Statistics and Machine Learning in Python

---

Edouard Duchesnay, Tommy Löfstedt, Feki Younes

Important links

• Github

• Latest pdf

• Official deposit for citation.

• Web page

Python ecosystem for data science

• Introduction to Python language
• Overview of Python ecosystem for data-science
• Development Environment
• Package & Environments Dependency Management
• Development with Integrated Development Environment (IDE) and JupyterLab

Python language

• Import libraries
• Basic operations
• Data types
• Execution control statements
• List comprehensions, iterators, etc.
• Functions
• Regular Expression
• System programming
• Scripts and argument parsing
• Networking
• Object Oriented Programming (OOP)
• Style guide for Python programming
• Documenting
• Modules and packages
• Unit testing
• Exercises

Data Manipulation and Visualization

• Numpy: Arrays and Matrices
• Pandas: data manipulation

Numerical Methods in Python

• Numerical Differentiation
• Numerical Integration
• Time Series
• Optimization (Minimization) by Gradient Descent

Statistics

• Univariate Statistics
• Hands-On: Brain volumes study
• Linear Mixed Models
• Multivariate Statistics
• Resampling and Monte Carlo Methods

Machine Learning Overview

• Introduction
• Overfitting and Regularization

Unsupervised Learning

• Linear Dimensionality Reduction and Feature Extraction
• Manifold learning: non-linear dimension reduction
• Clustering

Supervised Learning

• Linear Models for Regression
• Linear Models for Classification
• Non-Linear Kernel Methods and Support Vector Machines (SVM)
• Non-Linear Ensemble Learning

Resampling Methods and Model Evaluation

• Out-of-sample Validation for Model Selection and Evaluation
• Random Permutations: sample the null distribution
• Bootstrapping
• Parallel Computation
• Hands-On: Validation of Supervised Classification Pipelines

Deep Learning: Introduction

• Backpropagation
• Multilayer Perceptron (MLP)

Deep Learning for Computer Vision

• Convolutional Neural Networks (CNNs)
• Pretraining and Transfer Learning

Natural Language Processing

• Bag-of-Words Models

Mathematical Details and Demonstrations

• Negative Log-Likelihood (NLL) for Binary Classification with Sigmoid Activation

Indices and tables

• Index

• Module Index

• Search Page