-au-machine-learning (IML) 🇫🇷 (Arborescence limitée)
  • https://www.udemy.com/data-science-machine-learning-et-python (DSMLP) 🇫🇷 (Quelques notes)
  • https://www.udemy.com/le-deep-learning-de-a-a-z (DLAZ) 🇫🇷 (La Partie 8 - Deep Learning partiellement)
  • https://www.udemy.com/machinelearning (MLAZ) 🇬🇧/🇺🇸 (Arborescence complète)
  • https://www.udemy.com/intelligence-artificielle-az (IAAZ) 🇫🇷 (La Partie 11 - Intelligence Artificielle en totalité)
  • https://www.coursera.org/learn/tensorflow-sequences-time-series-and-prediction (TFTS) :en: (La Partie 12 - Time Series)

• Good links :

  • https://github.com/donnemartin/data-science-ipython-notebooks
  • https://medium.com/activewizards-machine-learning-company/top-20-python-libraries-for-data-science-in-2018-2ae7d1db8049
  • MindMap sur Machine Learning
  • Cours Cnam : Apprentissage, réseaux de neurones et modèles graphiques
  • Gradient Descent Algorithm : Explications et Implémentation en Python

• Ressources:

  • 2015_Book_EfficientLearningMachines.pdf
  • Most of Cheat-sheet from here
  • Veille technologique RNN (récupéré depuis http://nicolasmine.com/projects/)

Tutos & Liens pour Stats/ML/Python 🐍

• Stats :

  • StatQuest
  • Régression linéaire simple : quand les hypothèses ne sont pas satisfaites
  • Khan Academy
  • Evaluating a Linear Regression Model
  • Assumptions of Linear Regression
  • Assumptions of Linear Regression with Python
  • Linear Regression
  • Linear Regression 101 (Part 3 - Assumptions & Evaluation)

• Python :

  • Python pour le calcul scientifique/Statistiques
  • Bonnes pratiques et astuces Python
  • Structurer votre projet
  • Apprenez à programmer en Python

• Formations :

  • https://makina-corpus.com/formations/python-scientifique

License 📜

The content of this project itself has been sometimes written or sometimes commented by me, but was presented inside theses videos. Datasets were provided also by them. They can be found on Kaggle too.