csds

Différences

Ci-dessous, les différences entre deux révisions de la page.

Lien vers cette vue comparative

Les deux révisions précédentes Révision précédente
Prochaine révision		 Révision précédente
csds [2023/07/12 12:39] – oboironcsds [2026/09/01 12:06] (Version actuelle) – fschwander
Ligne 1: Ligne 1:
 +====== Introduction to Modern Data Science (24h) ======
  
 +===== Course Outline =====
 + 
 +
 +This course introduces students to the modern data science ecosystem, combining scientific computing, data engineering, machine
 +learning, and software engineering best practices. Students learn how to build reproducible data science projects using
 +contemporary Python tools while developing the practical skills required to acquire, process, analyze, visualize, and model real-world
 +data.
 +
 +The course covers the complete data science workflow, from environment setup and collaborative development to exploratory data
 +analysis, statistical inference, machine learning, REST APIs, containerization, and the fundamentals of MLOps. Emphasis is placed on
 +reproducibility, scalability, and industry-standard tools widely used in professional data science. 
 +
 +
 +=== 1. Computing Environment and Python Fundamentals Overview of Linux, Windows, and macOS ===
 +File systems, paths, and permissions
 +
 +Essential shell commands and scripting
 +
 +Python refresher: scripts, modules, and packages
 +
 +Command-line workflows for data science 
 +
 +=== 2. Modern Python Tooling === 
 +The modern Python ecosystem
 +
 +Virtual environments and dependency management with uv
 +
 +Code quality using ruff (linting and formatting)
 +
 +Development with Visual Studio Code
 +
 +AI-assisted programming and productivity tools
 +
 +Reproducible Python environments 
 +
 +=== 3. Data Manipulation and Processing === 
 +Working with tabular data using NumPy and Pandas
 +
 +Data formats: CSV, Parquet, and Apache Arrow
 +
 +Data cleaning and preprocessing
 +
 +Filtering, joins, aggregations, and missing values
 +
 +Performance considerations
 +
 +Introduction to scalable data processing with Polars, DuckDB, Dask, and Spark 
 +
 +=== 4. Exploratory Data Analysis and Statistical Thinking ===
 +Principles of exploratory data analysis (EDA)
 +
 +Descriptive statistics and data summarization
 +
 +Statistical inference with SciPy
 +
 +Data visualization using Matplotlib and Seaborn
 +
 +Interactive dashboards with Plotly and Streamlit
 +
 +Communicating insights through visualizations 
 +
 +=== 5. Version Control and Project Organization === 
 +Git fundamentals
 +
 +GitHub and collaborative development workflows
 +
 +Branching, merging, and pull requests
 +
 +Organizing reproducible data science projects
 +
 +Notebooks versus Python scripts
 +
 +Literate programming with Marimo 
 +
 +=== 6. Building Data Applications === 
 +Consuming REST APIs with requests
 +
 +Designing REST APIs with FastAPI
 +
 +Introduction to containerization with Docker (or Podman)
 +
 +Building portable and reproducible applications
 +
 +Deploying simple data services 
 +
 +=== 7. Introduction to Machine Learning and Artificial IntelligenceThe machine learning workflow ===
 +Supervised and unsupervised learning
 +
 +Feature engineering and model evaluation
 +
 +Introduction to deep learning
 +
 +Large Language Models (LLMs)
 +
 +Generative AI, prompt engineering, and AI agents 
 +
 +=== 8. MLOps and Production Machine Learning === 
 +From notebooks to production
 +
 +Model serving: batch and real-time inference
 +
 +Experiment tracking with MLflow
 +
 +CI/CD for machine learning projects
 +
 +Introduction to cloud-native ML workflows
 +
 +Model monitoring, drift detection, and retraining strategie